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.
14 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
15 we assumed that tags in the header were well-formed, and parsed the
16 element content after inspecting only the first char of the tag.
17 Assumptions at that stage could crash the receive process on malformed
20 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
21 While running the DKIM ACL we operate on the Permanent memory pool so that
22 variables created with "set" persist to the DATA ACL. Also (at any time)
23 DNS lookups that fail create cache records using the Permanent pool. But
24 expansions release any allocations made on the current pool - so a dnsdb
25 lookup expansion done in the DKIM ACL releases the memory used for the
26 DNS negative-cache, and bad things result. Solution is to switch to the
27 Main pool for expansions.
28 While we're in that code, add checks on the DNS cache during store_reset,
29 active in the testsuite.
30 Problem spotted, and debugging aided, by Wolfgang Breyha.
32 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
33 When none of the hosts presented to a transport match an already-open
34 connection, close it and proceed with the list. Previously we would
35 queue the message. Spotted by Lena with Yahoo, probably involving
38 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
39 Previously a spurious "250 OK id=" response was appended to the proper
46 JH/01 Rework error string handling in TLS interface so that the caller in
47 more cases is responsible for logging. This permits library-sourced
48 string to be attached to addresses during delivery, and collapses
49 pairs of long lines into single ones.
51 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
52 during configuration. Wildcards are allowed and expanded.
54 JH/02 Rework error string handling in DKIM to pass more info back to callers.
55 This permits better logging.
57 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
58 do not close it down and have the child transport start it up again on
59 the passed-on TCP connection. Instead, proxy the child (and any
60 subsequent ones) for TLS via a unix-domain socket channel. Logging is
61 affected: the continued delivery log lines do not have any DNSSEC, TLS
62 Certificate or OCSP information. TLS cipher information is still logged.
64 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
65 identical IP addresses on different listening ports. Will also affect
68 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
69 add noisy ifdef guards to special-case this sillyness.
70 Patch from Bernd Kuhls.
72 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
73 than 255 are no longer allowed.
75 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
76 Disable the session-cache too, which might reduce our load. Since we
77 currrectly use a new context for every connection, both as server and
78 client, there is no benefit for these.
79 GnuTLS appears to not support tickets server-side by default (we don't
80 call gnutls_session_ticket_enable_server()) but client side is enabled
81 by default on recent versions (3.1.3 +) unless the PFS priority string
84 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
85 <https://reproducible-builds.org/specs/source-date-epoch/>.
87 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
88 the check for any unsuccessful recipients did not notice the limit, and
89 erroneously found still-pending ones.
91 JH/08 Pipeline CHUNKING command and data together, on kernels that support
92 MSG_MORE. Only in-clear (not on TLS connections).
94 JH/09 Avoid using a temporary file during transport using dkim. Unless a
95 transport-filter is involved we can buffer the headers in memory for
96 creating the signature, and read the spool data file once for the
97 signature and again for transmission.
99 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
100 4.77 as the kernel support then wasn't solid, having issues in 64bit
101 mode. Now, it's been long enough. Add support for FreeBSD also.
103 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
104 case where the routing stage had gathered several addresses to send to
105 a host before calling the transport for the first, we previously failed
106 to close down TLS in the old transport process before passing the TCP
107 connection to the new process. The new one sent a STARTTLS command
108 which naturally failed, giving a failed delivery and bloating the retry
109 database. Investigation and fix prototype from Wolfgang Breyha.
111 JH/12 Fix check on SMTP command input synchronisation. Previously there were
112 false-negatives in the check that the sender had not preempted a response
113 or prompt from Exim (running as a server), due to that code's lack of
114 awareness of the SMTP input buffering.
116 PP/04 Add commandline_checks_require_admin option.
117 Exim drops privileges sanely, various checks such as -be aren't a
118 security problem, as long as you trust local users with access to their
119 own account. When invoked by services which pass untrusted data to
120 Exim, this might be an issue. Set this option in main configuration
121 AND make fixes to the calling application, such as using `--` to stop
124 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
125 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
126 responses to those, into a single TLS record each way (this usually means
127 a single packet). As a side issue, smtp_enforce_sync now works on TLS
130 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
131 affects you only if you're dancing at the edge of the param size limits.
132 If you are, and this message makes sense to you, then: raise the
133 configured limit or use OpenSSL 1.1. Nothing we can do for older
136 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
137 cmdline element to get the documented default one. Previously it was
140 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
141 are unsuable for non-message-specific reasons - eg. connection timeout,
142 banner-time rejection.
144 JH/16 Fix logging of delivery remote port, when specified by router, under
147 PP/06 Repair manualroute's ability to take options in any order, even if one
148 is the name of a transport.
151 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
153 JH/17 Change the list-building routines interface to use the expanding-string
154 triplet model, for better allocation and copying behaviour.
156 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
157 Previously it was constructed the first time a possibly-matching string
158 was met in the configuration file input during startup; now it is done
161 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
162 compatible one, to avoid the (poorly documented) possibility of a config
163 file in the working directory redirecting the DB files, possibly correpting
164 some existing file. CVE-2017-10140 assigned for BDB.
166 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
167 cache-hot. Previously, although the result was properly cached, the
168 initial verify call returned a defer.
170 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
171 the main verify for receipient in uncached-mode.
173 JH/22 Retire historical build files to an "unsupported" subdir. These are
174 defined as "ones for which we have no current evidence of testing".
176 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
177 if present. Previously it was ignored.
179 JH/24 Start using specified-initialisers in C structure init coding. This is
180 a C99 feature (it's 2017, so now considered safe).
182 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
183 if was a fixed-sized field and bitmask ops via macros; it is now more
186 PP/07 GitHub PR 56: Apply MariaDB build fix.
187 Patch provided by Jaroslav Škarvada.
189 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
190 during Coverity cleanups [4.87 JH/47]
191 Diagnosis and fix provided by Michael Fischer v. Mollard.
193 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
194 the right size to place the terminating semicolon on its own folded
195 line, the header hash was calculated to an incorrect value thanks to
196 the (relaxed) space the fold became.
198 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
199 and confused the parent.
201 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
202 which could crash as a result. This could lead to undeliverable messages.
204 JH/28 Logging: "next input sent too soon" now shows where input was truncated
207 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
208 matters on fast-turnover and PID-randomising systems, which were getting
209 out-of-order delivery.
211 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
212 a possibly-overlapping copy. The symptom was that "Remote host closed
213 connection in response to HELO" was logged instead of the actual 4xx
216 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
217 Previously only that bufferd was discarded, resulting in SYMTP command
220 JH/32 DKIM: when a message has multiple signatures matching an identity given
221 in dkim_verify_signers, run the dkim acl once for each. Previously only
222 one run was done. Bug 2189.
224 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
225 "panic the current process" to "deliberately defer". The panic log is
226 still written with the problem list name; the mail and reject logs now
227 get a temp-reject line for the message that was being handled, saying
228 something like "domains check lookup or other defer". The SMTP 451
229 message is still "Temporary local problem".
231 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
232 A crafted sequence of BDAT commands could result in in-use memory beeing
233 freed. CVE-2017-16943.
235 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
236 from SMTP input. Previously it was always done; now only done for DATA
237 and not BDAT commands. CVE-2017-16944.
239 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
240 to the message (such as an overlong header line). Previously this was
241 not done and we did not exit BDAT mode. Followon from the previous item
242 though a different problem.
248 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
249 than -2003 did; needs libidn2 in addition to libidn.
251 JH/02 The path option on a pipe transport is now expanded before use.
253 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
254 Patch provided by "Björn", documentation fix added too.
256 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
257 missing a wire-to-host endian conversion.
259 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
260 close after a BDAT command line could be taken as a following command,
261 giving a synch failure. Fix by only checking for synch immediately
262 before acknowledging the chunk.
264 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
265 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
266 macro. Patches provided by Josh Soref.
268 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
269 Previously we did not; the RFC seems ambiguous and VRFY is not listed
270 by IANA as a service extension. However, John Klensin suggests that we
273 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
274 the dkim code may be unix-mode line endings rather than smtp wire-format
275 CRLF, so prepend a CR to any bare LF.
277 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
278 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
280 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
281 the first were themselves being wrongly included in the feed into dkim
282 processing; with most chunk sizes in use this resulted in an incorrect
283 body hash calculated value.
285 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
286 DKIM signature block, for verification. Although advised against by
287 standards it is specifically not ruled illegal.
289 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
291 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
292 missing a body hash (the bh= tag).
294 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
295 It seems that HAProxy sends the Proxy Protocol information in clear and
296 only then does a TLS startup, so do the same.
298 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
299 TCP connections (such as for Spamd) unless the daemon successfully set
300 Fast Open mode on its listening sockets. This fixes breakage seen on
301 too-old kernels or those not configured for Fast Open, at the cost of
302 requiring both directions being enabled for TFO, and TFO never being used
303 by non-daemon-related Exim processes.
305 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
306 endings, at least on the first header line. Try to canonify any that get
307 past that check, despite the cost.
309 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
310 now limited to an arbitrary five deep, while parsing addresses with the
311 strip_excess_angle_brackets option enabled.
313 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
314 instead leave the unprompted TLS handshake in socket buffer for the
315 TLS library to consume.
317 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
319 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
321 JH/16 Drop variables when they go out of scope. Memory management drops a whole
322 region in one operation, for speed, and this leaves assigned pointers
323 dangling. Add checks run only under the testsuite which checks all
324 variables at a store-reset and panics on a dangling pointer; add code
325 explicitly nulling out all the variables discovered. Fixes one known
326 bug: a transport crash, where a dangling pointer for $sending_ip_address
327 originally assigned in a verify callout, is re-used.
329 PP/06 Drop '.' from @INC in various Perl scripts.
331 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
333 PP/08 Reduce a number of compilation warnings under clang; building with
334 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
335 should be warning-free.
337 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
339 HS/01 Fix portability problems introduced by PP/08 for platforms where
340 realloc(NULL) is not equivalent to malloc() [SunOS et al].
342 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
343 chunk. This allows us to accept broken chunked messages. We need a more
344 general solution here.
346 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
347 already-broken messages in the queue.
349 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
351 JH/19 Fix reference counting bug in routing-generated-address tracking.
357 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
358 supports it and a size is available (ie. the sending peer gave us one).
360 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
361 years of being deprecated). The replacements are the ACLs
362 acl_smtp_mime and acl_not_smtp_mime.
364 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
365 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
366 or even an in-clear connection were permitted. Now, if the host lookup
367 was dnssec and dane was requested then the host is only used if the
368 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
369 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
370 if one fails this test.
371 This means that a poorly-configured remote DNS will make it incommunicado;
372 but it protects against a DNS-interception attack on it.
374 JH/04 Bug 1810: make continued-use of an open smtp transport connection
375 non-noisy when a race steals the message being considered.
377 JH/05 If main configuration option tls_certificate is unset, generate a
378 self-signed certificate for inbound TLS connections.
380 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
381 in rewrites and routers.
383 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
384 and logged a warning sing 4.83; now they are a configuration file error.
386 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
387 (lacking @domain). Apply the same qualification processing as RCPT.
389 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
391 JH/10 Support ${sha256:} applied to a string (as well as the previous
394 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
395 a cutthrough deliver is pending, as we always want to make a connection.
396 This also avoids re-routing the message when later placing the cutthrough
397 connection after a verify cache hit.
398 Do not update it with the verify result either.
400 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
401 when routing results in more than one destination address.
403 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
404 signing (which inhibits the cutthrough capability). Previously only
405 the presence of an option was tested; now an expansion evaluating as
406 empty is permissible (obviously it should depend only on data available
407 when the cutthrough connection is made).
409 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
410 the relevant preceding SMTP command did not note the pipelining mode.
412 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
413 Previously they were not counted.
415 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
416 as one having no matching records. Previously we deferred the message
417 that needed the lookup.
419 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
420 distinguished as "(=".
422 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
423 for missing MX records. Previously it only worked for missing A records.
425 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
427 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
428 after the data-go-ahead and data-ack. Patch from Jason Betts.
430 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
431 even for a "none" policy. Patch from Tony Meyer.
433 JH/22 Fix continued use of a connection for further deliveries. If a port was
434 specified by a router, it must also match for the delivery to be
437 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
438 When one of the recipients of a message was unsuitable for the connection
439 (has no matching addresses), we lost track of needing to mark it
440 deferred. As a result mail would be lost.
442 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
444 JH/25 Decoding ACL controls is now done using a binary search; the source code
445 takes up less space and should be simpler to maintain. Merge the ACL
446 condition decode tables also, with similar effect.
448 JH/26 Fix problem with one_time used on a redirect router which returned the
449 parent address unchanged. A retry would see the parent address marked as
450 delivered, so not attempt the (identical) child. As a result mail would
453 JH/27 Fix a possible security hole, wherein a process operating with the Exim
454 UID can gain a root shell. Credit to http://www.halfdog.net/ for
455 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
458 JH/28 Enable {spool,log} filesystem space and inode checks as default.
459 Main config options check_{log,spool}_{inodes,space} are now
460 100 inodes, 10MB unless set otherwise in the configuration.
462 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
463 Previously it only applied to the main-section connection policy
466 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
468 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
469 by me. Added RFC7919 DH primes as an alternative.
471 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
472 are not in the system include path.
474 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
475 GnuTLS, when a session startup failed (eg because the client disconnected)
476 Exim did stdio operations after fclose. This was exposed by a recent
477 change which nulled out the file handle after the fclose.
479 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
480 signed directly by the cert-signing cert, rather than an intermediate
481 OCSP-signing cert. This is the model used by LetsEncrypt.
483 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
485 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
486 an incoming connection.
488 HS/02 Bug 1802: Do not half-close the connection after sending a request
491 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
492 fallback to "prime256v1".
494 JH/34 SECURITY: Use proper copy of DATA command in error message.
495 Could leak key material. Remotely exploitable. CVE-2016-9963.
501 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
502 and 3.4.4 - once the server is enabled to respond to an OCSP request
503 it does even when not requested, resulting in a stapling non-aware
504 client dropping the TLS connection.
506 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
507 support variable-length bit vectors. No functional change.
509 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
510 The I= interface field on outgoing lines is now after the H= remote
511 host field, same as incoming lines. There is a separate
512 outgoing_interface log selector which allows you to disable the
515 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
516 If not running log_selector +smtp_connection the mainlog would be held
517 open indefinitely after a "too many connections" event, including to a
518 deleted file after a log rotate. Leave the per net connection logging
519 leaving it open for efficiency as that will be quickly detected by the
520 check on the next write.
522 HS/01 Bug 1671: Fix post transport crash.
523 Processing the wait-<transport> messages could crash the delivery
524 process if the message IDs didn't exist for some reason. When
525 using 'split_spool_directory=yes' the construction of the spool
526 file name failed already, exposing the same netto behaviour.
528 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
529 mime_regex ACL conditions.
531 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
532 to DSN fail messages (bounces): remote IP, remote greeting, remote response
533 to HELO, local diagnostic string.
535 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
536 log line to debug. Even when configured with a tls authenticator many
537 client connections are expected to not authenticate in this way, so
538 an authenticate fail is not an error.
540 HS/02 Add the Exim version string to the process info. This way exiwhat
541 gives some more detail about the running daemon.
543 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
544 matter for fast-change records such as DNSBLs.
546 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
547 retry record, even if constant. There may be multiple transports with
548 different interface settings and the retry behaviour needs to be kept
551 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
553 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
555 JH/10 Bug 840: fix log_defer_output option of pipe transport
557 JH/11 Bug 830: use same host for all RCPTS of a message, even under
558 hosts_randomize. This matters a lot when combined with mua_wrapper.
560 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
561 ${quote_pgsql:<string>} operator.
563 JH/13 Bug 1708: avoid misaligned access in cached lookup.
565 JH/14 Change header file name for freeradius-client. Relevant if compiling
566 with Radius support; from the Gentoo tree and checked under Fedora.
568 JH/15 Bug 1712: Introduce $prdr_requested flag variable
570 JH/16 Bug 1714: Permit an empty string as expansion result for transport
571 option transport_filter, meaning no filtering.
573 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
575 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
576 defaults to "*" (all hosts). The variable is now available when not built
577 with TLS, default unset, mainly to enable keeping the testsuite sane.
578 If a server certificate is not supplied (via tls_certificate) an error is
579 logged, and clients will find TLS connections fail on startup. Presumably
580 they will retry in-clear.
581 Packagers of Exim are strongly encouraged to create a server certificate
582 at installation time.
584 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
585 with the $config_file variable.
587 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
588 in transport context, after the attempt, and per-recipient. The latter type
589 is per host attempted. The event data is the error message, and the errno
590 information encodes the lookup type (A vs. MX) used for the (first) host,
591 and the trailing two digits of the smtp 4xx response.
593 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
594 to write to mainlog (or rejectlog, paniclog) in the window between file
595 creation and permissions/ownership being changed. Particularly affects
596 installations where exicyclog is run as root, rather than exim user;
597 result is that the running daemon panics and dies.
599 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
601 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
602 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
603 "pri" and "weight". Note that the previous implicit priority given by the
604 list order is no longer honoured.
606 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
609 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
610 by defining SUPPORT_SOCKS.
612 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
613 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
614 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
615 variables are renamed to proxy_{local,external}_{address,port}.
617 JH/27 Move Internationalisation support from Experimental to mainline, enabled
618 for a build by defining SUPPORT_I18N
620 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
621 of the query string, and make ${quote_redis:} do that quoting.
623 JH/29 Move Events support from Experimental to mainline, enabled by default
624 and removable for a build by defining DISABLE_EVENT.
626 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
628 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
629 cached by the daemon.
631 JH/32 Move Redis support from Experimental to mainline, enabled for a build
632 by defining LOOKUP_REDIS. The libhiredis library is required.
634 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
635 keys are given for lookup.
637 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
638 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
639 only supported when built with TLS support. The PolarSSL SHA routines
640 are still used when the TLS library is too old for convenient support.
642 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
643 openssl_options), for security. OpenSSL forces this from version 1.1.0
644 server-side so match that on older versions.
646 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
647 allocation for $value could be released as the expansion processing
648 concluded, but leaving the global pointer active for it.
650 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
651 and to use the domains and local_parts ACL conditions.
653 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
654 incorrectly not doubled on cutthrough transmission, hence seen as a
655 body-termination at the receiving system - resulting in truncated mails.
656 Commonly the sender saw a TCP-level error, and retransmitted the message
657 via the normal store-and-forward channel. This could result in duplicates
658 received - but deduplicating mailstores were liable to retain only the
659 initial truncated version.
661 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
663 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
665 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
666 we're in there, support oversigning also; bug 1309.
668 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
670 HS/04 Add support for keep_environment and add_environment options.
672 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
673 either intentional arithmetic overflow during PRNG, or testing config-
676 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
677 delivery resulted in actual delivery. Cancel cutthrough before DATA
680 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
681 rejects a recipient: pass the reject to the originator.
683 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
684 Many were false-positives and ignorable, but it's worth fixing the
687 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
688 for the new environment-manipulation done at startup. Move the routines
689 from being local to tls.c to being global via the os.c file.
691 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
692 an extract embedded as result-arg for a map, the first arg for extract
693 is unavailable so we cannot tell if this is a numbered or keyed
694 extraction. Accept either.
700 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
703 JH/02 The smtp transport option "multi_domain" is now expanded.
705 JH/03 The smtp transport now requests PRDR by default, if the server offers
708 JH/04 Certificate name checking on server certificates, when exim is a client,
709 is now done by default. The transport option tls_verify_cert_hostnames
710 can be used to disable this per-host. The build option
711 EXPERIMENTAL_CERTNAMES is withdrawn.
713 JH/05 The value of the tls_verify_certificates smtp transport and main options
714 default to the word "system" to access the system default CA bundle.
715 For GnuTLS, only version 3.0.20 or later.
717 JH/06 Verification of the server certificate for a TLS connection is now tried
718 (but not required) by default. The verification status is now logged by
719 default, for both outbound TLS and client-certificate supplying inbound
722 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
725 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
726 Status Notification (bounce) messages are now MIME format per RFC 3464.
727 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
728 under the control of the dsn_advertise_hosts option, and routers may
729 have a dsn_lasthop option.
731 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
732 default, modifiable by a malware= option. The list separator for
733 the options can now be changed in the usual way. Bug 68.
735 JH/10 The smtp_receive_timeout main option is now expanded before use.
737 JH/11 The incoming_interface log option now also enables logging of the
738 local interface on delivery outgoing connections.
740 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
741 if the interface and destination host and port all match.
743 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
746 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
747 Patch from Andrew Lewis.
749 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
750 now supports optional time-restrictions, weighting, and priority
751 modifiers per server. Patch originally by <rommer@active.by>.
753 JH/16 The spamd_address main option now supports a mixed list of local
754 and remote servers. Remote servers can be IPv6 addresses, and
755 specify a port-range.
757 JH/17 Bug 68: The spamd_address main option now supports an optional
758 timeout value per server.
760 JH/18 Bug 1581: Router and transport options headers_add/remove can
761 now have the list separator specified.
763 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
766 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
769 JH/21 Support for the A6 type of dns record is withdrawn.
771 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
772 rather than the verbs used.
774 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
775 from 255 to 1024 chars.
777 JH/24 Verification callouts now attempt to use TLS by default.
779 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
780 are generic router options now. The defaults didn't change.
782 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
783 Original patch from Alexander Shikoff, worked over by JH.
785 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
786 files if ZCAT_COMMAND is not executable.
788 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
790 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
792 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
793 Normally benign, it bites when the pair was led to by a CNAME;
794 modern usage is to not canonicalize the domain to a CNAME target
795 (and we were inconsistent anyway for A-only vs AAAA+A).
797 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
799 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
800 when evaluating $sender_host_dnssec.
802 JH/31 Check the HELO verification lookup for DNSSEC, adding new
803 $sender_helo_dnssec variable.
805 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
807 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
809 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
811 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
812 documented as working, but never had. Support all but $spam_report.
814 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
815 added for tls authenticator.
817 HS/03 Add perl_taintmode main config option
823 TL/01 When running the test suite, the README says that variables such as
824 no_msglog_check are global and can be placed anywhere in a specific
825 test's script, however it was observed that placement needed to be near
826 the beginning for it to behave that way. Changed the runtest perl
827 script to read through the entire script once to detect and set these
828 variables, reset to the beginning of the script, and then run through
829 the script parsing/test process like normal.
831 TL/02 The BSD's have an arc4random API. One of the functions to induce
832 adding randomness was arc4random_stir(), but it has been removed in
833 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
834 function when detected.
836 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
837 cause callback expansion.
839 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
840 syntax errors in an expansion can be treated as a string instead of
841 logging or causing an error, due to the internal use of bool_lax
842 instead of bool when processing it.
844 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
845 server certificates when making smtp deliveries.
847 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
849 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
851 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
853 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
854 Merged patch from Sebastian Wiedenroth.
856 JH/05 Fix results-pipe from transport process. Several recipients, combined
857 with certificate use, exposed issues where response data items split
858 over buffer boundaries were not parsed properly. This eventually
859 resulted in duplicates being sent. This issue only became common enough
860 to notice due to the introduction of connection certificate information,
861 the item size being so much larger. Found and fixed by Wolfgang Breyha.
863 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
864 size buffer was used, resulting in syntax errors when an expansion
867 JH/07 Add support for directories of certificates when compiled with a GnuTLS
868 version 3.3.6 or later.
870 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
871 is EXPERIMENTAL_EVENT, the main-configuration and transport options
872 both become "event_action", the variables become $event_name, $event_data
873 and $event_defer_errno. There is a new variable $verify_mode, usable in
874 routers, transports and related events. The tls:cert event is now also
875 raised for inbound connections, if the main configuration event_action
878 TL/06 In test suite, disable OCSP for old versions of openssl which contained
879 early OCSP support, but no stapling (appears to be less than 1.0.0).
881 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
882 server certificate names available under the smtp transport option
883 "tls_verify_cert_hostname" now do not permit multi-component wildcard
886 JH/10 Time-related extraction expansions from certificates now use the main
887 option "timezone" setting for output formatting, and are consistent
888 between OpenSSL and GnuTLS compilations. Bug 1541.
890 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
891 encoded parameter in the incoming message. Bug 1558.
893 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
894 include certificate info, eximon was claiming there were spoolfile
897 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
899 JH/14 Log delivery-related information more consistently, using the sequence
900 "H=<name> [<ip>]" wherever possible.
902 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
903 are problematic for Debian distribution, omit them from the release
906 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
908 JH/16 Fix string representation of time values on 64bit time_t architectures.
911 JH/17 Fix a null-indirection in certextract expansions when a nondefault
912 output list separator was used.
917 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
918 checkers that were complaining about end of non-void function with no
921 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
922 This was a regression introduced in 4.83 by another bugfix.
924 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
926 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
927 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
933 TF/01 Correctly close the server side of TLS when forking for delivery.
935 When a message was received over SMTP with TLS, Exim failed to clear up
936 the incoming connection properly after forking off the child process to
937 deliver the message. In some situations the subsequent outgoing
938 delivery connection happened to have the same fd number as the incoming
939 connection previously had. Exim would try to use TLS and fail, logging
940 a "Bad file descriptor" error.
942 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
943 utilities have not been installed.
945 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
946 temporary space as the ACL may create new global variables.
948 TL/01 LDAP support uses per connection or global context settings, depending
949 upon the detected version of the libraries at build time.
951 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
952 to extract and use the src ip:port in logging and expansions as if it
953 were a direct connection from the outside internet. PPv2 support was
954 updated based on HAProxy spec change in May 2014.
956 JH/02 Add ${listextract {number}{list}{success}{fail}}.
958 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
959 Properly escape header and check for NULL return.
961 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
964 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
966 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
967 characters in header names, implemented as a verify condition.
968 Contributed by Michael Fischer v. Mollard.
970 TL/05 Rename SPF condition results err_perm and err_temp to standardized
971 results permerror and temperror. Previous values are deprecated but
972 still accepted. In a future release, err_perm and err_temp will be
973 completely removed, which will be a backward incompatibility if the
974 ACL tests for either of these two old results. Patch contributed by
975 user bes-internal on the mailing list.
977 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
979 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
980 selectors, in both main and reject logs.
982 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
985 JH/07 Add malware type "sock" for talking to simple daemon.
987 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
989 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
990 routers/transports under cutthrough routing.
992 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
993 numbers. Touch up "bool" conditional to keep the same definition.
995 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
997 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
999 JH/12 Expand items in router/transport headers_add or headers_remove lists
1000 individually rather than the list as a whole. Bug 1452.
1002 Required for reasonable handling of multiple headers_ options when
1003 they may be empty; requires that headers_remove items with embedded
1004 colons must have them doubled (or the list-separator changed).
1006 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1007 view the policy declared in the DMARC record. Currently, $dmarc_status
1008 is a combined value of both the record presence and the result of the
1011 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1013 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1014 dnslookup router and the smtp transport (applying to the forward
1017 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1018 of ldap servers used for a specific lookup. Patch provided by Heiko
1021 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1022 New variable $lookup_dnssec_authenticated for observability.
1024 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1025 Patch submitted by Lars Timman.
1027 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1029 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1030 Requires trusted mode and valid format message id, aborts otherwise.
1031 Patch contributed by Heiko Schlichting.
1033 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1034 certextract with support for various fields. Bug 1358.
1036 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1037 is requested by default, modifiable by smtp transport option
1040 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1041 operate on certificate variables to give certificate fingerprints
1042 Also new ${sha256:cert_variable}.
1044 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1046 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1047 Christian Aistleitner.
1049 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1051 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1052 file. Patch from Wolfgang Breyha.
1054 JH/25 Expand the coverage of the delivery $host and $host_address to
1055 client authenticators run in verify callout. Bug 1476.
1057 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1058 align with daemon_smtp_ports. Bug 72.
1060 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1061 support and error reporting did not work properly.
1063 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1064 and is readable. Patch from Andrew Colin Kissa.
1066 TL/14 Enhance documentation of ${run expansion and how it parses the
1067 commandline after expansion, particularly in the case when an
1068 unquoted variable expansion results in an empty value.
1070 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1072 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1073 by renaming away from it. Addresses GH issue 15, reported by
1076 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1078 TL/15 SECURITY: prevent double expansion in math comparison functions
1079 (can expand unsanitized data). Not remotely exploitable.
1086 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1088 PP/02 Make -n do something, by making it not do something.
1089 When combined with -bP, the name of an option is not output.
1091 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1094 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1095 $sender_host_name and config options to manage this, and basic check
1098 PP/05 DSCP support for outbound connections and control modifier for inbound.
1100 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1101 (Only plugin which currently uses this is kerberos4, which nobody should
1102 be using, but we should make it available and other future plugins might
1103 conceivably use it, even though it would break NAT; stuff *should* be
1104 using channel bindings instead).
1106 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1107 name; added for Sendmail compatibility; requires admin caller.
1108 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1109 just ignore it); requires trusted caller.
1110 Also parse but ignore: -Ac -Am -X<logfile>
1113 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1115 TL/02 Add +smtp_confirmation as a default logging option.
1117 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1118 Patch by Magnus Holmgren from 2007-02-20.
1120 TL/04 Bugzilla 1281 - Spec typo.
1121 Bugzilla 1283 - Spec typo.
1122 Bugzilla 1290 - Spec grammar fixes.
1124 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1126 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1128 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1129 mailing list by by Dmitry Isaikin.
1131 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1133 JH/02 Support "G" suffix to numbers in ${if comparisons.
1135 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1137 NM/01 Bugzilla 1197 - Spec typo
1138 Bugzilla 1196 - Spec examples corrections
1140 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1142 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1143 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1146 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1147 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1149 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1150 "acl {{name}{arg}...}", and optional args on acl condition
1153 JH/05 Permit multiple router/transport headers_add/remove lines.
1155 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1157 JH/07 Avoid using a waiting database for a single-message-only transport.
1158 Performance patch from Paul Fisher. Bugzilla 1262.
1160 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1163 JH/09 Add $headers_added variable, with content from use of ACL modifier
1164 add_header (but not yet added to the message). Bugzilla 199.
1166 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1167 Pulled from Bugzilla 817 by Wolfgang Breyha.
1169 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1171 (nb: this is the same fix as in Exim 4.80.1)
1173 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1176 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1179 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1181 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1182 advertise SMTP AUTH mechanism to us, instead of a generic
1183 protocol violation error. Also, make Exim more robust to bad
1184 data from the Dovecot auth socket.
1186 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1188 When a queue runner is handling a message, Exim first routes the
1189 recipient addresses, during which it prunes them based on the retry
1190 hints database. After that it attempts to deliver the message to
1191 any remaining recipients. It then updates the hints database using
1194 So if a recipient address works intermittently, it can get repeatedly
1195 deferred at routing time. The retry hints record remains fresh so the
1196 address never reaches the final cutoff time.
1198 This is a fairly common occurrence when a user is bumping up against
1199 their storage quota. Exim had some logic in its local delivery code
1200 to deal with this. However it did not apply to per-recipient defers
1201 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1203 This change adds a proper retry rule check during routing so that the
1204 final cutoff time is checked against the message's age. We only do
1205 this check if there is an address retry record and there is not a
1206 domain retry record; this implies that previous attempts to handle
1207 the address had the retry_use_local_parts option turned on. We use
1208 this as an approximation for the destination being like a local
1209 delivery, as in LMTP.
1211 I suspect this new check makes the old local delivery cutoff check
1212 redundant, but I have not verified this so I left the code in place.
1214 TF/02 Correct gecos expansion when From: is a prefix of the username.
1216 Test 0254 submits a message to Exim with the header
1220 When I ran the test suite under the user fanf2, Exim expanded
1221 the header to contain my full name, whereas it should have added
1222 a Resent-Sender: header. It erroneously treats any prefix of the
1223 username as equal to the username.
1225 This change corrects that bug.
1227 GF/01 DCC debug and logging tidyup
1228 Error conditions log to paniclog rather than rejectlog.
1229 Debug lines prefixed by "DCC: " to remove any ambiguity.
1231 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1233 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1234 Bug spotted by Jeremy Harris; was flawed since initial commit.
1235 Would have resulted in OCSP responses post-SNI triggering an Exim
1236 NULL dereference and crash.
1238 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1240 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1241 Bug detection, analysis and fix by Samuel Thibault.
1242 Bugzilla 1331, Debian bug #698092.
1244 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1246 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1247 Server implementation by Todd Lyons, client by JH.
1248 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1249 config variable "prdr_enable" controls whether the server
1250 advertises the facility. If the client requests PRDR a new
1251 acl_data_smtp_prdr ACL is called once for each recipient, after
1252 the body content is received and before the acl_smtp_data ACL.
1253 The client is controlled by both of: a hosts_try_prdr option
1254 on the smtp transport, and the server advertisement.
1255 Default client logging of deliveries and rejections involving
1256 PRDR are flagged with the string "PRDR".
1258 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1259 fclose(). Diagnosis by Todd Lyons.
1261 PP/17 Update configure.default to handle IPv6 localhost better.
1262 Patch by Alain Williams (plus minor tweaks).
1265 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1266 This is now consistent with GnuTLS, and is now documented: the
1267 previous undocumented portable approach to treating the option as
1268 unset was to force an expansion failure. That still works, and
1269 an empty string is now equivalent.
1271 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1272 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1273 not performing validation itself.
1275 PP/20 Added force_command boolean option to pipe transport.
1276 Patch from Nick Koston, of cPanel Inc.
1278 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1281 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1283 PP/21 Fix eximon continuous updating with timestamped log-files.
1284 Broken in a format-string cleanup in 4.80, missed when I repaired the
1285 other false fix of the same issue.
1286 Report and fix from Heiko Schlichting.
1289 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1290 Report from Prashanth Katuri.
1292 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1293 It's SecureTransport, so affects any MacOS clients which use the
1294 system-integrated TLS libraries, including email clients.
1296 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1297 using a MIME ACL for non-SMTP local injection.
1298 Report and assistance in diagnosis by Warren Baker.
1300 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1302 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1304 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1305 last id that failed so it may be referenced in subsequent ACL's.
1307 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1310 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1311 ldap library initialization, allowing self-signed CA's to be
1312 used. Also properly sets require_cert option later in code by
1313 using NULL (global ldap config) instead of ldap handle (per
1314 session). Bug diagnosis and testing by alxgomz.
1316 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1317 the src/util/ subdirectory.
1319 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1320 renamed to Transport Post Delivery Action by Jeremy Harris, as
1323 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1324 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1325 redis_servers = needs to be configured which will be used by the redis
1326 lookup. Patch from Warren Baker, of The Packet Hub.
1328 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1330 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1331 hostname or reverse DNS when processing a host list. Used suggestions
1332 from multiple comments on this bug.
1334 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1336 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1337 Missed a few lines, added it to make the runtest require no keyboard
1340 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1341 contains upper case chars. Make router use caseful_local_part.
1343 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1344 support when GnuTLS has been built with p11-kit.
1350 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1352 This, or similar/improved, will also be change PP/11 of 4.82.
1358 PP/01 Handle short writes when writing local log-files.
1359 In practice, only affects FreeBSD (8 onwards).
1360 Bugzilla 1053, with thanks to Dmitry Isaikin.
1362 NM/01 Bugzilla 949 - Documentation tweak
1364 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1367 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1369 PP/02 Implemented gsasl authenticator.
1371 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1373 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1374 `pkg-config foo` for cflags/libs.
1376 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1377 with rest of GSASL and with heimdal_gssapi.
1379 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1380 `pkg-config foo` for cflags/libs for the TLS implementation.
1382 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1383 properties get this fed in as external SSF. A number of robustness
1384 and debugging improvements to the cyrus_sasl authenticator.
1386 PP/08 cyrus_sasl server now expands the server_realm option.
1388 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1389 Patch by Jeremy Harris.
1391 PP/10 Added dbmjz lookup type.
1393 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1395 PP/12 MAIL args handles TAB as well as SP, for better interop with
1396 non-compliant senders.
1397 Analysis and variant patch by Todd Lyons.
1399 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1400 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1401 Patch from Dirk Mueller <dmueller@suse.com>
1403 PP/13 tls_peerdn now print-escaped for spool files.
1404 Observed some $tls_peerdn in wild which contained \n, which resulted
1405 in spool file corruption.
1407 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1408 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1409 or write after TLS renegotiation, which otherwise led to messages
1412 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1413 as a tracking header (ie: a signed header comes before the signature).
1414 Patch from Wolfgang Breyha.
1416 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1417 comma-sep list; embedded commas doubled.
1419 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1421 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1423 Report and patch from Dmitry Banschikov.
1425 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1426 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1427 failures. We appear to now support TLS1.1+ with Exim.
1429 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1430 lets Exim select keys and certificates based upon TLS SNI from client.
1431 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1432 before an outbound SMTP session. New log_selector, +tls_sni.
1434 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1435 NULL dereference. Report and patch from Alun Jones.
1437 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1438 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1439 Not seeing resolver debug output on NetBSD, but suspect this is a
1440 resolver implementation change.
1442 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1443 Left warnings. Added "eximon gdb" invocation mode.
1445 PP/21 Defaulting "accept_8bitmime" to true, not false.
1447 PP/22 Added -bw for inetd wait mode support.
1449 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1450 locate the relevant includes and libraries. Made this the default.
1452 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1453 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1455 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1456 This may cause build issues on older platforms.
1458 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1459 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1460 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1461 Added SNI support via GnuTLS too.
1462 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1464 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1466 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1467 Applied second patch from Janne, implementing suggestion to default
1468 multiple-strings-in-record handling to match SPF spec.
1470 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1472 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1473 read-only, out of scope).
1474 Patch from Wolfgang Breyha, report from Stuart Northfield.
1476 PP/29 Fix three issues highlighted by clang analyser static analysis.
1477 Only crash-plausible issue would require the Cambridge-specific
1478 iplookup router and a misconfiguration.
1479 Report from Marcin Mirosław.
1481 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1483 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1484 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1485 As part of this, removing so much warning spew let me fix some minor
1486 real issues in debug logging.
1488 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1489 assignment on my part. Fixed.
1491 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1492 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1493 Janne Snabb (who went above and beyond: thank you).
1495 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1496 string otherwise requires a connection and a bunch more work and it's
1497 relatively easy to get wrong. Should also expose TLS library linkage
1500 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1501 64-bit ${eval} (JH/03).
1503 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1504 GNU libc to support some of the 64-bit stuff, should not lead to
1505 conflicts. Defined before os.h is pulled in, so if a given platform
1506 needs to override this, it can.
1508 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1509 protection layer was required, which is not implemented.
1510 Bugzilla 1254, patch from Wolfgang Breyha.
1512 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1513 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1514 tls_dhparam take prime identifiers. Also unbreak combination of
1515 OpenSSL+DH_params+TLSSNI.
1517 PP/39 Disable SSLv2 by default in OpenSSL support.
1523 PP/01 Solaris build fix for Oracle's LDAP libraries.
1524 Bugzilla 1109, patch from Stephen Usher.
1526 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1528 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1531 TF/02 Fix a couple more cases where we did not log the error message
1532 when unlink() failed. See also change 4.74-TF/03.
1534 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1535 lock up or crash if it happened to be inside a call to libc when it
1536 got a SIGUSR1 from exiwhat.
1538 The SIGUSR1 handler appends the current process status to the process
1539 log which is later printed by exiwhat. It used to use the general
1540 purpose logging code to do this, but several functions it calls are
1541 not safe for signals.
1543 The new output code in the SIGUSR1 handler is specific to the process
1544 log, and simple enough that it's easy to inspect for signal safety.
1545 Removing some special cases also simplifies the general logging code.
1546 Removing the spurious timestamps from the process log simplifies
1549 TF/04 Improved ratelimit ACL condition.
1551 The /noupdate option has been deprecated in favour of /readonly which
1552 has clearer semantics. The /leaky, /strict, and /readonly update modes
1553 are mutually exclusive. The update mode is no longer included in the
1554 database key; it just determines when the database is updated. (This
1555 means that when you upgrade Exim will forget old rate measurements.)
1557 Exim now checks that the per_* options are used with an update mode that
1558 makes sense for the current ACL. For example, when Exim is processing a
1559 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1560 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1561 must specify per_mail/readonly. If you omit the update mode it defaults to
1562 /leaky where that makes sense (as before) or /readonly where required.
1564 The /noupdate option is now undocumented but still supported for
1565 backwards compatibility. It is equivalent to /readonly except that in
1566 ACLs where /readonly is required you may specify /leaky/noupdate or
1567 /strict/noupdate which are treated the same as /readonly.
1569 A useful new feature is the /count= option. This is a generalization
1570 of the per_byte option, so that you can measure the throughput of other
1571 aggregate values. For example, the per_byte option is now equivalent
1572 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1574 The per_rcpt option has been generalized using the /count= mechanism
1575 (though it's more complicated than the per_byte equivalence). When it is
1576 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1577 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1578 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1579 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1580 Note that using per_rcpt with a non-readonly update mode in more than
1581 one ACL will cause the recipients to be double-counted. (The per_mail
1582 and per_byte options don't have this problem.)
1584 The handling of very low rates has changed slightly. If the computed rate
1585 is less than the event's count (usually one) then this event is the first
1586 after a long gap. In this case the rate is set to the same as this event's
1587 count, so that the first message of a spam run is counted properly.
1589 The major new feature is a mechanism for counting the rate of unique
1590 events. The new per_addr option counts the number of different
1591 recipients that someone has sent messages to in the last time period. It
1592 behaves like per_rcpt if all the recipient addresses are different, but
1593 duplicate recipient addresses do not increase the measured rate. Like
1594 the /count= option this is a general mechanism, so the per_addr option
1595 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1596 example, measure the rate that a client uses different sender addresses
1597 with the options per_mail/unique=$sender_address. There are further
1598 details in the main documentation.
1600 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1602 TF/06 Removed a few PCRE remnants.
1604 TF/07 Automatically extract Exim's version number from tags in the git
1605 repository when doing development or release builds.
1607 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1608 Bugzilla 879. Patch from Paul Fisher.
1610 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1611 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1614 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1616 PP/05 Variable $av_failed, true if the AV scanner deferred.
1617 Bugzilla 1078. Patch from John Horne.
1619 PP/06 Stop make process more reliably on build failure.
1620 Bugzilla 1087. Patch from Heiko Schlittermann.
1622 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1623 Bugzilla 1089. Patch from Heiko Schlittermann.
1625 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1626 Bugzilla 1131. Patch from Holger Weiß.
1628 PP/09 Handle IPv6 addresses with SPF.
1629 Bugzilla 860. Patch from Wolfgang Breyha.
1631 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1633 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1636 PP/11 match_* no longer expand right-hand-side by default.
1637 New compile-time build option, EXPAND_LISTMATCH_RHS.
1638 New expansion conditions, "inlist", "inlisti".
1640 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1642 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1644 PP/14 fix log_write() format string regression from TF/03.
1645 Bugzilla 1152. Patch from Dmitry Isaikin.
1651 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1653 PP/02 Harmonised TLS library version reporting; only show if debugging.
1654 Layout now matches that introduced for other libraries in 4.74 PP/03.
1656 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1658 PP/04 New "dns_use_edns0" global option.
1660 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1663 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1664 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1666 TK/01 Updated PolarSSL code to 0.14.2.
1667 Bugzilla 1097. Patch from Andreas Metzler.
1669 PP/07 Catch divide-by-zero in ${eval:...}.
1670 Fixes bugzilla 1102.
1672 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1675 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1676 format-string attack -- SECURITY: remote arbitrary code execution.
1678 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1679 time unintentionally subject to list matching rules, letting the header
1680 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1681 arbitrary string expansion). This allowed for information disclosure.
1683 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1684 INT_MIN/-1 -- value coerced to INT_MAX.
1690 NM/01 Workaround for PCRE version dependency in version reporting
1693 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1694 This fixes portability to compilers other than gcc, notably
1695 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1697 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1698 makefiles for portability to HP-UX and POSIX correctness.
1700 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1701 Also via indented variable definition in the Makefile.
1702 (Debugging by Oliver Heesakkers).
1704 PP/02 Restore caching of spamd results with expanded spamd_address.
1705 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1707 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1708 Improves build reliability. Fix from: Frank Elsner
1710 NM/02 Fix wide character breakage in the rfc2047 coding
1711 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1713 NM/03 Allow underscore in dnslist lookups
1714 Fixes bug 1026. Patch from Graeme Fowler
1716 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1717 Code patches from Adam Ciarcinski of NetBSD.
1719 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1722 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1723 is logged, to avoid truncation. Patch from John Horne.
1725 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1726 Patch from Jakob Hirsch.
1728 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1729 SQL string expansion failure details.
1730 Patch from Andrey Oktyabrski.
1732 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1733 Patch from Simon Arlott.
1735 PP/09 New lookups functionality failed to compile on old gcc which rejects
1736 extern declarations in function scope.
1737 Patch from Oliver Fleischmann
1739 PP/10 Use sig_atomic_t for flags set from signal handlers.
1740 Check getgroups() return and improve debugging.
1741 Fixed developed for diagnosis in bug 927 (which turned out to be
1744 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1745 Patch from Mark Zealey.
1747 PP/12 Bugzilla 1056: Improved spamd server selection.
1748 Patch from Mark Zealey.
1750 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1751 Based on patch from Heiko Schlittermann.
1753 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1754 Patch from Uwe Doering, sign-off by Michael Haardt.
1756 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1757 variable declaration deep within a block. Bug and patch from
1760 PP/15 lookups-Makefile IRIX compatibility coercion.
1762 PP/16 Make DISABLE_DKIM build knob functional.
1764 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1765 Patch by Simon Arlott
1767 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1768 variable argument macros. Our copy now differs from upstream.
1774 TF/01 Failure to get a lock on a hints database can have serious
1775 consequences so log it to the panic log.
1777 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1778 controlled using the smtp_confirmation log selector.
1780 TF/03 Include the error message when we fail to unlink a spool file.
1782 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1783 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1784 for maintaining out-of-tree patches for some time.
1786 PP/01 Bugzilla 139: Documentation and portability issues.
1787 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1788 Handle per-OS dynamic-module compilation flags.
1790 PP/02 Let /dev/null have normal permissions.
1791 The 4.73 fixes were a little too stringent and complained about the
1792 permissions on /dev/null. Exempt it from some checks.
1793 Reported by Andreas M. Kirchwitz.
1795 PP/03 Report version information for many libraries, including
1796 Exim version information for dynamically loaded libraries. Created
1797 version.h, now support a version extension string for distributors
1798 who patch heavily. Dynamic module ABI change.
1800 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1801 privilege escalation vulnerability whereby the Exim run-time user
1802 can cause root to append content of the attacker's choosing to
1805 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1808 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1809 If dropping privileges for untrusted macros, we disabled normal logging
1810 on the basis that it would fail; for the Exim run-time user, this is not
1811 the case, and it resulted in successful deliveries going unlogged.
1812 Fixed. Reported by Andreas Metzler.
1818 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1819 only prepend for the Resent-* case. Fixes regression introduced in
1820 Exim 4.70 by NM/22 for Bugzilla 607.
1822 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1823 increasing numbers of administrators be bitten by this.
1825 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1827 PP/03 Bugzilla 994: added openssl_options main configuration option.
1829 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1831 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1833 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1835 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1836 without a peer certificate, leading to a segfault because of an
1837 assumption that peers always have certificates. Be a little more
1838 paranoid. Problem reported by Martin Tscholak.
1840 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1841 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1842 NB: ClamAV planning to remove STREAM in "middle of 2010".
1843 CL also introduces -bmalware, various -d+acl logging additions and
1844 more caution in buffer sizes.
1846 PP/09 Implemented reverse_ip expansion operator.
1848 PP/10 Bugzilla 937: provide a "debug" ACL control.
1850 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1852 PP/12 Bugzilla 973: Implement --version.
1854 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1856 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1858 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1860 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1861 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1862 ignore trailing whitespace.
1864 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1866 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1869 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1870 Notification from Dr Andrew Aitchison.
1872 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1873 ExtendedDetectionInfo response format.
1874 Notification from John Horne.
1876 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1879 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1880 XSL and documented dependency on system catalogs, with examples of how
1883 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1886 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1887 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1888 configuration file which is writeable by the Exim user or group.
1890 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1891 of configuration files to cover files specified with the -C option if
1892 they are going to be used with root privileges, not just the default
1895 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1896 option (effectively making it always true).
1898 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1899 files to be used while preserving root privileges.
1901 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1902 that rogue child processes cannot use them.
1904 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1905 run-time user, instead of root.
1907 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1908 Exim run-time user without dropping privileges.
1910 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1911 result string, instead of calling string_vformat() twice with the same
1914 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1915 for other users. Others should always drop root privileges if they use
1916 -C on the command line, even for a whitelisted configure file.
1918 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1920 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1926 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1927 $data_path, and $header_path variables; fixed documentation bugs and
1930 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1931 exipick to access non-standard spools, including the "frozen" queue
1934 NM/01 Bugzilla 965: Support mysql stored procedures.
1935 Patch from Alain Williams
1937 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1939 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1940 Patch from Andreas Metzler
1942 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1943 Patch from Kirill Miazine
1945 NM/05 Bugzilla 671: Added umask to procmail example.
1947 JJ/03 installed exipick 20100323.0, fixing doc bug
1949 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1950 directory. Notification and patch from Dan Rosenberg.
1952 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1954 TK/02 Improve log output when DKIM signing operation fails.
1956 MH/01 Treat the transport option dkim_domain as a colon separated
1957 list, not as a single string, and sign the message with each element,
1958 omitting multiple occurences of the same signer.
1960 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1961 Bugzilla 985, 986. Patch by Simon Arlott
1963 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1964 Patch by Simon Arlott
1966 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1967 MBX locking. Notification from Dan Rosenberg.
1973 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1975 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1977 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1979 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1981 NM/04 Bugzilla 915: Flag broken perl installation during build.
1987 TK/01 Added patch by Johannes Berg that expands the main option
1988 "spamd_address" if it starts with a dollar sign.
1990 TK/02 Write list of recipients to X-Envelope-Sender header when building
1991 the mbox-format spool file for content scanning (suggested by Jakob
1994 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1995 (http://www.dcc-servers.net/) support via dccifd. Activated by
1996 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1998 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1999 by Mark Daniel Reidel <mr@df.eu>.
2001 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2002 When building exim an external PCRE library is now needed -
2003 PCRE is a system library on the majority of modern systems.
2004 See entry on PCRE_LIBS in EDITME file.
2006 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2007 conversation. Added nologin parameter to request.
2008 Patch contributed by Kirill Miazine.
2010 TF/01 Do not log submission mode rewrites if they do not change the address.
2012 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2014 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2015 log files in place. Contributed by Roberto Lima.
2017 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2019 TF/03 Bugzilla 615: When checking the local_parts router precondition
2020 after a local_part_suffix or local_part_prefix option, Exim now
2021 does not use the address's named list lookup cache, since this
2022 contains cached lookups for the whole local part.
2024 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2025 Robert Millan. Documentation is in experimental-spec.txt.
2027 TF/04 Bugzilla 668: Fix parallel build (make -j).
2029 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2031 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2032 Patch provided by Jan Srzednicki.
2034 TF/05 Leading white space used to be stripped from $spam_report which
2035 wrecked the formatting. Now it is preserved.
2037 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2038 that they are available at delivery time.
2040 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2042 TF/08 TLS error reporting now respects the incoming_interface and
2043 incoming_port log selectors.
2045 TF/09 Produce a more useful error message if an SMTP transport's hosts
2046 setting expands to an empty string.
2048 NM/06 Bugzilla 744: EXPN did not work under TLS.
2049 Patch provided by Phil Pennock.
2051 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2052 Patch provided by Richard Godbee.
2054 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2055 acl_smtp_notquit, added index entry.
2057 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2058 Patch provided by Eugene Bujak.
2060 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2061 accept(). Patch provided by Maxim Dounin.
2063 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2064 Patch provided by Phil Pennock.
2066 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2068 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2069 Patch provided by Brad "anomie" Jorsch.
2071 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2072 Patch provided by Dean Brooks.
2074 TK/05 Add native DKIM support (does not depend on external libraries).
2076 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2077 Patch provided by Graeme Fowler.
2079 NM/16 Bugzilla 851: Documentation example syntax fix.
2081 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2083 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2086 NM/19 Bugzilla 745: TLS version reporting.
2087 Patch provided by Phil Pennock.
2089 NM/20 Bugzilla 167: bool: condition support.
2090 Patch provided by Phil Pennock.
2092 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2093 clients. Patch provided by Phil Pennock.
2095 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2096 Patch provided by Brad "anomie" Jorsch.
2098 NM/23 Bugzilla 687: Fix misparses in eximstats.
2099 Patch provided by Heiko Schlittermann.
2101 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2102 Patch provided by Heiko Schlittermann.
2104 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2105 plus update to original patch.
2107 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2109 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2110 Patch provided by David Brownlee.
2112 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2114 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2116 NM/30 Bugzilla 888: TLS documentation bugfixes.
2118 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2120 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2121 Unlike the original bugzilla I have changed all shell scripts in src tree.
2123 NM/33 Bugzilla 898: Transport filter timeout fix.
2124 Patch by Todd Rinaldo.
2126 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2127 Patch by Serge Demonchaux.
2129 NM/35 Bugzilla 39: Base64 decode bug fixes.
2130 Patch by Jakob Hirsch.
2132 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2134 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2136 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2138 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2144 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2145 ALT-N's libdkim that I have put here:
2146 http://duncanthrax.net/exim-experimental/
2148 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2149 were called 'true' and it seems that C99 defines that as a reserved
2150 keyword to be used with 'bool' variable types. That means you could
2151 not include C99-style headers which use bools without triggering
2152 build errors in sieve.c.
2154 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2155 as mailq or other aliases. Changed the --help handling significantly
2156 to do whats expected. exim_usage() emits usage/help information.
2158 SC/01 Added the -bylocaldomain option to eximstats.
2160 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2162 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2164 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2170 PH/01 Another patch from the Sieve maintainer.
2172 PH/02 When an IPv6 address is converted to a string for single-key lookup
2173 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2174 dots are used instead of colons so that keys in lsearch files need not
2175 contain colons. This was done some time before quoting was made available
2176 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2177 (notated using the quote facility) so as to distinguish them from IPv4
2178 keys. This meant that lookups for IP addresses in host lists did not work
2179 for iplsearch lookups.
2181 This has been fixed by arranging for IPv6 addresses to be expressed with
2182 colons if the lookup type is iplsearch. This is not incompatible, because
2183 previously such lookups could never work.
2185 The situation is now rather anomalous, since one *can* have colons in
2186 ordinary lsearch keys. However, making the change in all cases is
2187 incompatible and would probably break a number of configurations.
2189 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2192 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2193 conversion specification without a maximum field width, thereby enabling
2194 a rogue spamd server to cause a buffer overflow. While nobody in their
2195 right mind would setup Exim to query an untrusted spamd server, an
2196 attacker that gains access to a server running spamd could potentially
2197 exploit this vulnerability to run arbitrary code as the Exim user.
2199 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2200 $primary_hostname instead of what libspf2 thinks the hosts name is.
2202 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2203 a directory entry by the name of the lookup key. Previously, if a
2204 symlink pointed to a non-existing file or a file in a directory that
2205 Exim lacked permissions to read, a lookup for a key matching that
2206 symlink would fail. Now it is enough that a matching directory entry
2207 exists, symlink or not. (Bugzilla 503.)
2209 PH/03 The body_linecount and body_zerocount variables are now exported in the
2212 PH/04 Added the $dnslist_matched variable.
2214 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2215 This means they are set thereafter only if the connection becomes
2218 PH/06 Added the client_condition to authenticators so that some can be skipped
2219 by clients under certain conditions.
2221 PH/07 The error message for a badly-placed control=no_multiline_responses left
2222 "_responses" off the end of the name.
2224 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2226 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2227 (without spaces) instead of just copying the configuration text.
2229 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2231 PH/11 Added $max_received_linelength.
2233 PH/12 Added +ignore_defer and +include_defer to host lists.
2235 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2236 way in which PCRE > 7.0 is built.
2238 PH/14 Implemented queue_only_load_latch.
2240 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2241 MAIL command. The effect was to mangle the value on 64-bit systems.
2243 PH/16 Another patch from the Sieve maintainer.
2245 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2247 PH/18 If a system quota error occurred while trying to create the file for
2248 a maildir delivery, the message "Mailbox is full" was not appended to the
2249 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2250 only to a quota excession during the actual writing of the file.
2252 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2253 characters?) which causes problems in log lines. The DN values are now
2254 passed through string_printing() before being added to log lines.
2256 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2257 and InterBase are left for another time.)
2259 PH/21 Added message_body_newlines option.
2261 PH/22 Guard against possible overflow in moan_check_errorcopy().
2263 PH/23 POSIX allows open() to be a macro; guard against that.
2265 PH/24 If the recipient of an error message contained an @ in the local part
2266 (suitably quoted, of course), incorrect values were put in $domain and
2267 $local_part during the evaluation of errors_copy.
2273 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2274 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2277 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2278 issue a MAIL command.
2280 PH/02 In an ACL statement such as
2282 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2284 if a client was not listed at all, or was listed with a value other than
2285 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2286 the condition was not true (as it should be), so access was not denied.
2287 The bug was that the ! inversion was incorrectly passed on to the second
2288 item. This has been fixed.
2290 PH/03 Added additional dnslists conditions == and =& which are different from
2291 = and & when the dns lookup returns more than one IP address.
2293 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2294 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2296 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2297 FSYNC, which compiles an option called disable_fsync that allows for
2298 bypassing fsync(). The documentation is heavily laced with warnings.
2300 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2302 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2303 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2304 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2305 including adding "make clean"; (3) Added -fPIC when compiling the test
2306 dynamically loaded module, to get rid of a warning.
2308 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2309 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2310 = 0s. The bug is otherwise harmless.
2312 PH/07 There was a bug in the dovecot authenticator such that the value of
2313 $auth1 could be overwritten, and so not correctly preserved, after a
2314 successful authentication. This usually meant that the value preserved by
2315 the server_setid option was incorrect.
2317 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2319 PH/09 Installed PCRE release 7.0.
2321 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2322 run for batched SMTP input. It is now run at the start of every message
2323 in the batch. While fixing this I discovered that the process information
2324 (output by running exiwhat) was not always getting set for -bs and -bS
2325 input. This is fixed, and it now also says "batched" for BSMTP.
2327 PH/11 Added control=no_pipelining.
2329 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2330 patch, slightly modified), and move the expansion of helo_data till after
2331 the connection is made in the smtp transport (so it can use these
2334 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2336 PH/14 Added log_selector = +pid.
2338 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2340 PH/16 Add ${if forany and ${if forall.
2342 PH/17 Added dsn_from option to vary the From: line in DSNs.
2344 PH/18 Flush SMTP output before performing a callout, unless control =
2345 no_callout_flush is set.
2347 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2348 was true (the default) a successful delivery failed to delete the retry
2349 item, thus causing premature timeout of the address. The bug is now
2352 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2354 PH/21 Long custom messages for fakedefer and fakereject are now split up
2355 into multiline responses in the same way that messages for "deny" and
2356 other ACL rejections are.
2358 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2359 with slight modification.
2361 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2362 draft, changing the syntax and factoring some duplicate code".
2364 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2365 for deliveries of the second and subsequent messages over the same SMTP
2368 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2369 ${reduce, with only minor "tidies".
2371 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2373 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2374 expansion side effects.
2376 PH/27 When a message times out after an over-quota error from an Exim-imposed
2377 quota, the bounce message says "mailbox is full". This message was not
2378 being given when it was a system quota that was exceeded. It now should
2381 MH/03 Made $recipients available in local_scan(). local_scan() already has
2382 better access to the recipient list through recipients_list[], but
2383 $recipients can be useful in postmaster-provided expansion strings.
2385 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2386 in the case of a MAIL command with additional options following the
2387 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2388 were accidentally chopped off.
2390 PH/29 SMTP synchronization checks are implemented when a command is read -
2391 there is a check that no more input is waiting when there shouldn't be
2392 any. However, for some commands, a delay in an ACL can mean that it is
2393 some time before the response is written. In this time, more input might
2394 arrive, invalidly. So now there are extra checks after an ACL has run for
2395 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2396 pipelining has not been advertised.
2398 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2400 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2401 returned in the SMTP response when smtp_return_error_details was false.
2402 This has been fixed.
2404 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2405 instead of the C I/O that was originally supplied, because problems were
2406 reported on Solaris.
2408 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2409 Exim which did not show up earlier: it was assuming that a call to
2410 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2411 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2412 was a macro that became an assignment, so it seemed to work. This has
2413 changed to a proper function call with a void return, hence the compile
2414 error. Exim's code has been fixed.
2416 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2419 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2421 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2423 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2424 because it turns out to be too big or there is a timeout) there is no
2425 "Completed" line in the log. When some messages of this type were
2426 selected by exigrep, they were listed as "not completed". Others were
2427 picked up by some special patterns. I have improved the selection
2428 criteria to be more general.
2430 PH/38 The host_find_failed option in the manualroute router can now be set
2431 to "ignore", to completely ignore a host whose IP address cannot be
2432 found. If all hosts are ignored, the behaviour is controlled by the new
2433 host_all_ignored option.
2435 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2436 homing or because of multiple MX records with /mx) generated more than
2437 one IP address, and the following item turned out to be the local host,
2438 all the secondary addresses of the first item were incorrectly removed
2439 from the list, along with the local host and any following hosts (which
2440 is what is supposed to happen).
2442 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2443 whoever called Exim into the -H file. In the case of the daemon it was
2444 behaving confusingly. When first started, it used values for whoever
2445 started the daemon, but after a SIGHUP it used the Exim user (because it
2446 calls itself on a restart). I have changed the code so that it now always
2449 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2450 message are rejected with the same error (e.g. no authentication or bad
2451 sender address), and a DATA command is nevertheless sent (as can happen
2452 with PIPELINING or a stupid MUA), the error message that was given to the
2453 RCPT commands is included in the rejection of the DATA command. This is
2454 intended to be helpful for MUAs that show only the final error to their
2457 PH/42 Another patch from the Sieve maintainer.
2459 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2460 Eximstats - Fixed some broken HTML links and added missing column headers
2462 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2463 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2465 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2467 PH/43 Yet another patch from the Sieve maintainer.
2469 PH/44 I found a way to check for a TCP/IP connection going away before sending
2470 the response to the final '.' that terminates a message, but only in the
2471 case where the client has not sent further data following the '.'
2472 (unfortunately, this is allowed). However, in many cases there won't be
2473 any further data because there won't be any more messages to send. A call
2474 to select() can be used: if it shows that the input is "ready", there is
2475 either input waiting, or the socket has been closed. An attempt to read
2476 the next input character can distinguish the two cases. Previously, Exim
2477 would have sent an OK response which the client would never have see.
2478 This could lead to message repetition. This fix should cure that, at
2479 least in a lot of common cases.
2481 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2482 advertised in response to EHLO.
2488 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2489 fixed by 4.65/MH/01 (is this a record?) are fixed:
2491 (i) An empty string was always treated as zero by the numeric comparison
2492 operators. This behaviour has been restored.
2494 (ii) It is documented that the numeric comparison operators always treat
2495 their arguments as decimal numbers. This was broken in that numbers
2496 starting with 0 were being interpreted as octal.
2498 While fixing these problems I realized that there was another issue that
2499 hadn't been noticed. Values of message_size_limit (both the global option
2500 and the transport option) were treated as octal if they started with 0.
2501 The documentation was vague. These values are now always treated as
2502 decimal, and I will make that clear in the documentation.
2508 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2509 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2512 MH/01 Don't check that the operands of numeric comparison operators are
2513 integers when their expansion is in "skipping" mode (fixes bug
2514 introduced by 4.64-PH/07).
2516 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2517 child addresses, Exim now panics and dies. Previously, because the count
2518 is held in a short int, deliveries were likely to be lost. As such a
2519 large number of recipients for a single message is ridiculous
2520 (performance will be very, very poor), I have chosen to impose a limit
2521 rather than extend the field.
2527 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2528 leftover -K file (the existence of which was triggered by #402).
2529 While we were at it, introduced process PID as part of the -K
2530 filename. This should rule out race conditions when creating
2533 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2534 processing considerably. Previous code took too long for large mails,
2535 triggering a timeout which in turn triggers #401.
2537 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2538 in the DK code in transports.c. sendfile() is not really portable,
2539 hence the _LINUX specificness.
2541 TF/01 In the add_headers option to the mail command in an Exim filter,
2542 there was a bug that Exim would claim a syntax error in any
2543 header after the first one which had an odd number of characters
2546 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2547 callout verification, Exim cached a "reject" for the entire domain. This
2548 is correct for most verifications, but it is not correct for a recipient
2549 verification with use_sender or use_postmaster set, because in that case
2550 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2551 case of MAIL FROM:<> rejection from other early rejections (e.g.
2552 rejection of HELO). When verifying a recipient using a non-null MAIL
2553 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2554 Whatever the result of the callout, the value of the domain cache is
2555 left unchanged (for any other kind of callout, getting as far as trying
2556 RCPT means that the domain itself is ok).
2558 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2561 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2562 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2563 session, the daemon ignores ECONNECT errors and logs others; it now
2564 ignores EPIPE as well.
2566 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2567 (quoted-printable decoding).
2569 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2570 later the small subsequent patch to fix an introduced bug.
2572 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2574 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2576 PH/08 An error is now given if message_size_limit is specified negative.
2578 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2579 to be given (somewhat) arbitrary names.
2581 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2584 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2585 miscellaneous code fixes
2587 PH/10 Added the log_reject_target ACL modifier to specify where to log
2590 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2591 hostname. This is wrong, because it relates to the incoming message (and
2592 probably the interface on which it is arriving) and not to the outgoing
2593 callout (which could be using a different interface). This has been
2594 changed to use the value of the helo_data option from the smtp transport
2595 instead - this is what is used when a message is actually being sent. If
2596 there is no remote transport (possible with a router that sets up host
2597 addresses), $smtp_active_hostname is used.
2599 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2600 tweaks were necessary in order to get it to work (see also 21 below):
2601 (a) The code assumed that strncpy() returns a negative number on buffer
2602 overflow, which isn't the case. Replaced with Exim's string_format()
2604 (b) There were several signed/unsigned issues. I just did the minimum
2605 hacking in of casts. There is scope for a larger refactoring.
2606 (c) The code used strcasecmp() which is not a standard C function.
2607 Replaced with Exim's strcmpic() function.
2608 (d) The code set only $1; it now sets $auth1 as well.
2609 (e) A simple test gave the error "authentication client didn't specify
2610 service in request". It would seem that Dovecot has changed its
2611 interface. Fortunately there's a specification; I followed it and
2612 changed what the client sends and it appears to be working now.
2614 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2617 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2618 address A is aliased to B and C, where B exists and C does not. Without
2619 -v the output is "A verified" because verification stops after a
2620 successful redirection if more than one address is generated. However,
2621 with -v the child addresses are also verified. Exim was outputting "A
2622 failed to verify" and then showing the successful verification for C,
2623 with its parentage. It now outputs "B failed to verify", showing B's
2624 parentage before showing the successful verification of C.
2626 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2627 look up a TXT record in a specific list after matching in a combined
2630 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2631 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2632 they consult the DNS. I had assumed they would set it the way they
2633 wanted; and indeed my experiments on Linux seem to show that in some
2634 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2635 To be on the safe side, however, I have now made the interface to
2636 host_find_byname() similar to host_find_bydns(), with an argument
2637 containing the DNS resolver options. The host_find_byname() function now
2638 sets these options at its start, just as host_find_bydns() does. The smtp
2639 transport options dns_qualify_single and dns_search_parents are passed to
2640 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2641 of host_find_byname() use the default settings of RES_DEFNAMES
2642 (qualify_single) but not RES_DNSRCH (search_parents).
2644 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2645 spool_read_header() do less string testing, by means of a preliminary
2646 switch on the second character of optional "-foo" lines. (This is
2647 overdue, caused by the large number of possibilities that now exist.
2648 Originally there were few.) While I was there, I also converted the
2649 str(n)cmp tests so they don't re-test the leading "-" and the first
2650 character, in the hope this might squeeze out yet more improvement.
2652 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2653 flag allowing group syntax was set by the header_syntax check but not
2654 turned off, possible causing trouble later; (2) The flag was not being
2655 set at all for the header_verify test, causing "group"-style headers to
2656 be rejected. I have now set it in this case, and also caused header_
2657 verify to ignore an empty address taken from a group. While doing this, I
2658 came across some other cases where the code for allowing group syntax
2659 while scanning a header line wasn't quite right (mostly, not resetting
2660 the flag correctly in the right place). These bugs could have caused
2661 trouble for malformed header lines. I hope it is now all correct.
2663 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2664 with the "reply" argument non-NULL. The code, however (which originally
2665 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2666 but it didn't always do it. This confused somebody who was copying the
2667 code for some other use. I have removed all the tests.
2669 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2670 feature that was used to support insecure browsers during the U.S. crypto
2671 embargo. It requires special client support, and Exim is probably the
2672 only MTA that supported it -- and would never use it because real RSA is
2673 always available. This code has been removed, because it had the bad
2674 effect of slowing Exim down by computing (never used) parameters for the
2675 RSA_EXPORT functionality.
2677 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2678 authenticator to fail if there's a tab character in the incoming data
2679 (there should never be unless someone is messing about, as it's supposed
2680 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2681 if the connection is using TLS or if the remote IP is the same as the
2682 local IP, and the "valid-client-cert option" if a client certificate has
2685 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2686 authenticators. This can be used for authorization after authentication
2687 succeeds. (In the case of plaintext, it servers for both authentication
2690 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2691 if any retry times were supplied.
2693 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2694 connection, where there is no client IP address to check. In this
2695 situation, the verify now always succeeds.
2697 PH/25 Applied John Jetmore's -Mset patch.
2699 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2701 PH/27 In a string expansion for a processed (not raw) header when multiple
2702 headers of the same name were present, leading whitespace was being
2703 removed from all of them, but trailing whitespace was being removed only
2704 from the last one. Now trailing whitespace is removed from each header
2705 before concatenation. Completely empty headers in a concatenation (as
2706 before) are ignored.
2708 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2709 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2711 PH/29 [Removed. This was a change that I later backed out, and forgot to
2712 correct the ChangeLog entry (that I had efficiently created) before
2713 committing the later change.]
2715 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2716 address errors (4xx response to RCPT) over the same connection as other
2717 messages routed to the same hosts. Such deliveries are always "forced",
2718 so retry times are not inspected. This resulted in far too many retries
2719 for the affected addresses. The effect occurred only when there were more
2720 hosts than the hosts_max_try setting in the smtp transport when it had
2721 the 4xx errors. Those hosts that it had tried were not added to the list
2722 of hosts for which the message was waiting, so if all were tried, there
2723 was no problem. Two fixes have been applied:
2725 (i) If there are any address or message errors in an SMTP delivery, none
2726 of the hosts (tried or untried) are now added to the list of hosts
2727 for which the message is waiting, so the message should not be a
2728 candidate for sending over the same connection that was used for a
2729 successful delivery of some other message. This seems entirely
2730 reasonable: after all the message is NOT "waiting for some host".
2731 This is so "obvious" that I'm not sure why it wasn't done
2732 previously. Hope I haven't missed anything, but it can't do any
2733 harm, as the worst effect is to miss an optimization.
2735 (ii) If, despite (i), such a delivery is accidentally attempted, the
2736 routing retry time is respected, so at least it doesn't keep
2737 hammering the server.
2739 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2740 in ${readsocket because some servers need this prod.
2742 PH/32 Added some extra debug output when updating a wait-xxx database.
2744 PH/33 The hint "could be header name not terminated by colon", which has been
2745 given for certain expansion errors for a long time, was not being given
2746 for the ${if def:h_colon_omitted{... case.
2748 PH/34 The spec says: "With one important exception, whenever a domain list is
2749 being scanned, $domain contains the subject domain." There was at least
2750 one case where this was not true.
2752 PH/35 The error "getsockname() failed: connection reset by peer" was being
2753 written to the panic log as well as the main log, but it isn't really
2754 panic-worthy as it just means the connection died rather early on. I have
2755 removed the panic log writing for the ECONNRESET error when getsockname()
2758 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2759 runs only) independently of the message's sender address. This meant
2760 that, if the 4xx error was in fact related to the sender, a different
2761 message to the same recipient with a different sender could confuse
2762 things. In particular, this can happen when sending to a greylisting
2763 server, but other circumstances could also provoke similar problems.
2764 I have changed the default so that the retry time for these errors is now
2765 based a combination of the sender and recipient addresses. This change
2766 can be overridden by setting address_retry_include_sender=false in the
2769 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2770 remote server are returned as part of bounce messages. This was not
2771 happening for LMTP over a pipe (the lmtp transport), but now it is the
2772 same for both kinds of LMTP.
2774 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2775 in header lines that were in fact CNAMEs. This is no longer the case.
2777 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2778 and queue runs started by the daemon processed all messages. This has
2779 been fixed so that -R and -S can now usefully be given with -q<time>.
2781 PH/40 Import PCRE release 6.7 (fixes some bugs).
2783 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2785 PH/42 Give an error if -q is specified more than once.
2787 PH/43 Renamed the variables $interface_address and $interface_port as
2788 $received_ip_address and $received_port, to make it clear that these
2789 values apply to message reception, and not to the outgoing interface when
2790 a message is delivered. (The old names remain recognized, of course.)
2792 PH/44 There was no timeout on the connect() call when using a Unix domain
2793 socket in the ${readsocket expansion. There now is.
2795 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2796 be meaningful with "accept".
2798 SC/01 Eximstats V1.43
2799 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2801 SC/02 Eximstats V1.44
2802 Use a glob alias rather than an array ref in the generated
2803 parser. This improves both readability and performance.
2805 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2806 Collect SpamAssassin and rejection statistics.
2807 Don't display local sender or destination tables unless
2808 there is data to show.
2809 Added average volumes into the top table text output.
2811 SC/04 Eximstats V1.46
2812 Collect data on the number of addresses (recipients)
2813 as well as the number of messages.
2815 SC/05 Eximstats V1.47
2816 Added 'Message too big' to the list of mail rejection
2817 reasons (thanks to Marco Gaiarin).
2819 SC/06 Eximstats V1.48
2820 Mainlog lines which have GMT offsets and are too short to
2821 have a flag are now skipped.
2823 SC/07 Eximstats V1.49 (Alain Williams)
2824 Added the -emptyok flag.
2826 SC/08 Eximstats V1.50
2827 Fixes for obtaining the IP address from reject messages.
2829 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2830 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2831 whitespace changes from 4.64-PH/27
2833 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2836 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2837 are found, allow negative numbers in numeric criteria)
2839 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2841 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2842 to match changes made in 4.64-PH/43
2844 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2846 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2847 "domain" is actually an IP address, and force a failure. This locks out
2848 those revolvers/nameservers that support "A-for-A" lookups, in
2849 contravention of the specifications.
2851 PH/48 When a host name was looked up from an IP address, and the subsequent
2852 forward lookup of the name timed out, the host name was left in
2853 $sender_host_name, contrary to the specification.
2855 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2856 restricted to single-key lookups, Exim was not diagnosing an error if
2857 * or *@ was used with a query-style lookup.
2859 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2861 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2862 long ago, but noone interested enough thought of it. Let's just say that
2863 the "1.1" means that there are some new functions that weren't there at
2864 some point in the past.
2866 PH/51 Error processing for expansion failure of helo_data from an smtp
2867 transport during callout processing was broken.
2869 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2870 tested/used via the -bh/-bhc/-bs options.
2872 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2873 bug, fixed in subsequent PCRE releases).
2875 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2876 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2878 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2884 SC/01 Use a glob alias rather than an array ref in eximstats generated
2885 parser. This improves both readability and performance.
2887 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2888 Don't display local sender or destination tables in eximstats unless
2889 there is data to show.
2890 Added average volumes into the eximstats top table text output.
2892 SC/03 Collect data on the number of addresses (recipients) as well
2893 as the number of messages in eximstats.
2895 TF/01 Correct an error in the documentation for the redirect router. Exim
2896 does (usually) call initgroups() when daemonizing.
2898 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2899 with consistent privilege compared to when running as a daemon.
2901 TF/03 Note in the spec that $authenticated_id is not set for local
2902 submissions from trusted users.
2904 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2905 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2907 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2908 by adding some example configuration directives to the default
2909 configuration file. A little bit of work is required to uncomment the
2910 directives and define how usernames and passwords are checked, but
2911 there is now a framework to start from.
2913 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2914 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2915 without this. I don't know how relevant this is to other LDAP libraries.
2917 PH/02 Add the verb name to the "unknown ACL verb" error.
2919 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2921 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2923 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2924 directory not expanded when it should be if an expanded home directory
2925 was set for the address (which is overridden by the transport).
2927 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2930 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2931 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2932 because it is too late at that time, and has no effect.
2934 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2935 security issue with \' (bugzilla #107). I could not use the
2936 PQescapeStringConn() function, because it needs a PGconn value as one of
2939 PH/08 When testing addresses using -bt, indicate those final addresses that
2940 are duplicates that would not cause an additional delivery. At least one
2941 person was confused, thinking that -bt output corresponded to deliveries.
2942 (Suppressing duplicates isn't a good idea as you lose the information
2943 about possibly different redirections that led to the duplicates.)
2945 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2946 systems where poll() doesn't work, in particular OS X.
2948 PH/10 Added more information to debugging output for retry time not reached.
2950 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2951 operations in malware.c.
2953 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2956 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2957 syslog_duplication set false, log lines that would normally be written
2958 both the the main log and to the reject log were not written to syslog at
2961 PH/14 In the default configuration, change the use of "message" in ACL warn
2962 statements to "add_header".
2964 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2965 not followed by a command (e.g. "seen endif").
2967 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2968 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2971 PH/17 Added extra conditions to the default value of delay_warning_condition
2975 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2976 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2977 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2980 The Auto-Submitted: and various List- headers are standardised, whereas I
2981 don't think Precedence: ever was.
2983 PH/18 Refactored debugging code in route_finduser() to show more information,
2984 in particular, the error code if getpwnam() issues one.
2986 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2987 This is apparently needed in addition to the PH/07 change above to avoid
2988 any possible encoding problems.
2990 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2991 but not after initializing Perl.
2993 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2994 output them only if debugging. By default they are written stderr,
2995 apparently, which is not desirable.
2997 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3000 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3003 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3005 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3006 authenticated or an ident call has been made. Suppress the default
3007 values for $authenticated_id and $authenticated_sender (but permit -oMai
3008 and -oMas) when testing with -bh.
3010 PH/24 Re-jigged the order of the tests in the default configuration so that the
3011 tests for valid domains and recipients precede the DNS black list and CSA
3012 tests, on the grounds that those ones are more expensive.
3014 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3015 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3016 This bug exists in every version of Exim that I still have, right back to
3019 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3020 However, an attempt to turn on case-sensitivity in a regex key by
3021 including (?-i) didn't work because the subject string was already
3022 lowercased, and the effects were non-intuitive. It turns out that a
3023 one-line patch can be used to allow (?-i) to work as expected.
3029 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3030 other effects) broke the use of negated acl sub-conditions.
3032 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3035 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3036 "Deny" causes Exim to reject the incoming connection with a 554 error.
3037 Unfortunately, if there is a major crisis, such as a disk failure,
3038 tcp-wrappers gives "deny", whereas what one would like would be some
3039 kind of temporary error. A kludge has been added to help with this.
3040 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3041 554 error is used if errno is still zero or contains ENOENT (which occurs
3042 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3045 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3047 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3048 errors. Otherwise a message that provokes a temporary error (when other
3049 messages do not) can cause a whole host to time out.
3051 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3052 addresses were routed directly to files or pipes from a redirect router.
3053 File deliveries just didn't batch; pipe deliveries might have suffered
3056 PH/06 A failure to get a lock for a hints database would erroneously always say
3057 "Failed to get write lock", even when it was really a read lock.
3059 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3060 of 0600. This has been changed to use the value of the lockfile_mode
3061 option (which defaults to 0600).
3063 PH/08 Applied small patch from the Sieve maintainer.
3065 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3066 folder from quota calculations, a direct delivery into this folder messed
3067 up the contents of the maildirsize file. This was because the regex was
3068 used only to exclude .Trash (or whatever) when the size of the mailbox
3069 was calculated. There was no check that a delivery was happening into an
3070 excluded directory. This bug has been fixed by ignoring all quota
3071 processing for deliveries into excluded directories.
3073 PH/10 Added the maildirfolder_create_regex option to appendfile.
3079 PH/01 The code for finding all the local interface addresses on a FreeBSD
3080 system running IPv6 was broken. This may well have applied to all BSD
3081 systems, as well as to others that have similar system calls. The broken
3082 code found IPv4 interfaces correctly, but gave incorrect values for the
3083 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3084 that it would not match correctly against @[] and not recognize the IPv6
3087 PH/02 The ipliteral router was not recognizing addresses of the form user@
3088 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3090 PH/03 Added disable_ipv6.
3092 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3093 decoded form, because it is most often used to construct To: headers
3094 lines in autoreplies, and the decoded form may well be syntactically
3095 invalid. However, $reply_address has leading white space removed, and all
3096 newlines turned into spaces so that the autoreply transport does not
3099 PH/05 If group was specified without a user on a router, and no group or user
3100 was specified on a transport, the group from the router was ignored.
3102 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3103 for visible compile-time settings that can be used to change these
3104 numbers, for those that want even more. Backwards compatibility with old
3105 spool files has been maintained. However, going back to a previous Exim
3106 release will lost any variables that are in spool files.
3108 PH/07 Two small changes when running in the test harness: increase delay when
3109 passing a TCP/IP connection to a new process, in case the original
3110 process has to generate a bounce, and remove special handling of
3111 127.0.0.2 (sic), which is no longer necessary.
3113 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3114 be the same on different OS.
3116 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3119 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3120 whether --show-vars was specified or not
3122 JJ/02 exipick: Added support for new ACL variable spool format introduced
3125 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3126 syntactically invalid From: or Reply-to: line, and a filter used this to
3127 generate an autoreply, and therefore failed to obtain an address for the
3128 autoreply, Exim could try to deliver to a non-existent relative file
3129 name, causing unrelated and misleading errors. What now happens is that
3130 it logs this as a hard delivery error, but does not attempt to create a
3133 PH/11 The exinext utility has a -C option for testing purposes, but although
3134 the given file was scanned by exinext itself; it wasn't being passed on
3135 when Exim was called.
3137 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3138 an end-of-file indication when reading a command response.
3140 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3141 compiled. In many other places in Exim, IPv6 addresses are always
3142 recognized, so I have changed this. It also means that IPv4 domain
3143 literals of the form [IPV4:n.n.n.n] are now always recognized.
3145 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3146 used if the router is not running as root, for example, when verifying at
3147 ACL time, or when using -bh. The debugging output from this situation was
3148 non-existent - all you got was a failure to exec. I have made two
3151 (a) Failures to set uid/gid, the current directory, or a process leader
3152 in a subprocess such as that created by queryprogram now generate
3153 suitable debugging output when -d is set.
3155 (b) The queryprogram router detects when it is not running as root,
3156 outputs suitable debugging information if -d is set, and then runs
3157 the subprocess without attempting to change uid/gid.
3159 PH/15 Minor change to Makefile for building test_host (undocumented testing
3162 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3163 additional section of a DNS packet that returns MX or SRV records.
3164 Instead, it always explicitly searches for A/AAAA records. This avoids
3165 major problems that occur when a DNS server includes only records of one
3166 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3167 fixed another bug: if SRV records were looked up and the corresponding
3168 address records were *not* found in the additional section, the port
3169 values from the SRV records were lost.
3171 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3172 using the correct key (the original address) when searching the retry
3173 rules in order to find which one to use for generating the retry hint.
3175 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3176 adding the default one. Similarly, if it contains a Reply-To: header, the
3177 errors_reply_to option, if set, is not used.
3179 PH/19 When calculating a retry time, Exim used to measure the "time since
3180 failure" by looking at the "first failed" field in the retry record. Now
3181 it does not use this if it is later than than the arrival time of the
3182 message. Instead it uses the arrival time. This makes for better
3183 behaviour in cases where some deliveries succeed, thus re-setting the
3184 "first failed" field. An example is a quota failure for a huge message
3185 when small messages continue to be delivered. Without this change, the
3186 "time since failure" will always be short, possible causing more frequent
3187 delivery attempts for the huge message than are intended.
3188 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3190 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3191 $1, $2, $3) because the numerical variables can be reset during some
3192 expansion items (e.g. "match"), thereby losing the authentication data.
3194 PH/21 Make -bV show the size of off_t variables so that the test suite can
3195 decide whether to run tests for quotas > 2G.
3197 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3198 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3199 filecount value is greater than 2G or if a quota value is greater than 2G
3200 on a system where the size of off_t is not greater than 4, a panic error
3203 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3204 never match. The debug and -bh output now contains an explicit error
3205 message indicating a malformed IPv4 address or mask.
3207 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3208 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3209 PH/23 above applies.
3211 PH/25 Do not write to syslog when running in the test harness. The only
3212 occasion when this arises is a failure to open the main or panic logs
3213 (for which there is an explicit test).
3215 PH/26 Added the /no_tell option to "control=freeze".
3217 PH/27 If a host name lookup failed very early in a connection, for example, if
3218 the IP address matched host_lookup and the reverse lookup yielded a name
3219 that did not have a forward lookup, an error message of the form "no IP
3220 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3221 could be logged. Now it outputs the IP address instead of "NULL".
3223 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3224 allows the sender and the authenticated sender to be set when
3225 submitting a message from within Exim. Since child_open_exim() is
3226 documented for local_scan(), the new function should be too.
3228 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3229 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3230 results in an empty string is now treated as unset.
3232 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3234 PH/31 Added sender_verify_fail logging option.
3236 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3237 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3238 tidied the source and removed it altogether.
3240 PH/33 When a queue run was abandoned because the load average was too high, a
3241 log line was always written; now it is written only if the queue_run log
3242 selector is set. In addition, the log line for abandonment now contains
3243 information about the queue run such as the pid. This is always present
3244 in "start" and "stop" lines but was omitted from the "abandon" line.
3246 PH/34 Omit spaces between a header name and the colon in the error message that
3247 is given when verify = headers_syntax fails (if there are lots of them,
3248 the message gets confusing).
3250 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3251 names, as there are now some PTR records that contain slashes. This check
3252 is only to protect against broken name servers that fall over on strange
3253 characters, so the fact that it applies to all lookups doesn't matter.
3255 PH/36 Now that the new test suite is complete, we can remove some of the
3256 special code in Exim that was needed for the old test suite. For example,
3257 sorting DNS records because real resolvers return them in an arbitrary
3258 order. The new test suite's fake resolver always returns records in the
3261 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3262 bounces) except when queue_only is set, to avoid logging races between
3263 the different processes.
3265 PH/38 Panic-die if .include specifies a non-absolute path.
3267 PH/39 A tweak to the "H" retry rule from its user.
3269 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3270 a label. They prevented compilation on older perls.
3272 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3273 a warning to be raised on newish perls.
3275 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3276 on queue. Changes to match documented behaviour of showing count of
3277 messages matching specified criteria.
3279 PH/40 Changed the default ident timeout from 30s to 5s.
3281 PH/41 Added support for the use of login_cap features, on those BSD systems
3282 that have them, for controlling the resources used by pipe deliveries.
3284 PH/42 The content-scanning code uses fopen() to create files in which to put
3285 message data. Previously it was not paying any attention to the mode of
3286 the files. Exim runs with umask(0) because the rest of the code creates
3287 files with open(), and sets the required mode explicitly. Thus, these
3288 files were ending up world-writeable. This was not a big issue, because,
3289 being within the spool directory, they were not world-accessible. I have
3290 created a function called modefopen, which takes an additional mode
3291 argument. It sets umask(777), creates the file, chmods it to the required
3292 mode, then resets the umask. All the relevant calls to fopen() in the
3293 content scanning code have been changed to use this function.
3295 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3296 to 24 hours. This avoids potential overflow problems when processing G
3297 and H retry rules. I suspect nobody ever tinkers with this value.
3299 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3301 PH/45 When the plaintext authenticator is running as a client, the server's
3302 challenges are checked to ensure they are valid base64 strings. By
3303 default, the authentication attempt is cancelled if an invalid string is
3304 received. Setting client_ignore_invalid_base64 true ignores these errors.
3305 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3306 they are received. Thus, the responses can be made to depend on the
3307 challenges. If an invalid string is ignored, an empty string is placed in
3310 PH/46 Messages that are created by the autoreply transport now contains a
3311 References: header, in accordance with RFCs 2822 and 3834.
3313 PH/47 Added authenticated_sender_force to the smtp transport.
3315 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3317 PH/49 Installed latest patch from the Sieve maintainer.
3319 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3320 was also set, the appendfile transport was unnecessarily scanning a
3321 directory of message files (e.g. for maildir delivery) to find the count
3322 of files (along with the size), even though it did not need this
3323 information. It now does the scan only if it needs to find either the
3324 size of the count of files.
3326 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3328 PH/52 Two bugs concerned with error handling when the smtp transport is
3331 (i) Exim was not creating retry information for temporary errors given
3332 for individual recipients after the DATA command when the smtp transport
3333 was used in LMTP mode. This meant that they could be retried too
3334 frequently, and not timed out correctly.
3336 (ii) Exim was setting the flag that allows error details to be returned
3337 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3338 recipients that were returned after the DATA command.
3340 PH/53 This is related to PH/52, but is more general: for any failing address,
3341 when detailed error information was permitted to be returned to the
3342 sender, but the error was temporary, then after the final timeout, only
3343 "retry timeout exceeded" was returned. Now it returns the full error as
3344 well as "retry timeout exceeded".
3346 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3347 do this, and (what is worse) MTAs that accept it.
3349 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3350 will now be deprecated.
3352 PH/56 New os.c-cygwin from the Cygwin maintainer.
3354 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3355 formats (previously only available in exim formats via -bpr, -bpru,
3356 and -bpra. Now also available in native and exiqgrep formats)
3358 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3359 with very large, slow to parse queues
3361 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3363 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3365 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3366 responses to authentication challenges, though it was showing the
3367 challenges; (ii) I've removed the CR characters from the debug output for
3370 PH/58 Allow for the insertion of a newline as well as a space when a string
3371 is turned into more than one encoded-word during RFC 2047 encoding. The
3372 Sieve code now uses this.
3374 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3375 data_4xx, lost_connection, tls_required.
3377 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3378 message was being sent as an SMTP response.
3380 PH/61 Add -l and -k options to exicyclog.
3382 PH/62 When verifying, if an address was redirected to one new address, so that
3383 verification continued, and the new address failed or deferred after
3384 having set something in $address_data, the value of $address_data was not
3385 passed back to the ACL. This was different to the case when no
3386 redirection occurred. The value is now passed back in both cases.
3388 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3389 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3390 use login_cap.h, so on its own it isn't the distinguishing feature. The
3391 new name refers directly to the setclassresources() function.
3393 PH/65 Added configuration files for NetBSD3.
3395 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3397 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3398 is preferred over IPv4.
3400 PH/68 The bounce_return_message and bounce_return_body options were not being
3401 honoured for bounces generated during the reception of non-SMTP messages.
3402 In particular, this applied to messages rejected by the ACL. This bug has
3403 been fixed. However, if bounce_return_message is true and bounce_return_
3404 body is false, the headers that are returned for a non-SMTP message
3405 include only those that have been read before the error was detected.
3406 (In the case of an ACL rejection, they have all been read.)
3408 PH/69 The HTML version of the specification is now built in a directory called
3409 spec_html instead of spec.html, because the latter looks like a path with
3410 a MIME-type, and this confuses some software.
3412 PH/70 Catch two compiler warnings in sieve.c.
3414 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3415 function verify_get_ident() calls ip_connect() to connect a socket, but
3416 if the "connect()" function timed out, ip_connect() used to close the
3417 socket. However, verify_get_ident() also closes the socket later, and in
3418 between Exim writes to the log, which may get opened at this point. When
3419 the socket was closed in ip_connect(), the log could get the same file
3420 descriptor number as the socket. This naturally causes chaos. The fix is
3421 not to close the socket in ip_connect(); the socket should be closed by
3422 the function that creates it. There was only one place in the code where
3423 this was missing, in the iplookup router, which I don't think anybody now
3424 uses, but I've fixed it anyway.
3426 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3427 well as to direct DNS lookups. Otherwise the handling of names in host
3428 lists is inconsistent and therefore confusing.
3434 PH/01 Two changes to the default runtime configuration:
3436 (1) Move the checks for relay_from_hosts and authenticated clients from
3437 after to before the (commented out) DNS black list checks.
3439 (2) Add control=submission to the relay_from_hosts and authenticated
3440 clients checks, on the grounds that messages accepted by these
3441 statements are most likely to be submissions.
3443 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3445 (1) Generate an error if the third argument for the ${prvs expansion is
3448 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3451 (3) Reset the variables that are obtained from the first argument of
3452 ${prvscheck and used in the second argument before leaving the code,
3453 because their memory is reclaimed, so using them afterwards may do
3456 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3457 one (it's much easier than Tom thought :-).
3459 (5) Because of (4), we can now allow for the use of $prvscheck_result
3460 inside the third argument.
3462 PH/03 For some reason, the default setting of PATH when running a command from
3463 a pipe transport was just "/usr/bin". I have changed it to
3466 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3467 anything to be listed in the output from -bV.
3469 PH/05 When a filter generated an autoreply, the entire To: header line was
3470 quoted in the delivery log line, like this:
3472 => >A.N.Other <ano@some.domain> <original@ddress> ...
3474 This has been changed so that it extracts the operative address. There
3475 may be more than one such address. If so, they are comma-separated, like
3478 => >ano@some.domain,ona@other.domain <original@ddress> ...
3480 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3481 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3482 not being looked up in the rDNS to get a host name, Exim was showing the
3483 IP address twice in Received: lines, even though the IP addresses were
3484 identical. For example:
3486 Received: from [1.2.3.4] (helo=[1.2.3.4])
3488 However, if the real host name was known, it was omitting the HELO data
3489 if it matched the actual IP address. This has been tidied up so that it
3490 doesn't show the same IP address twice.
3492 PH/07 When both +timestamp and +memory debugging was on, the value given by
3493 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3494 called by the debug printing, thereby overwriting the timestamp buffer.
3495 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3497 PH/08 When the original message was included in an autoreply transport, it
3498 always said "this is a copy of the message, including all the headers",
3499 even if body_only or headers_only was set. It now gives an appropriate
3502 PH/09 Applied a patch from the Sieve maintainer which:
3504 o fixes some comments
3505 o adds the (disabled) notify extension core
3506 o adds some debug output for the result of if/elsif tests
3507 o points to the current vacation draft in the documentation
3508 and documents the missing references header update
3512 o fixes a bug in processing the envelope test (when testing
3513 multiple envelope elements, the last element determined the
3516 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3517 Electronic Mail") by including:
3519 Auto-submitted: auto-generated
3521 in the messages that it generates (bounce messages and others, such as
3522 warnings). In the case of bounce messages for non-SMTP messages, there was
3523 also a typo: it was using "Auto_submitted" (underscore instead of
3524 hyphen). Since every message generated by Exim is necessarily in response
3525 to another message, thes have all been changed to:
3527 Auto-Submitted: auto-replied
3529 in accordance with these statements in the RFC:
3531 The auto-replied keyword:
3533 - SHOULD be used on messages sent in direct response to another
3534 message by an automatic process,
3536 - MUST NOT be used on manually-generated messages,
3538 - MAY be used on Delivery Status Notifications (DSNs) and Message
3539 Disposition Notifications (MDNs),
3541 - MUST NOT be used on messages generated by automatic or periodic
3542 processes, except for messages which are automatic responses to
3545 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3546 to the default Received: header definition.
3548 PH/12 Added log selector acl_warn_skipped (default on).
3550 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3551 variables because (a) they are in the wrong storage pool and (b) even if
3552 they were copied, it wouldn't work properly because of the caching.
3554 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3555 checking when decoding. Apparently there are clients that generate
3556 overlong encoded strings. Why am I not surprised?
3558 PH/15 If the first argument of "${if match_address" was not empty, but did not
3559 contain an "@" character, Exim crashed. Now it writes a panic log message
3560 and treats the condition as false.
3562 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3564 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3565 extension "envelope-auth". The code is finished and in agreement with
3566 other implementations, but there is no documentation so far and in fact,
3567 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3568 not changing the active code.
3570 Print executed "if" and "elsif" statements when debugging is used. This
3571 helps a great deal to understand what a filter does.
3573 Document more things not specified clearly in RFC3028. I had all this
3574 sorted out, when out of a sudden new issues came to my mind. Oops."
3576 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3579 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3580 canonical form (as documented). However, after a host name lookup from
3581 the IP address, check_host() was doing a simple string comparison with
3582 addresses acquired from the DNS when checking that the found name did
3583 have the original IP as one of its addresses. Since any found IPv6
3584 addresses are likely to be in abbreviated form, the comparison could
3585 fail. Luckily, there already exists a function for doing the comparison
3586 by converting both addresses to binary, so now that is used instead of
3587 the text comparison.
3589 PH/20 There was another similar case to PH/19, when a complete host name was
3590 given in a host list; looking up its IP address could give an abbreviated
3591 form, whereas the current host's name might or might not be abbreviated.
3592 The same fix has been applied.
3598 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3599 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3602 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3603 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3605 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3607 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3608 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3609 Also, for "Auto-Submitted", treat anything other than "no" as
3610 non-personal, in accordance with RFC 3834. (Previously it treated
3611 anything starting "auto-" as non-personal.)
3613 TF/01 The control=submission/name=... option had a problem with syntax
3614 errors if the name included a slash character. The /name= option
3615 now slurps the rest of the string, so it can include any characters
3616 but it must come last in the list of options (after /sender_retain
3619 PH/05 Some modifications to the interface to the fake nameserver for the new
3627 TK/01 Added the "success_on_redirect" address verification option. See
3628 NewStuff for rationale and an example.
3630 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3632 PH/02 Patch to exigrep to allow it to work on syslog lines.
3634 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3635 fread() to skip over the body file's header line, because in Cygwin the
3636 header line is locked and is inaccessible.
3638 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3639 co-exist for some time) to make it clear that it is the Exim ID that is
3640 referenced, not the Message-ID: header line.
3642 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3643 string_format() function, because snprintf() does not exist on all
3646 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3647 ${stat: expansion item.
3649 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3650 protocol synchronization error", to keep the pedants happy.
3652 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3653 well as for IRIX systems, when gcc is being used. See the host.c source
3656 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3658 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3661 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3662 transport and to the smtp transport in LMTP mode.
3664 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3666 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3667 run a filter in a subprocess. This could lead to confusion in subsequent
3668 lookups in the parent process. There should also be a search_tidyup() at
3669 the end of the subprocess.
3671 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3672 only if the host matched helo_try_verify_hosts, which caused the
3673 verification to occur when the EHLO/HELO command was issued. The ACL just
3674 tested the remembered result. Now, if a previous verification attempt has
3675 not happened, "verify = helo" does it there and then.
3677 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3679 TK/03 Fix log output including CR from clamd.
3681 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3682 exist provoked a memory error which could cause a segfault.
3684 PH/15 Installed PCRE 6.2
3686 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3688 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3689 of the problem. Specifically, suggested +O2 rather than +O1 for the
3692 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3694 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3695 record, the port was not correctly logged when the outgoing_port log
3696 selector was set (it logged the transort's default port).
3698 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3699 fallback_hosts, and "hosts" in the smtp transport.
3701 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3702 host errors such as "Connection refused".
3704 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3705 authentication with radiusclient 0.4.9:
3707 - Error returned from rc_read_config was caught wrongly
3708 - Username/password not passed on to radius server due to wrong length.
3710 The presumption is that some radiusclient API changes for 4.51/PH/17
3711 were not taken care of correctly. The code is still untested by me (my
3712 Linux distribution still has 0.3.2 of radiusclient), but it was
3713 contributed by a Radius user.
3715 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3716 expanding the "port" option of the smtp transport.
3718 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3719 while reading a MIME header. Thanks to Tom Hughes for a patch.
3721 PH/24 Include config.h inside local_scan.h so that configuration settings are
3724 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3725 that in an ACL for RCPT (for example), you can examine exactly what was
3728 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3729 commands, but it was not correctly comparing the address with the actual
3730 client host address. Thus, it would show the EHLO address in Received:
3731 header lines when this was not necessary.
3733 PH/27 Added the % operator to ${eval:}.
3735 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3736 it should be ignoring failures (because with -C, for example, it has lost
3737 privilege). It wasn't ignoring creation failures other than "already
3740 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3743 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3744 because an input line was too long, either on its own, or by virtue of
3745 too many continuations, the temporary file was not being removed, and the
3746 return code was incorrect.
3748 PH/31 Missing "BOOL" in function definition in filtertest.c.
3750 PH/32 Applied Sieve patches from the maintainer.
3752 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3754 PH/33 Added "verify = not_blind".
3756 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3757 Local/Makefile (with some defaults set). These are used in built scripts
3758 such as exicyclog, but they have never been used in the exim_install
3759 script (though there are many overriding facilities there). I have
3760 arranged that the exim_install script now takes note of these two
3763 PH/35 Installed configuration files for Dragonfly.
3765 PH/36 When a locally submitted message by a trusted user did not contain a
3766 From: header, and the sender address was obtained from -f or from an SMTP
3767 MAIL command, and the trusted user did not use -F to supply a sender
3768 name, $originator_name was incorrectly used when constructing a From:
3769 header. Furthermore, $originator_name was used for submission mode
3770 messages from external hosts without From: headers in a similar way,
3771 which is clearly wrong.
3773 PH/37 Added control=suppress_local_fixups.
3775 PH/38 When log_selector = +received_sender was set, and the addition of the
3776 sender made the log line's construction buffer exactly full, or one byte
3777 less than full, an overflow happened when the terminating "\n" was
3780 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3781 when the result of a list match is failure because a DNS lookup failed.
3783 PH/40 RM_COMMAND is now used in the building process.
3785 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3786 the "build-* directories that it finds.
3788 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3789 address in a domain literal was a prefix of an interface address.
3791 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3792 when verifying a sender address, unless rewrite_headers is false.
3794 PH/44 Wrote a long comment about why errors_to addresses are verified as
3795 recipients, not senders.
3797 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3798 the ratelimit ACL was added.
3800 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3802 PH/46 Added extra information about PostgreSQL errors to the error string.
3804 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3805 avoiding the need to install special zones in a real server. This is
3806 backwards compatible; if it can't find the fake resolver, it drops back.
3807 Thus, both old and new test suites can be run.
3809 TF/02 Added util/ratelimit.pl
3811 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3812 clock is set back in time.
3814 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3815 Candler <B.Candler@pobox.com>.
3817 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3818 OK for addresses that are the result of redirections.
3820 PH/48 A number of further additions for the benefit of the new test suite,
3821 including a fake gethostbyname() that interfaces to the fake DNS resolver
3824 TF/06 The fix for widen_domains has also been applied to qualify_single and
3825 search_parents which are the other dnslookup options that can cause
3828 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3831 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3833 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3834 service" if it encountered an MX record with an empty target hostname.
3835 The message is now "an MX or SRV record indicated no SMTP service".
3837 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3838 if the DNS of the sending site is misconfigured. This is quite a
3839 common situation. This change restores the behaviour of treating a
3840 helo verification defer as a failure.
3842 PH/51 If self=fail was set on a router, the bounce message did not include the
3843 actual error message.
3849 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3851 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3852 command itself ended in error, the underlying message about the transport
3853 filter timeout was being overwritten with the pipe command error. Now the
3854 underlying error message should be appended to the second error message.
3856 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3858 PH/02 Reduce the amount of output that "make" produces by default. Full output
3859 can still be requested.
3861 PH/03 The warning log line about a condition test deferring for a "warn" verb
3862 was being output only once per connection, rather than after each
3863 occurrence (because it was using the same function as for successful
3864 "warn" verbs). This seems wrong, so I have changed it.
3866 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3867 it should not have, which might have caused a crash in the right
3868 circumstances, but probably never did.
3870 PH/04 Installed a modified version of Tony Finch's patch to make submission
3871 mode fix the return path as well as the Sender: header line, and to
3872 add a /name= option so that you can make the user's friendly name appear
3875 TF/03 Added the control = fakedefer ACL modifier.
3877 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3878 Mark Lowes for thorough testing.
3880 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3882 TK/03 Merged latest SRS patch from Miles Wilton.
3884 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3885 with the definition in sysexits.h (which is #included earlier).
3886 Fortunately, Exim does not actually use EX_OK. The code used to try to
3887 preserve the sysexits.h value, by assuming that macro definitions were
3888 scanned for macro replacements. I have been disabused of this notion,
3889 so now the code just undefines EX_OK before #including unistd.h.
3891 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3892 in the smtp transport. When a block could not be written in a single
3893 write() function, the timeout was being re-applied to each part-write.
3894 This seems wrong - if the receiver was accepting one byte at a time it
3895 would take for ever. The timeout is now adjusted when this happens. It
3896 doesn't have to be particularly precise.
3898 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3899 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3901 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3902 without a domain if the check to <postmaster@domain> fails.
3904 SC/01 Eximstats: added -xls and the ability to specify output files
3905 (patch written by Frank Heydlauf).
3907 SC/02 Eximstats: use FileHandles for outputting results.
3909 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3911 SC/04 Eximstats: fixed display of large numbers with -nvr option
3913 SC/05 Eximstats: fixed merging of reports with empty tables.
3915 SC/06 Eximstats: added the -include_original_destination flag
3917 SC/07 Eximstats: removed tabs and trailing whitespace.
3919 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3921 TK/06 MBOX spool code: Add real "From " MBOX separator line
3922 so the .eml file is really in mbox format (even though
3923 most programs do not really care). Patch from Alex Miller.
3925 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3926 The latter is generated from $received_to and is only set if the
3927 message has one envelope recipient. SA can use these headers,
3928 obviously out-of-the-box. Patch from Alex Miller.
3930 PH/08 The ${def test on a variable was returning false if the variable's
3931 value was "0", contrary to what the specification has always said!
3932 The result should be true unless the variable is empty.
3934 PH/09 The syntax error of a character other than { following "${if
3935 def:variable_name" (after optional whitespace) was not being diagnosed.
3936 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3937 accidental colon was present, for example, could give incorrect results.
3939 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3940 result is used (not including appendfile, where other changes are about
3943 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3944 This involved changing a lot of size variables from int to off_t. It
3945 should work with maildirs and everything.
3947 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3948 spamd dying while we are connected to it.
3950 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3953 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3954 function for BDB 4.3.
3956 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3958 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3959 the sources for PCRE have been re-arranged and more files are now
3962 PH/15 The code I had for printing potentially long long variables in PH/11
3963 above was not the best (it lost precision). The length of off_t variables
3964 is now inspected at build time, and an appropriate printing format (%ld
3965 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3966 to be "long long int" or "long int". This is needed for the internal
3967 formatting function string_vformat().
3969 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3970 the configuration file to be ":syslog", then the script "guesses" where
3971 the logs files are, rather than using the compiled in default. In our
3972 case the guess is not the same as the compiled default, so the script
3973 suddenly stopped working when I started to use syslog. The patch checks
3974 to see if log_file_path is "". If so, it attempts to read it from exim
3975 with no configuration file to get the compiled in version, before it
3976 falls back to the previous guessing code."
3978 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3979 implementing BATV in an Exim configuration. See NewStuff for the gory
3982 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3983 Makefile that are specific to HP-UX.
3985 PH/18 If the "use_postmaster" option was set for a recipient callout together
3986 with the "random" option, the postmaster address was used as the MAIL
3987 FROM address for the random test, but not for the subsequent recipient
3988 test. It is now used for both.
3990 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3991 patch removes a few documentation additions to RFC 3028, because the
3992 latest draft now contains them. It adds the new en;ascii-case comparator
3993 and a new error check for 8bit text in MIME parts. Comparator and
3994 require names are now matched exactly. I enabled the subaddress
3995 extension, but it is not well tested yet (read: it works for me)."
3997 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3998 rework some of the code of TK/09 above to avoid the hardwired use of
3999 "%lld" and "long long". Replaced the call to snprintf() with a call to
4002 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4003 records point to non-existent hosts", "retry timeout exceeded", and
4004 "retry time not reached for any host after a long failure period".
4006 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4007 experimental DomainKeys support:
4009 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4010 (2) On an error such as an illegally used "control", the wrong name for
4011 the control was given.
4013 These problems did NOT occur unless DomainKeys support was compiled.
4015 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4017 PH/24 Added ${if match_ip condition.
4019 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4020 databases so that it will be absolutely obvious if a crash occurs in the
4021 DB library. This is a regular occurrence (often caused by mis-matched
4024 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4025 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4026 picked up on a user's system that detects such things. There doesn't seem
4027 to be a gcc warning option for this - only an attribute that has to be
4028 put on the function's prototype. It seems that in Fedora Core 4 they have
4029 set this on a number of new functions. No doubt there will be more in due
4032 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4033 specify a transport. However, if an address that was verified by such a
4034 router was the subject of a callout, Exim crashed because it tried to
4035 read the rcpt_include_affixes from the non-existent transport. Now it
4036 just assumes that the setting of that option is false. This bug was
4037 introduced by 4.51/PH/31.
4039 PH/28 Changed -d+all to exclude +memory, because that information is very
4040 rarely of interest, but it makes the output a lot bigger. People tend to
4041 do -d+all out of habit.
4043 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4044 code in os-type was giving problems when libc.so lives in lib64, like on
4047 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4048 aren't the modern standard, and it seems that some systems' include files
4049 don't always have them. Exim was already checking for some of the newer
4050 ones like T_AAAA, and defining it itself. I've added checks for all the
4051 record types that Exim uses.
4053 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4054 not automatically generating a new one, as it is supposed to. This
4055 prevented TLS from working. If the file did exist, but contained invalid
4056 data, a new version was generated, as expected. It was only the case of a
4057 non-existent file that was broken.
4059 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4060 with a change in libdomainkeys > 0.64.
4062 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4063 from DNS. If the selector record carries the flag, it now has
4064 precedence over the domain-wide flag.
4066 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4068 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4069 the use of a port name that isn't defined in /etc/services) occurred, the
4070 message was deferred as in a normal delivery, and thus remained on the
4071 spool, instead of being failed because of the mua_wrapper setting. This
4072 is now fixed, and I tidied up some of the mua_wrapper messages at the
4075 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4076 the messages in a hash of arrays rather than using individual hashes.
4077 This is a bit cleaner and results in dramatic memory savings, albeit
4078 at a slight CPU cost.
4080 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4081 as requested by Marc Sherman.
4083 SC/10 Eximstats: added histograms for user specified patterns as requested
4086 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4088 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4089 fopen() in the content-scanning modules that did not already have it.
4095 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4096 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4098 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4100 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4102 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4103 to test Sieve filters that use "vacation".
4105 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4106 that changes the way the GnuTLS parameters are stored in the cache file.
4107 The new format can be generated externally. For backward compatibility,
4108 if the data in the cache doesn't make sense, Exim assumes it has read an
4109 old-format file, and it generates new data and writes a new file. This
4110 means that you can't go back to an older release without removing the
4113 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4114 work if there are any delivery delays because "one_time" forces the
4115 parent to be marked "delivered", so its unseen clone is never tried
4116 again. For this reason, Exim now forbids the simultaneous setting of
4119 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4120 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4121 entry if you want to know the details. The fix, however, neglected to
4122 consider the case when local delivery batching is involved. The test for
4123 "previously delivered" was not happening when checking to see if an
4124 address could be batched with a previous (undelivered) one; under
4125 certain circumstances this could lead to multiple deliveries to the same
4128 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4129 in its include files, and this causes problems building Exim.
4131 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4132 header_syntax) but Exim was just ignoring anything given after a slash.
4133 In particular, this caused confusion with an attempt to use "verify =
4134 reverse_host_lookup/defer_ok". An error is now given when options are
4135 supplied for verify items that do not have them. (Maybe reverse_host_
4136 lookup should have a defer_ok option, but that's a different point.)
4138 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4139 defined by RFC 821) to 2048, because there were problems with some AUTH
4140 commands, and RFC 1869 says the size should be increased for extended
4141 SMTP commands that take arguments.
4143 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4146 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4147 "unknown" error; now it says that the functionality isn't in the binary.
4149 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4150 an address' error message when a string expansion fails (syntax or
4151 whatever). Otherwise the password may appear in the log. Following change
4152 PH/42 below, there is no longer a chance of it appearing in a bounce
4155 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4157 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4158 of its addresses was ever tried. (Bugzilla bug #2.)
4160 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4161 the result incorrectly in the debug output. (It correctly added a newline
4162 to what was transported.)
4164 TF/01 Added $received_time.
4166 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4167 commented out examples of how to interface to a virus scanner and to
4168 SpamAssassin. Also added commented examples of av_scanner and
4169 spamd_address settings.
4171 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4172 and controls are allowed in which ACLs. There were a couple of minor
4173 errors. Some of the entries in the conditions table (which is a table of
4174 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4175 negation of where the condition IS allowed.
4177 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4179 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4180 header file does not have a version number, so I've had to invent a new
4181 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4182 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4183 radiusclient), but it was contributed by a Radius user.
4185 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4186 files or directories, for OpenSSL.
4188 PH/19 When an Exim process that is running as root has to create an Exim log
4189 file, it does so in a subprocess that runs as exim:exim so as to get the
4190 ownership right at creation (otherwise, other Exim processes might see
4191 the file with the wrong ownership). There was no test for failure of this
4192 fork() call, which would lead to the process getting stuck as it waited
4193 for a non-existent subprocess. Forks do occasionally fail when resources
4194 run out. I reviewed all the other calls to fork(); they all seem to check
4197 PH/20 When checking for unexpected SMTP input at connect time (before writing
4198 the banner), Exim was not dealing correctly with a non-positive return
4199 from the read() function. If the client had disconnected by this time,
4200 the result was a log entry for a synchronization error with an empty
4201 string after "input=" when read() returned zero. If read() returned -1
4202 (an event I could not check), uninitialized data bytes were printed.
4203 There were reports of junk text (parts of files, etc) appearing after
4206 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4208 PH/22 Added support for macro redefinition, and (re)definition in between
4209 driver and ACL definitions.
4211 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4212 forgetting to use the resulting value; it was using the unexpanded value.
4214 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4215 hadn't been configured. The fix is from Juergen Kreileder, who
4216 understands it better than I do:
4218 "Here's what I see happening with three configured cyrus_sasl
4219 authenticators configured (plain, login, cram-md5):
4221 On startup auth_cyrus_sasl_init() gets called for each of these.
4222 This means three calls to sasl_listmech() without a specified mech_list.
4223 => SASL tests which mechs of all available mechs actually work
4224 => three warnings about OTP not working
4225 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4227 With the patch, sasl_listmech() also gets called three times. But now
4228 SASL's mech_list option is set to the server_mech specified in the the
4229 authenticator. Or in other words, the answer from sasl_listmech()
4230 gets limited to just the mech you're testing for (which is different
4232 => the return list contains just 'plain' or 'login', 'cram-md5' or
4233 nothing depending on the value of ob->server_mech.
4235 I've just tested the patch: Authentication still works fine,
4236 unavailable mechs specified in the exim configuration are still
4237 caught, and the auth.log warnings about OTP are gone."
4239 PH/25 When debugging is enabled, the contents of the command line are added
4240 to the debugging output, even when log_selector=+arguments is not
4243 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4244 answer is "GNU", and only if the return is "GNU/something" is the answer
4247 PH/27 $acl_verify_message is now set immediately after the failure of a
4248 verification in an ACL, and so is available in subsequent modifiers. In
4249 particular, the message can be preserved by coding like this:
4251 warn !verify = sender
4252 set acl_m0 = $acl_verify_message
4254 Previously, $acl_verify_message was set only while expanding "message"
4255 and "log_message" when a very denied access.
4257 PH/28 Modified OS/os.c-Linux with
4259 -#ifndef OS_LOAD_AVERAGE
4260 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4262 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4263 nomenclature these days.)
4265 PH/29 Installed patch from the Sieve maintainer that adds the options
4266 sieve_useraddress and sieve_subaddress to the redirect router.
4268 PH/30 In these circumstances:
4269 . Two addresses routed to the same list of hosts;
4270 . First host does not offer TLS;
4271 . First host accepts first address;
4272 . First host gives temporary error to second address;
4273 . Second host offers TLS and a TLS session is established;
4274 . Second host accepts second address.
4275 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4276 and peerdn, if requested) that were in fact used only for the second
4279 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4280 attention to any local part prefix or suffix that was matched by the
4281 router that accepted the address. It now behaves in the same way as it
4282 does for delivery: the affixes are removed from the local part unless
4283 rcpt_include_affixes is set on the transport.
4285 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4286 timeout during the DATA phase of an incoming message.
4288 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4289 applied a patch sanctioned by the Sieve maintainer.
4291 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4292 the uid or gid is negative. A case of a negative gid caused this to be
4293 noticed. The fix allows for either to be negative.
4295 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4296 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4299 PH/36 The change PH/12 above was broken. Fixed it.
4301 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4302 the grounds that routing the same address twice would always produce the
4303 same answer. This might have been true once, but it is certainly no
4304 longer true now. Routing a child address may depend on the previous
4305 routing that produced that child. Some complicated redirection strategies
4306 went wrong when messages had multiple recipients, and made Exim's
4307 behaviour dependent on the order in which the addresses were given.
4309 I have moved the duplicate checking until after the routing is complete.
4310 Exim scans the addresses that are assigned to local and remote
4311 transports, and removes any duplicates. This means that more work will be
4312 done, as duplicates will always all be routed, but duplicates are
4313 presumably rare, so I don't expect this is of any significance.
4315 For deliveries to pipes, files, and autoreplies, the duplicate checking
4316 still happens during the routing process, since they are not going to be
4319 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4320 It corrects a timeout issue with spamd. This is Ian's comment: "The
4321 background is that sometimes spamd either never reads data from a
4322 connection it has accepted, or it never writes response data. The exiscan
4323 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4324 blindly assumes that writes won't block so it may never time out."
4326 PH/39 Allow G after quota size as well as K and M.
4328 PH/40 The value set for $authenticated_id in an authenticator may not contain
4329 binary zeroes or newlines because the value is written to log lines and
4330 to spool files. There was no check on this. Now the value is run through
4331 the string_printing() function so that such characters are converted to
4332 printable escape sequences.
4334 PH/41 $message_linecount is a new variable that contains the total number of
4335 lines in the message. Compare $body_linecount, which is the count for the
4338 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4339 bounce and delay warning messages, except in certain special cases, which
4342 (a) An SMTP error message from a remote host;
4343 (b) A message specified in a :fail: redirection;
4344 (c) A message specified in a "fail" command in a system filter;
4345 (d) A message specified in a FAIL return from the queryprogram router;
4346 (e) A message specified by the cannot_route_message router option.
4348 In these cases only, Exim does include the error details in bounce and
4349 warning messages. There are also a few cases where bland messages such
4350 as "unrouteable address" or "local delivery error" are given.
4352 PH/43 $value is now also set for the "else" part of a ${run expansion.
4354 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4355 being worked on, but at least Exim now implements the latest version to
4358 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4359 process to complete was treated as a delivery failure, a timeout while
4360 writing the message to the pipe was logged, but erroneously treated as a
4361 successful delivery. Such timeouts include transport filter timeouts. For
4362 consistency with the overall process timeout, these timeouts are now
4363 treated as errors, giving rise to delivery failures by default. However,
4364 there is now a new Boolean option for the pipe transport called
4365 timeout_defer, which, if set TRUE, converts the failures into defers for
4366 both kinds of timeout. A transport filter timeout is now identified in
4369 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4370 systems where "make" and "gmake" are different, calling "gmake" at top
4371 level broke things. I've arranged for the value of $(MAKE) to be passed
4372 from the Makefile to this script so that it can call the same version of
4376 A note about Exim versions 4.44 and 4.50
4377 ----------------------------------------
4379 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4380 changes of various kinds. As a consequence, a big documentation update was
4381 needed. This delayed the release for rather longer than seemed good, especially
4382 in the light of a couple of (minor) security issues. Therefore, the changes
4383 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4384 release. So 4.44 and 4.50 are in effect two different branches that both start
4387 I have left the 4.50 change log unchanged; it contains all the changes since
4388 4.43. The change log for 4.44 is below; many of its items are identical to
4389 those for 4.50. This seems to be the most sensible way to preserve the
4390 historical information.
4396 1. Minor wording change to the doc/README.SIEVE file.
4398 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4399 computation of the current number of files was incorrect.
4401 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4402 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4405 4. Give more explanation in the error message when the command for a transport
4406 filter fails to execute.
4408 5. There are several places where Exim runs a non-Exim command in a
4409 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4410 was being done only for the command run by the queryprogram router. It is
4411 now done for all such subprocesses. The other cases are: ${run, transport
4412 filters, and the commands run by the lmtp and pipe transports.
4414 6. Added CONFIGURE_GROUP build-time option.
4416 7. Some older OS have a limit of 256 on the maximum number of file
4417 descriptors. Exim was using setrlimit() to set 1000 as a large value
4418 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4419 systems. I've change it so that if it can't get 1000, it tries for 256.
4421 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4422 was an oversight, and furthermore, ever since the addition of extra
4423 controls (e.g. 4.43/32), the checks on when to allow different forms of
4424 "control" were broken. There should now be diagnostics for all cases when a
4425 control that does not make sense is encountered.
4427 9. Added the /retain_sender option to "control=submission".
4429 10. $recipients is now available in the predata ACL (oversight).
4431 11. Tidy the search cache before the fork to do a delivery from a message
4432 received from the command line. Otherwise the child will trigger a lookup
4433 failure and thereby defer the delivery if it tries to use (for example) a
4434 cached ldap connection that the parent has called unbind on.
4436 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4437 of $address_data from the recipient verification was clobbered by the
4438 sender verification.
4440 13. The value of address_data from a sender verification is now available in
4441 $sender_address_data in subsequent conditions in the ACL statement.
4443 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4445 15. Added a new option "connect=<time>" to callout options, to set a different
4448 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4449 was its contents. (It was OK if the option was not defined at all.)
4451 17. A "Completed" log line is now written for messages that are removed from
4452 the spool by the -Mrm option.
4454 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4455 information about exactly what failed.
4457 19. Added -dd to debug only the daemon process.
4459 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4460 handles timeouts, both on the server side and network timeouts. Renamed the
4461 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4463 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4464 It is now set to "smtps".
4466 22. $host_address is now set to the target address during the checking of
4467 ignore_target_hosts.
4469 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4470 being passed; this would have caused $sender_host_name to have been used if
4471 matching the list had actually called for a host name (not very likely,
4472 since this list is usually IP addresses). A host name is now passed as
4475 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4476 code that specifies a non-restarting handler (typically sigaction() in
4477 modern systems) in an attempt to fix a rare and obscure crash bug.
4479 25. Narrowed the window for a race in the daemon that could cause it to ignore
4480 SIGCHLD signals. This is not a major problem, because they are used only to
4481 wake it up if nothing else does.
4483 26. A malformed maildirsize file could cause Exim to calculate negative values
4484 for the mailbox size or file count. Odd effects could occur as a result.
4485 The maildirsize information is now recalculated if the size or filecount
4488 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4489 support for a long time. Removed HAVE_SYS_VFS_H.
4491 28. Installed the latest version of exipick from John Jetmore.
4493 29. In an address list, if the pattern was not a regular expression, an empty
4494 subject address (from a bounce message) matched only if the pattern was an
4495 empty string. Non-empty patterns were not even tested. This was the wrong
4496 because it is perfectly reasonable to use an empty address as part of a
4497 database query. An empty address is now tested by patterns that are
4498 lookups. However, all the other forms of pattern expect the subject to
4499 contain a local part and a domain, and therefore, for them, an empty
4500 address still always fails if the pattern is not itself empty.
4502 30. Exim went into a mad DNS loop when attempting to do a callout where the
4503 host was specified on an smtp transport, and looking it up yielded more
4504 than one IP address.
4506 31. Re-factored the code for checking spool and log partition space into a
4507 function that finds that data and another that does the check. The former
4508 is then used to implement four new variables: $spool_space, $log_space,
4509 $spool_inodes, and $log_inodes.
4511 32. The RFC2047 encoding function was originally intended for short strings
4512 such as real names; it was not keeping to the 75-character limit for
4513 encoded words that the RFC imposes. It now respects the limit, and
4514 generates multiple encoded words if necessary. To be on the safe side, I
4515 have increased the buffer size for the ${rfc2047: expansion operator from
4518 33. It is now permitted to omit both strings after an "if" condition; if the
4519 condition is true, the result is "true". As before, when the second string
4520 is omitted, a false condition yields an empty string. This makes it less
4521 cumbersome to write custom ACL and router conditions.
4523 34. Failure to deliver a bounce message always caused it to be frozen, even if
4524 there was an errors_to setting on the router. The errors_to setting is now
4527 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4528 canonical form (fully expanded) before being placed in
4529 $sender_host_address.
4531 36. The table in the code that translates DNS record types into text (T_A to
4532 "A" for instance) was missing entries for NS and CNAME. It is just possible
4533 that this could have caused confusion if both these types were looked up
4534 for the same domain, because the text type is used as part of Exim's
4535 per-process caching. But the chance of anyone hitting this buglet seems
4538 37. The dnsdb lookup has been extended in a number of ways.
4540 (1) There is a new type, "zns", which walks up the domain tree until it
4541 finds some nameserver records. It should be used with care.
4543 (2) There is a new type, "mxh", which is like "mx" except that it returns
4544 just the host names, not the priorities.
4546 (3) It is now possible to give a list of domains (or IP addresses) to be
4547 looked up. The behaviour when one of the lookups defers can be
4548 controlled by a keyword.
4550 (4) It is now possible to specify the separator character for use when
4551 multiple records are returned.
4553 38. The dnslists ACL condition has been extended: it is now possible to supply
4554 a list of IP addresses and/or domains to be looked up in a particular DNS
4557 39. Added log_selector=+queue_time_overall.
4559 40. When running the queue in the test harness, wait just a tad after forking a
4560 delivery process, to get repeatability of debugging output.
4562 41. Include certificate and key file names in error message when GnuTLS fails
4563 to set them up, because the GnuTLS error message doesn't include the name
4564 of the failing file when there is a problem reading it.
4566 42. Allow both -bf and -bF in the same test run.
4568 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4570 44. The "Exiscan patch" is now merged into the mainline Exim source.
4572 45. Sometimes the final signoff response after QUIT could fail to get
4573 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4574 < 0 before doing a fflush(). This bug looks as though it goes back to the
4575 introduction of TLS in release 3.20, but "sometimes" must have been rare
4576 because the tests only now provoked it.
4578 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4579 (this can affect the format of dates).
4581 47. exim_tidydb, when checking for the continued existence of a message for
4582 which it has found a message-specific retry record, was not finding
4583 messages that were in split spool directories. Consequently, it was
4584 deleting retry records that should have stayed in existence.
4586 48. Steve fixed some bugs in eximstats.
4588 49. The SPA authentication driver was not abandoning authentication and moving
4589 on to the next authenticator when an expansion was forced to fail,
4590 contradicting the general specification for all authenticators. Instead it
4591 was generating a temporary error. It now behaves as specified.
4593 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4594 (the order specifies the preference for clients). The order is now AES256,
4595 AES128, 3DES, ARCFOUR128.
4597 51. Small patch to Sieve code - explicitly set From: when generating an
4600 52. Exim crashed if a remote delivery caused a very long error message to be
4601 recorded - for instance if somebody sent an entire SpamAssassin report back
4602 as a large number of 550 error lines. This bug was coincidentally fixed by
4603 increasing the size of one of Exim's internal buffers (big_buffer) that
4604 happened as part of the Exiscan merge. However, to be on the safe side, I
4605 have made the code more robust (and fixed the comments that describe what
4608 53. Now that there can be additional text after "Completed" in log lines (if
4609 the queue_time_overall log selector is set), a one-byte patch to exigrep
4610 was needed to allow it to recognize "Completed" as not the last thing in
4613 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4614 patch that reportedly fixes this has been added. I am not expert enough to
4615 create a test for it. This is what the patch creator wrote:
4617 "I found a little strange behaviour of ldap code when working with
4618 Windows 2003 AD Domain, where users was placed in more than one
4619 Organization Units. When I tried to give exim partial DN, the exit code
4620 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4621 But simultaneously result of request was absolutely normal ldap result,
4622 so I produce this patch..."
4624 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4625 REFERENCE, so I have modified the code to exclude the patch when that macro
4628 55. Some experimental protocols are using DNS PTR records for new purposes. The
4629 keys for these records are domain names, not reversed IP addresses. The
4630 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4631 leaves it alone. Component reversal etc. now happens only for IP addresses.
4634 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4636 57. Double the size of the debug message buffer (to 2048) so that more of very
4637 long debug lines gets shown.
4639 58. The exicyclog utility now does better if the number of log files to keep
4640 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4642 59. Two changes related to the smtp_active_hostname option:
4644 (1) $smtp_active_hostname is now available as a variable.
4645 (2) The default for smtp_banner uses $smtp_active_hostname instead
4646 of $primary_hostname.
4648 60. The host_aton() function is supposed to be passed a string that is known
4649 to be a valid IP address. However, in the case of IPv6 addresses, it was
4650 not checking this. This is a hostage to fortune. Exim now panics and dies
4651 if the condition is not met. A case was found where this could be provoked
4652 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4653 components; fortuitously, this particular loophole had already been fixed
4654 by change 4.50/55 above.
4656 If there are any other similar loopholes, the new check in host_aton()
4657 itself should stop them being exploited. The report I received stated that
4658 data on the command line could provoke the exploit when Exim was running as
4659 exim, but did not say which command line option was involved. All I could
4660 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4661 running as the user.
4664 61. There was a buffer overflow vulnerability in the SPA authentication code
4665 (which came originally from the Samba project). I have added a test to the
4666 spa_base64_to_bits() function which I hope fixes it.
4669 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4670 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4672 63. The daemon start-up calls getloadavg() while still root for those OS that
4673 need the first call to be done as root, but it missed one case: when
4674 deliver_queue_load_max is set with deliver_drop_privilege. This is
4675 necessary for the benefit of the queue runner, because there is no re-exec
4676 when deliver_drop_privilege is set.
4678 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4679 This has been fixed.
4681 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4682 was in use, was not putting the data itself into the right store pool;
4683 consequently, it could be overwritten for a subsequent message in the same
4684 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4687 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4689 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4690 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4691 the function were treating the return as a boolean value, which happened to
4692 work because 0=false and not-0=true, but is not correct code.
4694 68. The host_aton() function was not handling scoped IPv6 addresses (those
4695 with, for example, "%eth0" on the end) correctly.
4697 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4698 negated items (that is, ~something) in unsigned ints. Some compilers
4699 apparently mutter when there is no cast.
4701 70. If an address verification called from an ACL failed, and did not produce a
4702 user-specific message (i.e. there was only a "system" message), nothing was
4703 put in $acl_verify_message. In this situation, it now puts the system
4706 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4707 session; change 4.31/43 added the unwanted input to the log line - except
4708 that it did not do this in the start of session case. It now does.
4710 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4711 This is wrong and can cause the other end to generate a synchronization
4712 error if it is another Exim or anything else that does the synchronization
4713 check. A QUIT command is no longer sent after a timeout.
4715 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4716 during host lookups.
4718 74. The defer_ok option of callout verification was not working if it was used
4719 when verifying addresses in header lines, that is, for this case:
4721 verify = header_sender/callout=defer_ok
4723 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4724 those file descriptors could be used for SMTP connections. If anything
4725 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4726 could be sent to the SMTP client, causing chaos. The daemon now opens
4727 stdin, stdout, and stderr to /dev/null when it puts itself into the
4730 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4731 log by default. The user can override this with suitable Perl magic.
4733 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4734 the log message when discard triggers, was not working for the DATA ACL or
4735 for the non-SMTP ACL.
4737 78. Error message wording change in sieve.c.
4739 79. If smtp_accept_max_per_host was set, the number of connections could be
4740 restricted to fewer than expected, because the daemon was trying to set up
4741 a new connection before checking whether the processes handling previous
4742 connections had finished. The check for completed processes is now done
4743 earlier. On busy systems, this bug wouldn't be noticed because something
4744 else would have woken the daemon, and it would have reaped the completed
4747 80. If a message was submitted locally by a user whose login name contained one
4748 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4749 It caused a spool format error. I have fixed the spool reading code. A
4750 related problem was that the "from" clause in the Received: line became
4751 illegal because of the space(s). It is now covered by ${quote_local_part.
4753 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4756 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4758 83. Patch from Sieve maintainer to fix unterminated string problem in
4759 "vacation" handling.
4761 84. Some minor changes to the Linux configuration files to help with other
4762 OS variants using glibc.
4764 85. One more patch for Sieve to update vacation handling to latest spec.
4767 ----------------------------------------------------
4768 See the note above about the 4.44 and 4.50 releases.
4769 ----------------------------------------------------
4775 1. Change 4.43/35 introduced a bug that caused file counts to be
4776 incorrectly computed when quota_filecount was set in an appendfile
4779 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4780 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4783 3. Give more explanation in the error message when the command for a transport
4784 filter fails to execute.
4786 4. There are several places where Exim runs a non-Exim command in a
4787 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4788 was being done only for the command run by the queryprogram router. It is
4789 now done for all such subprocesses. The other cases are: ${run, transport
4790 filters, and the commands run by the lmtp and pipe transports.
4792 5. Some older OS have a limit of 256 on the maximum number of file
4793 descriptors. Exim was using setrlimit() to set 1000 as a large value
4794 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4795 systems. I've change it so that if it can't get 1000, it tries for 256.
4797 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4798 was an oversight, and furthermore, ever since the addition of extra
4799 controls (e.g. 4.43/32), the checks on when to allow different forms of
4800 "control" were broken. There should now be diagnostics for all cases when a
4801 control that does not make sense is encountered.
4803 7. $recipients is now available in the predata ACL (oversight).
4805 8. Tidy the search cache before the fork to do a delivery from a message
4806 received from the command line. Otherwise the child will trigger a lookup
4807 failure and thereby defer the delivery if it tries to use (for example) a
4808 cached ldap connection that the parent has called unbind on.
4810 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4811 of $address_data from the recipient verification was clobbered by the
4812 sender verification.
4814 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4815 was its contents. (It was OK if the option was not defined at all.)
4817 11. A "Completed" log line is now written for messages that are removed from
4818 the spool by the -Mrm option.
4820 12. $host_address is now set to the target address during the checking of
4821 ignore_target_hosts.
4823 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4824 being passed; this would have caused $sender_host_name to have been used if
4825 matching the list had actually called for a host name (not very likely,
4826 since this list is usually IP addresses). A host name is now passed as
4829 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4830 code that specifies a non-restarting handler (typically sigaction() in
4831 modern systems) in an attempt to fix a rare and obscure crash bug.
4833 15. Narrowed the window for a race in the daemon that could cause it to ignore
4834 SIGCHLD signals. This is not a major problem, because they are used only to
4835 wake it up if nothing else does.
4837 16. A malformed maildirsize file could cause Exim to calculate negative values
4838 for the mailbox size or file count. Odd effects could occur as a result.
4839 The maildirsize information is now recalculated if the size or filecount
4842 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4843 support for a long time. Removed HAVE_SYS_VFS_H.
4845 18. Updated exipick to current release from John Jetmore.
4847 19. Allow an empty sender to be matched against a lookup in an address list.
4848 Previously the only cases considered were a regular expression, or an
4851 20. Exim went into a mad DNS lookup loop when doing a callout where the
4852 host was specified on the transport, if the DNS lookup yielded more than
4855 21. The RFC2047 encoding function was originally intended for short strings
4856 such as real names; it was not keeping to the 75-character limit for
4857 encoded words that the RFC imposes. It now respects the limit, and
4858 generates multiple encoded words if necessary. To be on the safe side, I
4859 have increased the buffer size for the ${rfc2047: expansion operator from
4862 22. Failure to deliver a bounce message always caused it to be frozen, even if
4863 there was an errors_to setting on the router. The errors_to setting is now
4866 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4867 canonical form (fully expanded) before being placed in
4868 $sender_host_address.
4870 24. Updated eximstats to version 1.33
4872 25. Include certificate and key file names in error message when GnuTLS fails
4873 to set them up, because the GnuTLS error message doesn't include the name
4874 of the failing file when there is a problem reading it.
4876 26. Expand error message when OpenSSL has problems setting up cert/key files.
4879 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4880 (this can affect the format of dates).
4882 28. exim_tidydb, when checking for the continued existence of a message for
4883 which it has found a message-specific retry record, was not finding
4884 messages that were in split spool directories. Consequently, it was
4885 deleting retry records that should have stayed in existence.
4887 29. eximstats updated to version 1.35
4888 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4889 1.35 - bugfix such that pie charts by volume are generated correctly
4891 30. The SPA authentication driver was not abandoning authentication and moving
4892 on to the next authenticator when an expansion was forced to fail,
4893 contradicting the general specification for all authenticators. Instead it
4894 was generating a temporary error. It now behaves as specified.
4896 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4897 (the order specifies the preference for clients). The order is now AES256,
4898 AES128, 3DES, ARCFOUR128.
4900 31. Small patch to Sieve code - explicitly set From: when generating an
4903 32. Exim crashed if a remote delivery caused a very long error message to be
4904 recorded - for instance if somebody sent an entire SpamAssassin report back
4905 as a large number of 550 error lines. This bug was coincidentally fixed by
4906 increasing the size of one of Exim's internal buffers (big_buffer) that
4907 happened as part of the Exiscan merge. However, to be on the safe side, I
4908 have made the code more robust (and fixed the comments that describe what
4911 33. Some experimental protocols are using DNS PTR records for new purposes. The
4912 keys for these records are domain names, not reversed IP addresses. The
4913 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4914 leaves it alone. Component reversal etc. now happens only for IP addresses.
4917 34. The host_aton() function is supposed to be passed a string that is known
4918 to be a valid IP address. However, in the case of IPv6 addresses, it was
4919 not checking this. This is a hostage to fortune. Exim now panics and dies
4920 if the condition is not met. A case was found where this could be provoked
4921 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4922 components; fortuitously, this particular loophole had already been fixed
4923 by change 4.50/55 or 4.44/33 above.
4925 If there are any other similar loopholes, the new check in host_aton()
4926 itself should stop them being exploited. The report I received stated that
4927 data on the command line could provoke the exploit when Exim was running as
4928 exim, but did not say which command line option was involved. All I could
4929 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4930 running as the user.
4933 35. There was a buffer overflow vulnerability in the SPA authentication code
4934 (which came originally from the Samba project). I have added a test to the
4935 spa_base64_to_bits() function which I hope fixes it.
4938 36. The daemon start-up calls getloadavg() while still root for those OS that
4939 need the first call to be done as root, but it missed one case: when
4940 deliver_queue_load_max is set with deliver_drop_privilege. This is
4941 necessary for the benefit of the queue runner, because there is no re-exec
4942 when deliver_drop_privilege is set.
4944 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4945 was in use, was not putting the data itself into the right store pool;
4946 consequently, it could be overwritten for a subsequent message in the same
4947 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4950 38. Sometimes the final signoff response after QUIT could fail to get
4951 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4952 < 0 before doing a fflush(). This bug looks as though it goes back to the
4953 introduction of TLS in release 3.20, but "sometimes" must have been rare
4954 because the tests only now provoked it.
4960 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4961 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4962 Now it returns an integer. A number of calls were still expecting a T/F
4963 return. Fortuitously, in all cases, the tests worked in OK situations,
4964 which is the norm. However, things would have gone wrong on any write
4965 failures on the smtp file descriptor. This function is used when sending
4966 messages over SMTP and also when doing verify callouts.
4968 2. When Exim is called to do synchronous delivery of a locally submitted
4969 message (the -odf or -odi options), it no longer closes stderr before doing
4972 3. Implemented the mua_wrapper option.
4974 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4976 5. Implemented the functions header_remove(), header_testname(),
4977 header_add_at_position(), and receive_remove_recipient(), and exported them
4980 6. If an ACL "warn" statement specified the addition of headers, Exim already
4981 inserted X-ACL-Warn: at the start if there was no header name. However, it
4982 was not making this test for the second and subsequent header lines if
4983 there were newlines in the string. This meant that an invalid header could
4984 be inserted if Exim was badly configured.
4986 7. Allow an ACL "warn" statement to add header lines at the start or after all
4987 the Received: headers, as well as at the end.
4989 8. Added the rcpt_4xx retry error code.
4991 9. Added postmaster_mailfrom=xxx to callout verification option.
4993 10. Added mailfrom=xxxx to the callout verification option, for verify=
4996 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4997 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4999 12. Inserted some casts to stop certain compilers warning when using pointer
5000 differences as field lengths or precisions in printf-type calls (mostly
5001 affecting debugging statements).
5003 13. Added optional readline() support for -be (dynamically loaded).
5005 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5006 same clock tick as a message's arrival, so that its received time was the
5007 same as the "first fail" time on the retry record, and that message
5008 remained on the queue past the ultimate address timeout, every queue runner
5009 would try a delivery (because it was past the ultimate address timeout) but
5010 after another failure, the ultimate address timeout, which should have then
5011 bounced the address, did not kick in. This was a "< instead of <=" error;
5012 in most cases the first failure would have been in the next clock tick
5013 after the received time, and all would be well.
5015 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5016 being recognized when the domain list was tested by the match_domain
5017 condition in an expansion string.
5019 16. Added the ${str2b64: operator.
5021 17. Exim was always calling setrlimit() to set a large limit for the number of
5022 processes, without checking whether the existing limit was already
5023 adequate. (It did check for the limit on file descriptors.) Furthermore,
5024 errors from getrlimit() and setrlimit() were being ignored. Now they are
5025 logged to the main and panic logs, but Exim does carry on, to try to do its
5026 job under whatever limits there are.
5028 18. Imported PCRE 5.0.
5030 19. Trivial typo in log message " temporarily refused connection" (the leading
5033 20. If the log selector return_path_on_delivery was set and an address was
5034 redirected to /dev/null, the delivery process crashed because it assumed
5035 that a return path would always be set for a "successful" delivery. In this
5036 case, the whole delivery is bypassed as an optimization, and therefore no
5039 21. Internal re-arrangement: the function for sending a challenge and reading
5040 a response while authentication was assuming a zero-terminated challenge
5041 string. It's now changed to take a pointer and a length, to allow for
5042 binary data in such strings.
5044 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5046 23. Exim was not respecting finduser_retries when seeking the login of the
5047 uid under which it was called; it was always trying 10 times. (The default
5048 setting of finduser_retries is zero.) Also, it was sleeping after the final
5049 failure, which is pointless.
5051 24. Implemented tls_on_connect_ports.
5053 25. Implemented acl_smtp_predata.
5055 26. If the domain in control=submission is set empty, Exim assumes that the
5056 authenticated id is a complete email address when it generates From: or
5057 Sender: header lines.
5059 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5060 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5061 chown and chgrp in /bin and hostname in /usr/bin.
5063 28. Exim was keeping the "process log" file open after each use, just as it
5064 does for the main log. This opens the possibility of it remaining open for
5065 long periods when the USR1 signal hits a daemon. Occasional processlog
5066 errors were reported, that could have been caused by this. Anyway, it seems
5067 much more sensible not to leave this file open at all, so that is what now
5070 29. The long-running daemon process does not normally write to the log once it
5071 has entered its main loop, and it closes the log before doing so. This is
5072 so that log files can straightforwardly be renamed and moved. However,
5073 there are a couple of unusual error situations where the daemon does write
5074 log entries, and I had neglected to close the log afterwards.
5076 30. The text of an SMTP error response that was received during a remote
5077 delivery was being truncated at 512 bytes. This is too short for some of
5078 the long messages that one sometimes sees. I've increased the limit to
5081 31. It is now possible to make retry rules that apply only when a message has a
5082 specific sender, in particular, an empty sender.
5084 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5085 it possible to be selective about when SMTP synchronization is enforced.
5087 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5089 32. Implemented hosts_connection_nolog.
5091 33. Added an ACL for QUIT.
5093 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5096 35. Added mailbox_size and mailbox_filecount to appendfile.
5098 36. Added control = no_multiline_responses to ACLs.
5100 37. There was a bug in the logic of the code that waits for the clock to tick
5101 in the case where the clock went backwards by a substantial amount such
5102 that the microsecond fraction of "now" was more than the microsecond
5103 fraction of "then" (but the whole seconds number was less).
5105 38. Added support for the libradius Radius client library this is found on
5106 FreeBSD (previously only the radiusclient library was supported).
5112 1. When certain lookups returned multiple values in the form name=value, the
5113 quoting of the values was not always being done properly. Specifically:
5114 (a) If the value started with a double quote, but contained no whitespace,
5116 (b) If the value contained whitespace other than a space character (i.e.
5117 tabs or newlines or carriage returns) it was not quoted.
5118 This fix has been applied to the mysql and pgsql lookups by writing a
5119 separate quoting function and calling it from the lookup code. The fix
5120 should probably also be applied to nisplus, ibase and oracle lookups, but
5121 since I cannot test any of those, I have not disturbed their existing code.
5123 2. A hit in the callout cache for a specific address caused a log line with no
5124 reason for rejecting RCPT. Now it says "Previous (cached) callout
5125 verification failure".
5127 3. There was an off-by-one bug in the queryprogram router. An over-long
5128 return line was truncated at 256 instead of 255 characters, thereby
5129 overflowing its buffer with the terminating zero. As well as fixing this, I
5130 have increased the buffer size to 1024 (and made a note to document this).
5132 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5133 when Exim is waiting for an SMTP response from a remote server, Exim
5134 restarts its select() call on the socket, thereby resetting its timeout.
5135 This is not a problem when such interrupts are rare. Somebody set up a cron
5136 job to run exiwhat every 2 minutes, which is less than the normal select()
5137 timeout (5 or 10 minutes). This meant that the select() timeout never
5138 kicked in because it was always reset. I have fixed this by comparing the
5139 time when an interrupt arrives with the time at the start of the first call
5140 to select(). If more time than the timeout has elapsed, the interrupt is
5141 treated as a timeout.
5143 5. Some internal re-factoring in preparation for the addition of Sieve
5144 extensions (by MH). In particular, the "personal" test is moved to a
5145 separate function, and given an option for scanning Cc: and Bcc: (which is
5146 not set for Exim filters).
5148 6. When Exim created an email address using the login of the caller as the
5149 local part (e.g. when creating a From: or Sender: header line), it was not
5150 quoting the local part when it contained special characters such as @.
5152 7. Installed new OpenBSD configuration files.
5154 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5155 try to make them clearer.
5157 9. Callout options, other than the timeout value, were being ignored when
5158 verifying sender addresses in header lines. For example, when using
5160 verify = header_sender/callout=no_cache
5162 the cache was (incorrectly) being used.
5164 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5165 only the Cygwin environment.
5167 11. When return_path_on_delivery was set as a log selector, if different remote
5168 addresses in the same message used different return paths and parallel
5169 remote delivery occurred, the wrong values would sometimes be logged.
5170 (Whenever a remote delivery process finished, the return path value from
5171 the most recently started remote delivery process was logged.)
5173 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5174 lines when AUTH and/or TLS are in use. This is the "received protocol"
5175 field. Exim used to use "asmtp" for authenticated SMTP, without any
5176 indication (in the protocol name) for TLS use. Now it follows the RFC and
5177 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5178 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5179 names appear in log lines as well as in Received: header lines.
5181 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5182 extensions, and comparison tests, and to fix some bugs.
5184 14. Changes to the "personal" filter test:
5186 (1) The test was buggy in that it was just doing the equivalent of
5187 "contains" tests on header lines. For example, if a user's address was
5188 anne@some.where, the "personal" test would incorrectly be true for
5190 To: susanne@some.where
5192 This test is now done by extracting each address from the header in turn,
5193 and checking the entire address. Other tests that are part of "personal"
5194 are now done using regular expressions (for example, to check local parts
5195 of addresses in From: header lines).
5197 (2) The list of non-personal local parts in From: addresses has been
5198 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5199 taken from the Sieve specification recommendations.
5201 (3) If the message contains any header line starting with "List-" it is
5202 treated as non-personal.
5204 (4) The test for "circular" in the Subject: header line has been removed
5205 because it now seems ill-conceived.
5207 15. Minor typos in src/EDITME comments corrected.
5209 16. Installed latest exipick from John Jetmore.
5211 17. If headers_add on a router specified a text string that was too long for
5212 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5213 of string_sprintf() is now avoided.
5215 18. $message_body_size was not set (it was always zero) when running the DATA
5216 ACL and the local_scan() function.
5218 19. For the "mail" command in an Exim filter, no default was being set for
5219 the once_repeat time, causing a random time value to be used if "once" was
5220 specified. (If the value happened to be <= 0, no repeat happened.) The
5221 default is now 0s, meaning "never repeat". The "vacation" command was OK
5222 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5223 (I found it when inspecting the code).
5225 20. There is now an overall timeout for performing a callout verification. It
5226 defaults to 4 times the callout timeout, which applies to individual SMTP
5227 commands during the callout. The overall timeout applies when there is more
5228 than one host that can be tried. The timeout is checked before trying the
5229 next host. This prevents very long delays if there are a large number of
5230 hosts and all are timing out (e.g. when the network connections are timing
5231 out). The value of the overall timeout can be changed by specifying an
5232 additional sub-option for "callout", called "maxwait". For example:
5234 verify = sender/callout=5s,maxwait=20s
5236 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5237 to the end before writing, but this should make it even safer).
5239 22. Exim was forgetting that it had advertised PIPELINING for the second and
5240 subsequent messages on an SMTP connection. It was also not resetting its
5241 memory on STARTTLS and an internal HELO.
5243 23. When Exim logs an SMTP synchronization error within a session, it now
5244 records whether PIPELINING has been advertised or not.
5246 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5247 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5248 rather than long int.
5250 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5252 26. Added the never_mail option to autoreply.
5258 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5259 crash if the getsockname() call failed; this can happen if a connection is
5260 closed very soon after it is established. The problem was simply in the
5261 order in which certain operations were done, causing Exim to try to write
5262 to the SMTP stream before it had set up the file descriptor. The bug has
5263 been fixed by making things happen in the correct order.
5269 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5270 before the connection was closed, thus losing the rejection response.
5272 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5273 some early Solaris releases, but causes trouble in current releases where
5274 socklen_t is defined.
5276 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5279 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5282 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5283 that releasing the top of it at the end releases what was used for sub-
5284 expansions (unless the block got too big). However, discard this block if
5285 the first thing is a variable or header, so that we can use its block when
5286 it is dynamic (useful for very large $message_headers, for example).
5288 6. Lookups now cache *every* query, not just the most recent. A new, separate
5289 store pool is used for this. It can be recovered when all lookup caches are
5290 flushed. Lookups now release memory at the end of their result strings.
5291 This has involved some general refactoring of the lookup sources.
5293 7. Some code has been added to the store_xxx() functions to reduce the amount
5294 of flapping under certain conditions.
5296 8. log_incoming_interface used to affect only the <= reception log lines. Now
5297 it causes the local interface and port to be added to several more SMTP log
5298 lines, for example "SMTP connection from", and rejection lines.
5300 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5302 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5304 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5305 could be overwritten at the end of the current message (or the start of a
5306 new message if it was set in a HELO ACL). The value is now preserved for
5307 the duration of the SMTP connection.
5309 12. If a transport had a headers_rewrite setting, and a matching header line
5310 contained an unqualified address, that address was qualified, even if it
5311 did not match any rewriting rules. The underlying bug was that the values
5312 of the flags that permit the existence of unqualified sender and recipient
5313 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5314 non-local messages, and by -bnq for local messages) were not being
5315 preserved with the message after it was received.
5317 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5318 "next input=" as part of the text comprising the host identity instead of
5319 the correct text. The code was using the same buffer for two different
5320 strings. However, depending on which order the printing function evaluated
5321 its arguments, the bug did not always show up. Under Linux, for example, my
5322 test suite worked just fine.
5324 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5325 doesn't work with some older versions of Perl. It has been changed to "my",
5326 which in any case is probably the better facility to use.
5328 15. A really picky compiler found some instances of statements for creating
5329 error messages that either had too many or two few arguments for the format
5332 16. The size of the buffer for calls to the DNS resolver has been increased
5333 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5334 for addresses that have a lot of PTR records. This alleviates a problem; it
5335 does not fully solve it.
5337 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5338 buffer now truncates the list and logs the incident, which is the same
5339 action as happens when Exim is looking up a host name and its aliases.
5340 Previously in this situation something unpredictable would happen;
5341 sometimes it was "internal error: store_reset failed".
5343 18. If a server dropped the connection unexpectedly when an Exim client was
5344 using GnuTLS and trying to read a response, the client delivery process
5345 crashed while trying to generate an error log message.
5347 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5348 string, for example:
5350 warn message = H1: something\nH2: something
5352 the text was added as a single header line from Exim's point of view
5353 though it ended up OK in the delivered message. However, searching for the
5354 second and subsequent header lines using $h_h2: did not work. This has been
5355 fixed. Similarly, if a system filter added multiple headers in this way,
5356 the routers could not see them.
5358 20. Expanded the error message when iplsearch is called with an invalid key to
5359 suggest using net-iplsearch in a host list.
5361 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5362 ACLs are no longer actually imposed (and a message to that effect is
5365 22. If a "gecos" field in a passwd entry contained escaped characters, in
5366 particular, if it contained a \" sequence, Exim got it wrong when building
5367 a From: or a Sender: header from that name. A second bug also caused
5368 incorrect handling when an unquoted " was present following a character
5369 that needed quoting.
5371 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5372 was not being matched caselessly.
5374 24. Arranged for all hyphens in the exim.8 source to be escaped with
5377 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5378 properly. Recipient callout cache records were still being keyed to include
5379 the sender, even when use_sender was set false. This led to far more
5380 callouts that were necessary. The sender is no longer included in the key
5381 when use_sender is false.
5383 26. Added "control = submission" modifier to ACLs.
5385 27. Added the ${base62d: operator to decode base 62 numbers.
5387 28. dnsdb lookups can now access SRV records.
5389 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5390 the configuration file.
5392 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5393 (-v) mode. This makes the output for a verbose queue run more intelligible.
5395 31. Added a use_postmaster feature to recipient callouts.
5397 32. Added the $body_zerocount variable, containing the number of binary zero
5398 bytes in the message body.
5400 33. The time of last modification of the "new" subdirectory is now used as the
5401 "mailbox time last read" when there is a quota error for a maildir
5404 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5406 35. Added +ignore_unknown as a special item in host lists.
5408 36. Code for decoding IPv6 addresses in host lists is now included, even if
5409 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5410 address was recognized as an IP address, but was then not correctly decoded
5411 into binary, causing unexpected and incorrect effects when compared with
5418 1. Very minor rewording of debugging text in manualroute to say "list of
5419 hosts" instead of "hostlist".
5421 2. If verify=header_syntax was set, and a header line with an unqualified
5422 address (no domain) and a large number of spaces between the end of the
5423 name and the colon was received, the reception process suffered a buffer
5424 overflow, and (when I tested it) crashed. This was caused by some obsolete
5425 code that should have been removed. The fix is to remove it!
5427 3. When running in the test harness, delay a bit after writing a bounce
5428 message to get a bit more predictability in the log output.
5430 4. Added a call to search_tidyup() just before forking a reception process. In
5431 theory, someone could use a lookup in the expansion of smtp_accept_max_
5432 per_host which, without the tidyup, could leave open a database connection.
5434 5. Added the variables $recipient_data and $sender_data which get set from a
5435 lookup success in an ACL "recipients" or "senders" condition, or a router
5436 "senders" option, similar to $domain_data and $local_part_data.
5438 6. Moved the writing of debug_print from before to after the "senders" test
5441 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5442 problems for message scanning, either using a data ACL, or using
5443 local_scan() because the Received: header was not generated till after they
5444 were called (in order to set the time as the time of reception completion).
5445 I have revised the way this works. The header is now generated after the
5446 body is received, but before the ACL or local_scan() are called. After they
5447 are run, the timestamp in the header is updated.
5453 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5454 before starting a queue runner without re-exec. This happened only when
5455 deliver_drop_privilege was set or when the Exim user was set to root. The
5456 effect of the bug was that timeouts during subsequent deliveries caused
5457 crashes instead of being properly handled. The handler is now left at its
5458 default (and expected) setting.
5460 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5461 message, again when deliver_drop_privilege is set or Exim is run as root.
5462 The bug described in (1) was not present in this case, but the tidying up
5463 of the other signals was missing. I have made the two cases consistent.
5465 3. The ignore_target_hosts setting on a manualroute router was being ignored
5466 for hosts that were looked up using the /MX notation.
5468 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5471 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5472 operated on the sender address. After changing the $sender_address to <>
5473 for the sender address verify, Exim was re-instated it as the original
5474 (before rewriting) address, but remembering that it had rewritten it, so it
5475 wasn't rewriting it again. This bug also had the effect of breaking the
5476 sender address verification caching when the sender address was rewritten.
5478 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5479 This has been changed so that if the ip literal address matches
5480 ignore_target_hosts, the router declines.
5482 7. Added expansion conditions match_domain, match_address, and match_local_
5483 part (NOT match_host).
5485 8. The placeholder for the Received: header didn't have a length field set.
5487 9. Added code to Exim itself and to exim_lock to test for a specific race
5488 condition that could lead to file corruption when using MBX delivery. The
5489 issue is with the lockfile that is created in /tmp. If this file is removed
5490 after a process has opened it but before that process has acquired a lock,
5491 there is the potential for a second process to recreate the file and also
5492 acquire a lock. This could lead to two Exim processes writing to the file
5493 at the same time. The added code performs the same test as UW imapd; it
5494 checks after acquiring the lock that its file descriptor still refers to
5495 the same named file.
5497 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5498 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5501 11. Added the smtp_active_hostname option. If used, this will typically be made
5502 to depend on the incoming interface address. Because $interface_address is
5503 not set up until the daemon has forked a reception process, error responses
5504 that can happen earlier (such as "too many connections") no longer contain
5507 12. If an expansion in a condition on a "warn" statement fails because a lookup
5508 defers, the "warn" statement is abandoned, and the next ACL statement is
5509 processed. Previously this caused the whole ACL to be aborted.
5511 13. Added the iplsearch lookup type.
5513 14. Added ident_timeout as a log selector.
5515 15. Added tls_certificate_verified as a log selector.
5517 16. Added a global option tls_require_ciphers (compare the smtp transport
5518 option of the same name). This controls incoming TLS connections.
5520 17. I finally figured out how to make tls_require_ciphers do a similar thing
5521 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5522 before starting the TLS session.
5524 18. Tabs are now shown as \t in -bP output.
5526 19. If the log selector return_path_on_delivery was set, Exim crashed when
5527 bouncing a message because it had too many Received: header lines.
5529 20. If two routers both had headers_remove settings, and the first one included
5530 a superfluous trailing colon, the final name in the first list and the
5531 first name in the second list were incorrectly joined into one item (with a
5532 colon in the middle).
5538 1. Added -C and -D options to the exinext utility, mainly to make it easier
5539 to include in the automated testing, but these could be helpful when
5540 multiple configurations are in use.
5542 2. The exinext utility was not formatting the output nicely when there was
5543 an alternate port involved in the retry record key, nor when there was a
5544 message id as well (for retries that were specific to a specific message
5545 and a specific host). It was also confused by IPv6 addresses, because of
5546 the additional colons they contain. I have fixed the IPv4 problem, and
5547 patched it up to do a reasonable job for IPv6.
5549 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5550 delivery, the log line now contains "pipelined" if PIPELINING was used.
5552 4. An SMTP transport process used to panic and die if the bind() call to set
5553 an explicit outgoing interface failed. This has been changed; it is now
5554 treated in the same way as a connect() failure.
5556 5. A reference to $sender_host_name in the part of a conditional expansion
5557 that was being skipped was still causing a DNS lookup. This no longer
5560 6. The def: expansion condition was not recognizing references to header lines
5561 that used bh_ and bheader_.
5563 7. Added the _cache feature to named lists.
5565 8. The code for checking quota_filecount in the appendfile transport was
5566 allowing one more file than it should have been.
5568 9. For compatibility with Sendmail, the command line option
5576 and sets the incoming protocol and host name (for trusted callers). The
5577 host name and its colon can be omitted when only the protocol is to be set.
5578 Note the Exim already has two private options, -pd and -ps, that refer to
5579 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5580 "s", but I don't think that's a major issue.
5582 10. A number of refactoring changes to the code, none of which should affect
5585 (a) The number of logging options was getting close to filling up the
5586 32-bit word that was used as a bit map. I have split them into two classes:
5587 those that are passed in the argument to log_write(), and those that are
5588 only ever tested independently outside of that function. These are now in
5589 separate 32-bit words, so there is plenty of room for expansion again.
5590 There is no change in the user interface or the logging behaviour.
5592 (b) When building, for example, log lines, the code previously used a
5593 macro that called string_cat() twice, in order to add two strings. This is
5594 not really sufficiently general. Furthermore, there was one instance where
5595 it was actually wrong because one of the argument was used twice, and in
5596 one call a function was used. (As it happened, calling the function twice
5597 did not affect the overall behaviour.) The macro has been replaced by a
5598 function that can join an arbitrary number of extra strings onto a growing
5601 (c) The code for expansion conditions now uses a table and a binary chop
5602 instead of a serial search (which was left over from when there were very
5603 few conditions). Also, it now recognizes conditions like "pam" even when
5604 the relevant support is not compiled in: a suitably worded error message is
5605 given if an attempt is made to use such a condition.
5607 11. Added ${time_interval:xxxxx}.
5609 12. A bug was causing one of the ddress fields not to be passed back correctly
5610 from remote delivery subprocesses. The field in question was not being
5611 subsequently used, so this caused to problems in practice.
5613 13. Added new log selectors queue_time and deliver_time.
5615 14. Might have fixed a bug in maildirsizefile handling that threw up
5616 "unexpected character" debug warnings, and recalculated the data
5617 unnecessarily. In any case, I expanded the warning message to give more
5620 15. Added the message "Restricted characters in address" to the statements in
5621 the default ACL that block characters like @ and % in local parts.
5623 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5624 Three changes have been made:
5626 (a) There was a serious bug; a negative response to MAIL caused the whole
5627 recipient domain to be cached as invalid, thereby blocking all messages
5628 to all local parts at the same domain, from all senders. This bug has
5629 been fixed. The domain is no longer cached after a negative response to
5630 MAIL if the sender used is not empty.
5632 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5635 (c) A new callout option, "use_sender" has been added for people who want
5636 the modified behaviour.
5642 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5645 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5646 indeed breaks things for older releases.
5648 3. Added additional logging to the case where there is a problem reading data
5649 from a filter that is running in a subprocess using a pipe, in order to
5650 try to track down a specific problem.
5652 4. Testing facility fudge: when running in the test harness and attempting
5653 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5654 getting "No route to host". Convert this to a timeout.
5656 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5659 6. Some OS don't have socklen_t but use size_t instead. This affects the
5660 fifth argument of getsockopt() amongst other things. This is now
5661 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5662 can be set for individual OS. I have set it for SunOS5, OSF1, and
5663 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5664 some earlier ones do not.
5666 7. Change 4.30/15 was not doing the test caselessly.
5668 8. The standard form for an IPv6 address literal was being rejected by address
5669 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5670 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5671 this, as well as the form without the "IPv6" on the front (but only when
5672 address literals are enabled, of course).
5674 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5676 10. Exim crashed if a message with an empty sender address specified by -f
5677 encountered a router with an errors_to setting. This could be provoked only
5678 by a command such as
5682 where an empty string was supplied; "<>" did not hit this bug.
5684 11. Installed PCRE release 4.5.
5686 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5687 remained set. It is now erased.
5689 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5690 times from message ids (which are base 36 rather than the normal 62).
5692 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5693 were being counted as actual protocol errors, and logged if the log
5694 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5695 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5696 and DATA following a set of rejected RCPTs do not count as protocol errors.
5697 In other words, Exim assumes they were pipelined, though this may not
5698 actually be the case. Of course, in all cases the client gets an
5699 appropriate error code.
5701 15. If a lookup fails in an ACL condition, a message about the failure may
5702 be available; it is used if testing the ACL cannot continue, because most
5703 such messages specify what the cause of the deferral is. However, some
5704 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5705 that caused an old message to be retained and used if a later statement
5706 caused a defer, replacing the real cause of the deferral.
5708 16. If an IP address had so many PTR records that the DNS lookup buffer
5709 was not large enough to hold them, Exim could crash while trying to process
5710 the truncated data. It now detects and logs this case.
5712 17. Further to 4.21/58, another change has been made: if (and only if) the
5713 first line of a message (the first header line) ends with CRLF, a bare LF
5714 in a subsequent header line has a space inserted after it, so as not to
5715 terminate the header.
5717 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5718 unnecessarily, used atoi() instead of strtol(), and didn't check the
5719 termination when getting file sizes from file names by regex.
5721 19. Completely re-implemented the support for maildirsize files, in the light
5722 of a number of problems with the previous contributed implementation
5723 (4.30/29). In particular:
5725 . If the quota is zero, the maildirsize file is maintained, but no quota is
5728 . If the maildir directory does not exist, it is created before any attempt
5729 to write a maildirsize file.
5731 . The quota value in the file is just a cache; if the quota is changed in
5732 the transport, the new value overrides.
5734 . A regular expression is available for excluding directories from the
5737 20. The autoreply transport checks the characters in options that define the
5738 message's headers; it allows continued headers, but it was checking with
5739 isspace() after an embedded newline instead of explicitly looking for a
5742 21. If all the "regular" hosts to which an address was routed had passed their
5743 expiry times, and had not reached their retry times, the address was
5744 bounced, even if fallback hosts were defined. Now Exim should go on to try
5747 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5748 equivalent code in the SMTP transport. Some hosts send humungous responses
5749 to HELO/EHLO, more than 1024 it seems.
5751 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5752 gives compiler warnings in some environments. I've now done it "properly",
5755 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5756 (because of problems with the built-in one) was declared to return uschar *
5757 instead of char *, causing compiler failure.
5759 25. Fixed a file descriptor leak when processing alias/forward files.
5761 26. Fixed a minor format string issue in dbfn.c.
5763 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5765 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5766 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5769 29. When a sender address is verified, it is cached, to save repeating the test
5770 when there is more than one recipient in a message. However, when the
5771 verification involves a callout, it is possible for different callout
5772 options to be set for different recipients. It is too complicated to keep
5773 track of this in the cache, so now Exim always runs a verification when a
5774 callout is required, relying on the callout cache for the optimization.
5775 The overhead is duplication of the address routing, but this should not be
5778 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5779 to be verified with callout=postmaster, and the main callout worked but the
5780 postmaster check failed, the verification correctly failed. However, if a
5781 subsequent RCPT command asked for sender verification *without* the
5782 postmaster check, incorrect caching caused this verification also to fail,
5785 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5786 it was not caching the DNS options (qualify_single, search_parents) that
5787 were used when the lookup failed. A subsequent lookup with different
5788 options therefore always gave the same answer, though there were cases
5789 where it should not have. (Example: a "domains = !$mx_any" option on a
5790 dnslookup router: the "domains" option is always processed without any
5791 widening, but the router might have qualify_single set.) Now Exim uses the
5792 cached value only when the same options are set.
5794 32. Added John Jetmore's "exipick" utility to the distribution.
5796 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5797 than a timeout (e.g. a certificate is required, and is not provided), an
5798 Exim server now closes the connection immediately. Previously it waited for
5799 the client to close - but if the client is SSL, it seems that they each
5800 wait for each other, leading to a delay before one of them times out.
5802 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5803 maintained 0.8.x compatibility because I don't think many are using it, and
5804 it is clearly obsolete.
5806 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5809 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5810 client certificate was expired. A simple patch fixes this, though I don't
5811 understand the full logic of why the verify callback is called multiple
5814 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5815 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5816 which causes problems with some clients (such as the Certicom SSL Plus
5817 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5818 disables the coutermeasure allowing Eudora to connect."
5820 38. Exim was not checking that a write() to a log file succeeded. This could
5821 lead to Bad Things if a log got too big, in particular if it hit a file
5822 size limit. Exim now panics and dies if it cannot write to a log file, just
5823 as it does if it cannot open a log file.
5825 39. Modified OS/Makefile-Linux so that it now contains
5827 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5829 The two -D definitions ensure that Exim is compiled with large file
5830 support, which makes it possible to handle log files that are bigger than
5833 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5834 instance) a domain was checked against a named list that involved a lookup,
5835 causing $domain_data to be set, then another domain was checked against the
5836 same list, then the first domain was re-checked, the value of $domain_data
5837 after the final check could be wrong. In particular, if the second check
5838 failed, it could be set empty. This bug probably also applied to
5841 41. The strip_trailing_dot option was not being applied to the address given
5842 with the -f command-line option.
5844 42. The code for reading a message's header from the spool was incrementing
5845 $received_count, but never initializing it. This meant that the value was
5846 incorrect (doubled) while delivering a message in the same process in which
5847 it was received. In the most common configuration of Exim, this never
5848 happens - a fresh exec is done - but it can happen when
5849 deliver_drop_privilege is set.
5851 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5852 it now includes up to 150 characters of the unexpected data in the log
5855 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5856 and building data strings. The size of both of these buffers was 10 000
5857 bytes - far larger than anybody would *ever* want, thought I. Needless to
5858 say, somebody hit the limit. I have increased the maximum line length to
5859 20 000 and the maximum data length of concatenated lines to 100 000. I have
5860 also fixed two bugs, because there was no checking on these buffers. Tsk,
5861 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5862 buffer is too small.
5864 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5865 lsearch lookups. Now it does.
5867 46. When parsing a route_list item in a manualroute router, a fixed-length
5868 buffer was used for the list of hosts. I made this 1024 bytes long,
5869 thinking that nobody would ever have a list of hosts that long. Wrong.
5870 Somebody had a whole pile of complicated expansion conditions, and the
5871 string was silently truncated, leading to an expansion error. It turns out
5872 that it is easier to change to an unlimited length (owing to other changes
5873 that have happened since this code was originally written) than to build
5874 structure for giving a limitation error. The length of the item that
5875 expands into the list of hosts is now unlimited.
5877 47. The lsearch lookup could not handle data where the length of text line was
5878 more than 4095 characters. Such lines were truncated, leading to shortened
5879 data being returned. It should now handle lines of any length.
5881 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5882 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5885 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5886 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5887 no longer happens in comment lines. A list of replacements is now placed
5888 at the head of all of the source files, except those whose only change is
5889 to replace PERL_COMMAND in the very first #! line.
5891 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5892 messages on the queue, with a bottom-up merge sort, using code contributed
5893 by Michael Haardt. This should make operations like -bp somewhat faster on
5894 large queues. It won't affect queue runners, except when queue_run_in_order
5897 51. Installed eximstats 1.31 in the distribution.
5899 52. Added support for SRV lookups to the dnslookup router.
5901 53. If an ACL referred to $message_body or $message_body_end, the value was not
5902 reset for any messages that followed in the same SMTP session.
5904 54. The store-handling optimization for building very long strings was not
5905 differentiating between the different store pools. I don't think this
5906 actually made any difference in practice, but I've tidied it.
5908 55. While running the routers to verify a sender address, $sender_address
5909 was still set to the sender address. This is wrong, because when routing to
5910 send a bounce to the sender, it would be empty. Therefore, I have changed
5911 it so that, while verifying a sender address, $sender_address is set to <>.
5912 (There is no change to what happens when verifying a recipient address.)
5914 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5915 target A or AAAA records (if not already returned) without resetting the
5916 qualify_single or search_parents options of the DNS resolver. These are
5917 inappropriate in this case because the targets of MX and SRV records must
5918 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5919 target that, when qualified, matched something in the local domain. These
5920 two options are now turned off when doing these lookups.
5922 57. It seems that at least some releases of Reiserfs (which does not have the
5923 concept of a fixed number of inodes) returns zero and not -1 for the
5924 number of available inodes. This interacted badly with check_spool_inodes,
5925 which assumed that -1 was the "no such thing" setting. What I have done is
5926 to check that the total number of inodes is greater than zero before doing
5927 the test of how many are available.
5929 58. When a "warn" ACL statement has a log_message modifier, the message is
5930 remembered, and not repeated. This is to avoid a lot of repetition when a
5931 message has many recipients that cause the same warning to be written.
5932 However, Exim was preserving the list of already written lines for an
5933 entire SMTP session, which doesn't seem right. The memory is now reset if a
5934 new message is started.
5936 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5937 debugging output unless log_rewrite was also set.
5939 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5940 of (int)(handle) when we know that handle contains (void *)(-1).
5942 61. The Exim daemon panic-logs an error return when it closes the incoming
5943 connection. However "connection reset by peer" seems to be common, and
5944 isn't really an error worthy of noting specially, so that particular error
5947 62. When Exim is trying to find all the local interfaces, it used to panic and
5948 die if the ioctl to get the interface flags failed. However, it seems that
5949 on at least one OS (Solaris 9) it is possible to have an interface that is
5950 included in the list of interfaces, but for which you get a failure error
5951 for this call. This happens when the interface is not "plumbed" into a
5952 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5953 failure of the "get flags" call assumes that the interface is down.
5955 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5956 makes life easier for people who are doing arithmetic on fields extracted
5957 from dates, where you often get leading zeros that should not be
5958 interpreted as octal.
5960 64. Added qualify_domain to the redirect router, to override the global
5963 65. If a pathologically long header line contained very many addresses (the
5964 report of this problem mentioned 10 000) and each of them was rewritten,
5965 Exim could use up a very large amount of memory. (It kept on making new
5966 copies of the header line as it rewrote, and never released the old ones.)
5967 At the expense of a bit more processing, the header rewriting function has
5968 been changed so that it no longer eats memory in this way.
5970 66. The generation of the Received: header has been moved from the time that a
5971 message starts to be received, to the time that it finishes. The timestamp
5972 in the Received: header should now be very close to that of the <= log
5973 line. There are two side-effects of this change:
5975 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5976 logged header lines no longer include the local Received: line, because
5977 it has not yet been created. The same applies to a copy of the message
5978 that is returned to a non-SMTP sender when a message is rejected.
5980 (b) When a filter file is tested using -bf, no additional Received: header
5981 is added to the test message. After some thought, I decided that this
5984 This change does not affect the value of $received_for. It is still set
5985 after address rewriting, but before local_scan() is called.
5987 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5989 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5990 gave an unhelpful panic error message, and a defer error. I have managed to
5991 change this behaviour so that it now rejects any supplied certificate,
5992 which seems right, as the list of acceptable certificates is empty.
5994 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5995 gave an unhelpful defer error. I have not managed to make this reject any
5996 supplied certificates, but the error message it gives is "no certificate
5997 supplied", which is not helpful.
5999 70. exigrep's output now also includes lines that are not associated with any
6000 message, but which match the given pattern. Implemented by a patch from
6001 Martin Sluka, which also tidied up the Perl a bit.
6003 71. Recipient callout verification, like sender verification, was using <> in
6004 the MAIL FROM command. This isn't really the right thing, since the actual
6005 sender may affect whether the remote host accepts the recipient or not. I
6006 have changed it to use the actual sender in the callout; this means that
6007 the cache record is now keyed on a recipient/sender pair, not just the
6008 recipient address. There doesn't seem to be a real danger of callout loops,
6009 since a callout by the remote host to check the sender would use <>.
6010 [SEE ABOVE: changed after hitting problems.]
6012 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6013 temporary errors. However, in the case of such a code being given after
6014 the end of a data transmission (i.e. after ".") Exim was failing to write
6015 a retry record for the message. (Yes, there was some broken host that was
6016 actually sending 8xx at this point.)
6018 73. An unknown lookup type in a host list could cause Exim to panic-die when
6019 the list was checked. (An example that provoked this was putting <; in the
6020 middle of a list instead of at the start.) If this happened during a DATA
6021 ACL check, a -D file could be left lying around. This kind of configuration
6022 error no longer causes Exim to die; instead it causes a defer error. The
6023 incident is still logged to the main and panic logs.
6025 74. Buglet left over from Exim 3 conversion. The message "too many messages
6026 in one connection" was written to the rejectlog but not the mainlog, except
6027 when address rewriting (yes!) was being logged.
6029 75. Added write_rejectlog option.
6031 76. When a system filter was run not as root (that is, when system_filter_user
6032 was set), the values of the $n variables were not being returned to the
6033 main process; thus, they were not subsequently available in the $sn
6036 77. Added +return_path_on_delivery log selector.
6038 78. A connection timeout was being treated differently from recipients deferred
6039 when testing hosts_max_try with a message that was older than the host's
6040 retry timeout. (The host should not be counted, thus allowing all hosts to
6041 be tried at least once before bouncing.) This may have been the cause of an
6042 occasionally reported bug whereby a message would remain on the queue
6043 longer than the retry timeout, but would be bounced if a delivery was
6044 forced. I say "may" because I never totally pinned down the problem;
6045 setting up timeout/retry tests is difficult. See also the next item.
6047 79. The ultimate address timeout was not being applied to errors that involved
6048 a combination of host plus message (for example, a timeout on a MAIL
6049 command). When an address resolved to a number of possible hosts, and they
6050 were not all tried for each delivery (e.g. because of hosts_max_try), a
6051 message could remain on the queue longer than the retry timeout.
6053 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6056 81. Fixed an obscure SMTP outgoing bug which required at least the following
6057 conditions: (a) there was another message waiting for the same server;
6058 (b) the server returned 5xx to all RCPT commands in the first message so
6059 that the message was not completed; (c) the server dropped the connection
6060 or gave a negative response to the RSET that Exim sends to abort the
6061 transaction. The observed case was a dropped connection after DATA that had
6062 been sent in pipelining mode. That is, the server had advertised PIPELINING
6063 but was not implementing it correctly. The effect of the bug was incorrect
6064 behaviour, such as trying another host, and this could lead to a crash.
6070 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6071 and daemon.c were passed as pointers to ints; they should have been
6072 pointers to socklen_t variables (which are typically unsigned ints).
6074 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6077 3. Fixed a really odd bug that affected only the testing scheme; patching a
6078 certain fixed string in the binary changed the value of another string that
6079 happened to be identical to the end of the original first string.
6081 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6082 name", it returns that address as the IP address. On some operating
6083 systems (e.g. Solaris), it also passes back the IP address string as the
6084 "host name". However, on others (e.g. Linux), it passes back an empty
6085 string. Exim wasn't checking for this, and was changing the host name to an
6086 empty string, assuming it had been canonicalized.
6088 5. Although rare, it is permitted to have more than one PTR record for a given
6089 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6090 all the names associated with an address, because they do in Solaris.
6091 However, it seems that they do not in Linux for data that comes from the
6092 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6093 I found this out when I moved to a new Linux workstation and tried to run
6094 the Exim test suite.
6096 To get round this problem I have changed the code so that it now does its
6097 own call to the DNS to look up PTR records when searching for a host name.
6098 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6099 addresses that are only in /etc/hosts are still found.
6101 This behaviour is, however, controlled by an option called host_lookup_
6102 order, which defaults to "bydns:byaddr". If people want to use the other
6103 order, or indeed, just use one or the other means of lookup, they can
6104 specify it in this variable.
6106 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6107 some operating systems, this comes back from gethostbyaddr() as an empty
6108 string, and this is what Exim used to test for. However, it seems that in
6109 other systems, "." is yielded. Exim now tests for this case too.
6111 7. The values of check_spool_space and check_log_space are now held internally
6112 as a number of kilobytes instead of an absolute number of bytes. If a
6113 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6114 kilobyte. This means that much larger values can be stored.
6116 8. Exim monitor: an attempt to get the action menu when not actually pointing
6117 at a message produces an empty menu entitled "No message selected". This
6118 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6119 no entries in it ("Shell widget menu has zero width and/or height"). So I
6120 have added a single, blank menu entry in this case.
6122 9. Added ${quote_local_part.
6124 10. MIME decoding is now applied to the contents of Subject: header lines when
6127 11. Now that a reference to $sender_host_address automatically causes a reverse
6128 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6129 host lookup before query-style lookups in lists that might use this
6130 variable. This has therefore been abolished, and the "net-" prefix is no
6131 longer necessary for query-style lookups.
6133 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6134 have been a typo for LFLAGS, so it has been changed.
6136 13. The install script calls Exim with "-C /dev/null" in order to find the
6137 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6138 to be output. However, since Exim outputs its version number before the
6139 error, it didn't break the script. It just looked ugly. I fixed this by
6140 always allowing "-C /dev/null" if the caller is root.
6142 14. Ignore overlarge ACL variable number when reading spool file - insurance
6143 against a later release with more variables having written the file.
6145 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6146 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6147 this, as well as the form without the "IPv6" on the front.
6149 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6150 OS/Makefile-Darwin file.
6152 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6153 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6155 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6156 variable. However, some OS use ints for inodes. I've added cast to long int
6157 to get rid of the compiler warning.
6159 19. I had forgotten to lock out "/../" in configuration file names when
6160 ALT_CONFIG_PREFIX was set.
6162 20. Routers used for verification do not need to specify transports. However,
6163 if such a router generated a host list, and callout was configured, Exim
6164 crashed, because it could not find a port number from the (non-existent)
6165 transport. It now assumes port 25 in this circumstance.
6167 21. Added the -t option to exigrep.
6169 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6170 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6171 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6172 them is set, LOOKUP_LSEARCH is forced.
6174 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6176 24. Added sender and host information to the "rejected by local_scan()" log
6177 line; previously there was no indication of these.
6179 25. Added .include_if_exists.
6181 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6182 Linux. It turns out that not all file systems support this. Apparently some
6183 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6184 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6185 not supported on the file descriptor, is now ignored when Exim is trying to
6186 sync a directory. This applies only to Linux.
6188 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6190 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6191 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6194 29. Added support for maildirsize files from supplied patch (modified a bit).
6196 30. The use of :fail: followed by an empty string could lead Exim to respond to
6197 sender verification failures with (e.g.):
6199 550 Verification failed for <xxx>
6200 550 Sender verify failed
6202 where the first response line was missing the '-' that indicates it is not
6203 the final line of the response.
6205 31. The loop for finding the name of the user that called Exim had a hardwired
6206 limit of 10; it now uses the value of finduser_retries, which is used for
6207 all other user lookups.
6209 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6212 33. Exim was neglecting to zero errno before one call of strtol() when
6213 expanding a string and expecting an integer value. On some systems this
6214 resulted in spurious "integer overflow" errors. Also, it was casting the
6215 result into an int without checking.
6217 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6218 did not work. The code looks as if it has *never* worked, though it appears
6219 to have been documented since at least release 1.62. I have made it work.
6221 35. The "timeout_DNS" error in retry rules, also documented since at least
6222 1.62, also never worked. As it isn't clear exactly what this means, and
6223 clearly it isn't a major issue, I have abolished the feature by treating it
6224 as "timeout", and writing a warning to the main and panic logs.
6226 36. The display of retry rules for -brt wasn't always showing the error code
6229 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6230 timeout_connect_A, timeout_connect_MX.
6232 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6233 to the empty sender.
6235 39. The daemon was not analysing the content of -oX till after it had closed
6236 stderr and disconnected from the controlling terminal. This meant that any
6237 syntax errors were only noted on the panic log, and the return code from
6238 the command was 0. By re-arranging the code a little, I've made the
6239 decoding happen first, so such errors now appear on stderr, and the return
6240 code is 1. However, the actual setting up of the sockets still happens in
6241 the disconnected process, so errors there are still only recorded on the
6244 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6245 connections (as happens on some IP stacks) was logged at start up time as
6246 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6247 it from "IPv6 and IPv4", which means that two separate sockets are being
6250 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6251 says whether AF_INET or AF_INET6 was passed as an argument.
6253 42. Exiwhat output was messed up when time zones were included in log
6256 43. Exiwhat now gives more information about the daemon's listening ports,
6257 and whether -tls-on-connect was used.
6259 44. The "port" option of the smtp transport is now expanded.
6261 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6262 silently ignored. Now an error message is written to the main and panic
6265 46. There's a new ACL modifier called "logwrite" which writes to a log file
6266 as soon as it is encountered.
6268 47. Added $local_user_uid and $local_user_gid at routing time.
6270 48. Exim crashed when trying to verify a sender address that was being
6273 49. Exim was recognizing only a space character after ".include". It now also
6274 recognizes a tab character.
6276 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6277 extracting the relevant information from the specification. The man page no
6278 longer contains scrambled data for the -d option, and I've added a section
6279 at the front about calling Exim under different names.
6281 51. Added "extra_headers" argument to the "mail" command in filter files.
6283 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6286 53. Installed eximstats 1.29.
6288 54. Added transport_filter_timeout as a generic transport option.
6290 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6291 Cc: header lines. This was required by RFC 822, but it not required by RFC
6294 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6295 incoming messages that did not have them. Now it does so only if the
6296 message originates locally, that is, if there is no associated remote host
6297 address. When Resent- header lines are present, this applies to the Resent-
6298 lines rather than the non-Resent- lines.
6300 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6301 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6303 58. Messages for configuration errors now include the name of the main
6304 configuration file - useful now that there may be more than one file in a
6305 list (.included file names were always shown).
6307 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6308 for those rare installations that do not start the daemon as root or run it
6309 setuid root. I've cut out the call to initgroups() if the daemon is not
6312 60. The Exim user and group can now be bound into the binary as text strings
6313 that are looked up at the start of Exim's processing.
6315 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6317 62. Added $mailstore_basename variable.
6319 63. Installed patch to sieve.c from Michael Haardt.
6321 64. When Exim failed to open the panic log after failing to open the main log,
6322 the original message it was trying to log was written to stderr and debug
6323 output, but if they were not available (the usual case in production), it
6324 was lost. Now it is written to syslog before the two lines that record the
6325 failures to open the logs.
6327 65. Users' Exim filters run in subprocesses under the user's uid. It is
6328 possible for a "deliver" command or an alias in a "personal" command to
6329 provoke an address rewrite. If logging of address rewriting is configured,
6330 this fails because the process is not running as root or exim. There may be
6331 a better way of dealing with this, but for the moment (because 4.30 needs
6332 to be released), I have disabled address rewrite logging when running a
6333 filter in a non-root, non-exim process.
6339 1. The buildconfig auxiliary program wasn't quoting the value set for
6340 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6341 not defined. This bug was masked in 4.22 by the effect that was fixed in
6344 2. Some messages that were rejected after a message id was allocated were
6345 shown as "incomplete" by exigrep. It no longer does this for messages that
6346 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6348 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6349 have allow_domain_literals set, the ID did not get logged in the <= line.
6350 Domain literals are now always recognized in Message-ID: header lines.
6352 4. The first argument for a ${extract expansion item is the key name or field
6353 number. Leading and trailing spaces in this item were not being ignored,
6354 causing some misleading effects.
6356 5. When deliver_drop_privilege was set, single queue runner processes started
6357 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6358 same command in the process it spins off) were not dropping privilege.
6360 6. When the daemon running as "exim" started a queue runner, it always
6361 re-executed Exim in the spun-off process. This is a waste of effort when
6362 deliver_drop_privilege is set. The new process now just calls the
6363 queue-runner function directly.
6369 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6372 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6373 ignored. Though the use of -oP was forcing the writing of a pid file, it
6374 was always written to the default place.
6376 3. If the message "no IP address found for host xxxx" is generated during
6377 incoming verification, it is now followed by identification of the incoming
6378 connection (so you can more easily find what provoked it).
6380 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6382 5. Added some features to "harden" Exim a bit more against certain attacks:
6384 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6385 be put in Local/Makefile. This is like the never_users runtime option,
6386 but it cannot be overridden. The default setting is "root".
6388 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6389 prefix string with which any file named in a -C command line option
6392 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6393 is retained for -C and -D only if the caller of Exim is root. Without
6394 it, the exim user may also use -C and -D and retain privilege.
6396 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6397 command line option is disabled.
6399 6. Macro names set by the -D option must start with an upper case letter, just
6400 like macro names defined in the configuration file.
6402 7. Added "dereference=" facility to LDAP.
6404 8. Two instances of the typo "uknown" in the source files are fixed.
6406 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6407 the Configure-Makefile script screwed up while processing it.
6409 10. Incorporated PCRE 4.4.
6411 11. The SMTP synchronization check was not operating right at the start of an
6412 SMTP session. For example, it could not catch a HELO sent before the client
6413 waited for the greeting. There is now a check for outstanding input at the
6414 point when the greeting is written. Because of the duplex, asynchronous
6415 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6416 way, but not yet received, when the check is performed.
6418 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6419 on TCP/IP sockets, because this apparently causes some broken clients to
6422 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6423 unchanged) from the Cygwin maintainer.
6425 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6426 was supported instead of testing for mbx. Effectively a typo.
6428 15. The spa authenticator server code was not checking that the input it
6429 received was valid base64.
6431 16. The debug output line for the "set" modifier in ACLs was not showing the
6432 name of the variable that was being set.
6434 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6436 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6437 lookup on its own. Something else was needed to trigger the lookup. For
6438 example, a match in host_lookup or the need for a host name in a host list.
6439 Now, if $sender_host_name is referenced and the host name has not yet been
6440 looked up, a lookup is performed. If the lookup fails, the variable remains
6441 empty, and $host_lookup_failed is set to "1".
6443 19. Added "eqi" as a case-independent comparison operator.
6445 20. The saslauthd authentication condition could segfault if neither service
6446 nor realm was specified.
6448 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6449 error message that was logged was misleading, and incoming SMTP
6450 connections were dropped. The message is now more accurate, and temporary
6451 errors are given to SMTP connections.
6453 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6454 (except RSET) with a 421 error, until QUIT is received. However, it was
6455 failing to send a response to QUIT.
6457 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6458 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6459 ACL is now run after the helo_try_verify_hosts code.
6461 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6462 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6463 case-independent, so other case variants are also recognized). Apparently
6464 some systems use these upper case variants.
6466 25. If more than two messages were waiting for the same host, and a transport
6467 filter was specified for the transport, Exim sent two messages over the
6468 same TCP/IP connection, and then failed with "socket operation on non-
6469 socket" when it tried to send the third.
6471 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6473 27. The extern definition of crypt16() in expand.c was not being excluded when
6474 the OS had its own crypt16() function.
6476 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6477 as a preferred synonym for return_size_limit, both as an option and as an
6480 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6482 30. Changed the default configuration ACL to relax the local part checking rule
6483 for addresses that are not in any local domains. For these addresses,
6484 slashes and pipe symbols are allowed within local parts, but the sequence
6485 /../ is explicitly forbidden.
6487 31. SPA server authentication was not clearing the challenge buffer before
6490 32. log_message in a "warn" ACL statement was writing to the reject log as
6491 well as to the main log, which contradicts the documentation and doesn't
6492 seem right (because no rejection is happening). So I have stopped it.
6494 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6495 However, I am unable to do any testing of this.
6497 34. Fixed an infelicity in the appendfile transport. When checking directories
6498 for a mailbox, to see if any needed to be created, it was accidentally
6499 using path names with one or more superfluous leading slashes; tracing
6500 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6502 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6503 contents are added to the log line that is written for every discarded
6504 recipient. (Previously a log_message setting was ignored.)
6506 36. The ${quote: operator now quotes the string if it is empty.
6508 37. The install script runs exim in order to find its version number. If for
6509 some reason other than non-existence or emptiness, which it checks, it
6510 could not run './exim', it was installing it with an empty version number,
6511 i.e. as "exim-". This error state is now caught, and the installation is
6514 38. An argument was missing from the function that creates an error message
6515 when Exim fails to connect to the socket for saslauthd authentication.
6516 This could cause Exim to crash, or give a corrupted message.
6518 39. Added isip, isip4, and isip6 to ${if conditions.
6520 40. The ACL variables $acl_xx are now saved with the message, and can be
6521 accessed later in routers, transports, and filters.
6523 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6524 strings in the file are not string-expanded.
6526 42. If a MAIL command specified a SIZE value that was too large to fit into an
6527 int variable, the check against message_size_limit failed. Such values are
6528 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6529 day this will have to be increased, but I don't think I want to be around
6530 when emails are that large.
6537 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6538 iconv() is not standard in FreeBSD.
6540 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6541 IPv6 enabled. The observed symptom was a segmentation fault on return from
6542 the function os_common_find_running_interfaces() in src/os.c.
6544 3. In the check_special_case() function in daemon.c I had used "errno" as an
6545 argument name, which causes warnings on some systems. This was basically a
6546 typo, since it was named "eno" in the comments!
6548 4. The code that waits for the clock to tick (at a resolution of some fraction
6549 of a second) so as to ensure message-id uniqueness was always waiting for
6550 at least one whole tick, when it could have waited for less. [This is
6551 almost certainly not relevant at current processor speeds, where it is
6552 unlikely to ever wait at all. But we try to future-proof.]
6554 5. The function that sleeps for a time interval that includes fractions of a
6555 second contained a race. It did not block SIGALRM between setting the
6556 timer, and suspending (a couple of lines later). If the interval was short
6557 and the sigsuspend() was delayed until after it had expired, the suspension
6558 never ended. On busy systems this could lead to processes getting stuck for
6561 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6562 process, before it forks any delivery processes. The open lookup caching
6563 mechanism meant that the open file or database connection was passed into
6564 the delivery process. The problem was that delivery processes always tidy
6565 up cached lookup data. This could cause a problem for the next delivery
6566 process started by the queue runner, because the external queue runner
6567 process does not know about the closure. So the next delivery process
6568 still has data in the lookup cache. In the case of a file lookup, there was
6569 no problem because closing a file descriptor in a subprocess doesn't affect
6570 the parent. However, if the lookup was caching a connection to a database,
6571 the connection was closed, and the second delivery process was likely to
6572 see errors such as "PGSQL: query failed: server closed the connection
6573 unexpectedly". The problem has been fixed by closing all cached lookups
6574 in a queue runner before running a delivery process.
6576 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6577 seem to have the "const" qualifier which it has on other OS. I've
6580 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6581 *of the same type* (client or server) with the same public name that an
6582 error should be diagnosed.
6584 9. When Exim looked up a host name for an IP address, but failed to find the
6585 original IP address when looking up the host name (a safety check), it
6586 output the message "<ip address> does not match any IP for NULL", which was
6587 confusing, to say the least. The bug was that the host name should have
6588 appeared instead of "NULL".
6590 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6591 user that is built into the binary, and the -C or -D options is used, root
6592 privilege is dropped before the configuration file is read. In addition,
6593 logging is switched to stderr instead of the normal log files. If the
6594 configuration then re-defines the Exim user, the unprivileged environment
6595 is probably not what is expected, so Exim logs a panic warning message (but
6598 However, if deliver_drop_privilege is set, the unprivileged state may well
6599 be exactly what is intended, so the warning has been cut out in that case,
6600 and Exim is allowed to try to write to its normal log files.
6606 1. smtp_return_error_details was not giving details for temporary sender
6607 or receiver verification errors.
6609 2. Diagnose a configuration error if two authenticators have the same public
6612 3. Exim used not to create the message log file for a message until the first
6613 delivery attempt. This could be confusing when incoming messages were held
6614 for policy or load reasons. The message log file is now created at the time
6615 the message is received, and an initial "Received" line is written to it.
6617 4. The automatically generated man page for command line options had a minor
6618 bug that caused no ill effects; however, a more serious problem was that
6619 the procedure for building the man page automatically didn't always
6620 operate. Consequently, release 4.20 contains an out-of-date version. This
6621 shouldn't happen again.
6623 5. When building Exim with embedded Perl support, the script that builds the
6624 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6625 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6627 6. The freeze_tell option was not being used for messages that were frozen on
6628 arrival, either by an ACL or by local_scan().
6630 7. Added the smtp_incomplete_transaction log selector.
6632 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6633 was accepting AUTH without a new EHLO.
6635 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6636 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6639 10. Logging of TCP/IP connections (when configured) now happens in the main
6640 daemon process instead of the child process, so that the TCP/IP connection
6641 count is more accurate (but it can never be perfect).
6643 11. The use of "drop" in a nested ACL was not being handled correctly in the
6644 outer ACL. Now, if condition failure induced by the nested "drop" causes
6645 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6646 or "require"), the connection is dropped.
6648 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6649 that yield "discard" can now be used with an "accept" or a "discard" verb,
6650 but an error is generated for any others (because I can't see a useful way
6651 to define what should happen).
6653 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6654 are now processed in the same way as lines in the Exim configuration file.
6655 In particular, continuation lines are supported.
6657 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6659 15. Added -ti meaning -t -i.
6661 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6662 domains, and warn by logging if others are found.
6664 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6665 returned from GIFCONF when Exim is trying to find the list of interfaces on
6666 a host. The code in os.c has been modified to copy each ifreq to an aligned
6667 structure in all cases.
6669 Also, in some cases, the returned ifreq's were being copied to a 'struct
6670 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6671 means the last couple of bytes of an IPv6 address could be chopped if the
6672 ifreq contained only a normal sockaddr (14 bytes storage).
6674 18. Named domain lists were not supported in the hosts_treat_as_local option.
6675 An entry such as +xxxx was not recognized, and was treated as a literal
6678 19. Ensure that header lines added by a DATA ACL are included in the reject log
6679 if the ACL subsequently rejects the message.
6681 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6682 MD5 (which is deprecated).
6684 21. When testing a filter file using -bf, Exim was writing a message when it
6685 took the sender from a "From " line in the message, but it was not doing so
6686 when it took $return_path from a Return-Path: header line. It now does.
6688 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6689 with a valid header line field name (a series of printing characters
6690 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6692 23. Changed "disc" in the source to "disk" to conform to the documentation and
6693 the book and for uniformity.
6695 24. Ignore Sendmail's -Ooption=value command line item.
6697 25. When execve() failed while trying to run a command in a pipe transport,
6698 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6699 could be confused with a return value of 69 from the command itself. This
6700 has been changed to 127, the value the shell returns if it is asked to run
6701 a non-existent command. The wording for the related log line suggests a
6702 non-existent command as the problem.
6704 26. If received_header_text expands to an empty string, do not add a Received:
6705 header line to the message. (Well, it adds a token one on the spool, but
6706 marks it "old" so that it doesn't get used or transmitted.)
6708 27. Installed eximstats 1.28 (addition of -nt option).
6710 28. There was no check for failure on the call to getsockname() in the daemon
6711 code. This can fail if there is a shortage of resources on the system, with
6712 ENOMEM, for example. A temporary error is now given on failure.
6714 29. Contrary to the C standard, it seems that in some environments, the
6715 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6716 program. Exim now does this explicitly; it affects the formatting of
6717 timestamps using strftime().
6719 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6720 complaints. I've now initialized all the variables, to avoid this.
6722 32. Header lines added by a system filter were not being "seen" during
6723 transport-time rewrites.
6725 33. The info_callback() function passed to OpenSSL is set up with type void
6726 (*)(SSL *, int, int), as described somewhere. However, when calling the
6727 function (actually a macro) that sets it up, the type void(*)() is
6728 expected. I've put in a cast to prevent warnings from picky compilers.
6730 34. If a DNS black list lookup found a CNAME record, but there were no A
6731 records associated with the domain it pointed at, Exim crashed.
6733 35. If a DNS black list lookup returned more than one A record, Exim ignored
6734 all but the first. It now scans all returned addresses if a particular IP
6735 value is being sought. In this situation, the contents of the
6736 $dnslist_value variable are a list of all the addresses, separated by a
6739 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6740 to accept a host name and all its aliases if the forward lookup for any of
6741 them yielded the IP address of the incoming connection. Now it accepts only
6742 those names whose forward lookup yields the correct IP address. Any other
6743 names are discarded. This closes a loophole whereby a rogue DNS
6744 administrator could create reverse DNS records to break through a
6745 wildcarded host restriction in an ACL.
6747 37. If a user filter or a system filter that ran in a subprocess used any of
6748 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6749 the wrong values were passed to the pipe command ($thisaddress had the
6750 value of $0, $0 had the value of $1, etc). This bug was introduced by
6751 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6753 38. Improved the line breaking for long SMTP error messages from ACLs.
6754 Previously, if there was no break point between 40 and 75 characters, Exim
6755 left the rest of the message alone. Two changes have been made: (a) I've
6756 reduced the minimum length to 35 characters; (b) if it can't find a break
6757 point between 35 and 75 characters, it looks ahead and uses the first one
6758 that it finds. This may give the occasional overlong line, but at least the
6759 remaining text gets split now.
6761 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6762 file descriptors might be low, and that setting 1000 would always raise it.
6763 It turns out that in some environments, the limit is already over 1000 and
6764 that lowering it causes trouble. So now Exim takes care not to decrease it.
6766 40. When delivering a message, the value of $return_path is set to $sender_
6767 address at the start of routing (routers may change the value). By an
6768 oversight, this default was not being set up when an address was tested by
6769 -bt or -bv, which affected the outcome if any router or filter referred to
6772 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6773 log only when "message" or "log_message" are set. However, if one of the
6774 conditions was an address verification, or a call to a nested ACL, the
6775 messages generated by the underlying test were being passed through. This
6776 no longer happens. The underlying message is available in $acl_verify_
6777 message for both "message" and "log_message" expansions, so it can be
6778 passed through if needed.
6780 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6781 new expansion $bh_ to give the encoded byte string without charset
6782 translation. Translation happens only if iconv() is available; HAVE_ICONV
6783 indicates this at build time. HEADERS_CHARSET gives the charset to
6784 translate to; headers_charset can change it in the configuration, and
6785 "headers charset" can change it in an individual filter file.
6787 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6788 that creates RFC 2047 encoded "words" labels them as that charset instead
6789 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6790 expansion operator; (ii) when Exim creates a From: line for a local
6791 message; (iii) when a header line is rewritten to include a "phrase" part.
6793 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6794 buggy, causing it to skip the first lines of messages whose message ID
6795 ended in 'D'. This would not have bitten before Exim release 4.14, because
6796 message IDs were unlikely to end in 'D' before then. The effect was to have
6797 incorrect size information for certain domains.
6799 45. #include "config.h" was missing at the start of the crypt16.c module. This
6800 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6803 46. If there was a timeout during a "random" callout check, Exim treated it as
6804 a failure of the random address, and carried on sending RSET and the real
6805 address. If the delay was just some slowness somewhere, the response to the
6806 original RCPT would be taken as a response to RSET and so on, causing
6807 mayhem of various kinds.
6809 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6810 when I implemented it. It didn't allow for the fact that some option values
6811 may legitimately be negative (e.g. size_addition), and it didn't even do
6812 the right test for positive values.
6814 48. Domain names in DNS records are case-independent. Exim always looks them up
6815 in lower case. Some resolvers return domain names in exactly the case they
6816 appear in the zone file, that is, they may contain uppercase letters. Not
6817 all resolvers do this - some return always lower case. Exim was treating a
6818 change of case by a resolver as a change of domain, similar to a widening
6819 of a domain abbreviation. This triggered its re-routing code and so it was
6820 trying to route what was effectively the same domain again. This normally
6821 caused routing to fail (because the router wouldn't handle the domain
6822 twice). Now Exim checks for this case specially, and just changes the
6823 casing of the domain that it ultimately uses when it transmits the message
6826 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6829 50. If a filter generated a file delivery with a non-absolute name (possible if
6830 no home directory exists for the router), the forbid_file option was not
6833 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6834 the existing equality matching.
6836 52. Exim was using ints instead of ino_t variables in some places where it was
6837 dealing with inode numbers.
6839 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6840 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6841 called TMPDIR, and if it finds it is different, it changes its value.
6843 54. The smtp_printf() function is now made available to local_scan() so
6844 additional output lines can be written before returning. There is also an
6845 smtp_fflush() function to enable the detection of a dropped connection.
6846 The variables smtp_input and smtp_batched_input are exported to
6849 55. Changed the default runtime configuration: the message "Unknown user"
6850 has been removed from the ACL, and instead placed on the localuser router,
6851 using the cannot_route_message feature. This means that any verification
6852 failures that generate their own messages won't get overridden. Similarly,
6853 the "Unrouteable address" message that was in the ACL for unverifiable
6854 relay addresses has also been removed.
6856 56. Added hosts_avoid_esmtp to the smtp transport.
6858 57. The exicyclog script was not checking for the esoteric option
6859 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6860 will work only if exicyclog is run under the appropriate euid.
6862 58. Following a discussion on the list, the rules by which Exim recognises line
6863 endings on incoming messages have been changed. The -dropcr and drop_cr
6864 options are now no-ops, retained only for backwards compatibility. The
6865 following line terminators are recognized: LF CRLF CR. However, special
6866 processing applies to CR:
6868 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6869 nor a local message in the state where . is a terminator.
6871 (ii) If a bare CR is encountered in a header line, an extra space is added
6872 after the line terminator so as not to end the header. The reasoning
6873 behind this is that bare CRs in header lines are most likely either
6874 to be mistakes, or people trying to play silly games.
6876 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6877 was being incorrectly given as 18 bytes larger than it should have been.
6878 This is a VOB (very old bug).
6880 60. This may never have affected anything current, but just in case it has:
6881 When the local host is found other than at the start of a list of hosts,
6882 the local host, those with the same MX, and any that follow, are discarded.
6883 When the list in question was part of a longer list of hosts, the following
6884 hosts (not currently being processed) were also being discarded. This no
6885 longer happens. I'm not sure if this situation could ever has previously
6888 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6891 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6892 header. This is something that is recommended in a new Internet Draft, and
6893 is something that is documented as being done by Sendmail. There are two
6894 possible values. For messages generated by the autoreply transport, Exim
6897 Auto-Submitted: auto-replied
6899 whereas for all other generated messages (e.g. bounces) it adds
6901 Auto-Submitted: auto-generated
6903 63. The "personal" condition in filters now includes a test for the
6904 Auto-Submitted: header. If it contains the string "auto-" the message it
6905 not considered personal.
6907 64. Added rcpt_include_affixes as a generic transport option.
6909 65. Added queue_only_override (default true).
6911 66. Added the syslog_duplication option.
6913 67. If what should have been the first header line of a message consisted of
6914 a space followed by a colon, Exim was mis-interpreting it as a header line.
6915 It isn't of course - it is syntactically invalid and should therefore be
6916 treated as the start of the message body. The misbehaviour could have
6917 caused a number of strange effects, including loss of data in subsequent
6918 header lines, and spool format errors.
6920 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6921 client host had authenticated. This control can now be exercised by an ACL
6922 for more flexibility.
6924 69. By default, callouts do not happen when testing with -bh. There is now a
6925 variant, -bhc, which does actually run the callout code, including
6926 consulting and updating the callout cache.
6928 70. Added support for saslauthd authentication, courtesy of Alexander
6931 71. If statvfs() failed on the spool or log directories while checking their
6932 size for availability, Exim confusingly gave the error "space shortage".
6933 Furthermore, in debugging mode it crashed with a floating point exception.
6934 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6935 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6936 serious problem, Exim now writes to the main and panic logs when this
6937 happens, with details of the failure. It then refuses to accept the
6938 incoming message, giving the message "spool directory problem" or "log
6939 directory problem" with a 421 code for SMTP messages.
6941 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6942 0, 1, and 2 exist, because some OS complain for execs without them (see
6943 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6944 descriptors. However, the code omitted to check that the open succeeded,
6945 causing mysterious errors if for some reason the permissions on /dev/null
6946 got screwed. Now Exim writes a message to the main and panic logs, and
6947 bombs out if it can't open /dev/null.
6949 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6950 interact so that it is all more flexible. It is supposed to remain
6951 backwards compatible. Also added extra_local_interfaces.
6953 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6954 to bomb out with an assertion failure - to the client this appears as a
6955 connection drop. This problem occurs in the part of the code that was taken
6956 from the Samba project. Fortunately, the assertion is in a very simple
6957 function, so I have fixed this by reproducing the function inline in the
6958 one place where it is called, and arranging for authentication to fail
6959 instead of killing the process with assert().
6961 75. The SPA client code was not working when the server requested OEM rather
6962 than Unicode encoding.
6964 76. Added code to make require_files with a specific uid setting more usable in
6965 the case where statting the file as root fails - usually a non-root-mounted
6966 NFS file system. When this happens and the failure is EACCES, Exim now
6967 forks a subprocess and does the per-uid checking as the relevant uid.
6969 77. Added process_log_path.
6971 78. If log_file_path was not explicitly set, a setting of check_log_space or
6972 check_log_inodes was ignored.
6974 79. If a space check for the spool or log partitions fails, the incident is now
6975 logged. Of course, in the latter case the data may get lost...
6977 80. Added the %p formatting code to string_format() so that it can be used to
6978 print addresses in debug_print(). Adjusted all the address printing in the
6979 debugging in store.c to use %p rather than %d.
6981 81. There was a concern that a line of code in smtp_in.c could overflow a
6982 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6983 initially expressed, the concern was not well-founded, because trailing
6984 spaces are removed early. However, if the trailing spaces were followed by
6985 a NULL, they did not get removed, so the overflow was possible. Two fixes
6988 (a) I re-wrote the offending code in a cleaner fashion.
6989 (b) If an incoming SMTP command contains a NULL character, it is rejected
6992 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6993 runs initgroups(), so that if the Exim user is in any additional groups,
6994 they will be used during message reception.
7000 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob