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
42 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
43 rows affected is given instead).
45 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
46 SMTP connection. Previously, when one had more receipients than the
47 first, an abortive onward connection was made. Move to full support for
48 multiple onward connections in sequence, handling cutthrough connection
49 for all multi-message initiating connections.
51 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
52 routers. Previously, a multi-recipient message would fail to match the
53 onward-connection opened for the first recipient, and cause its closure.
55 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
56 a timeout on read on a GnuTLS initiating connection, resulting in the
57 initiating connection being dropped. This mattered most when the callout
58 was marked defer_ok. Fix to keep the two timeout-detection methods
61 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
62 metadata, resulting in a crash in free().
64 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
65 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
66 Broken also in d185889f4, with init system revamp.
72 JH/01 Rework error string handling in TLS interface so that the caller in
73 more cases is responsible for logging. This permits library-sourced
74 string to be attached to addresses during delivery, and collapses
75 pairs of long lines into single ones.
77 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
78 during configuration. Wildcards are allowed and expanded.
80 JH/02 Rework error string handling in DKIM to pass more info back to callers.
81 This permits better logging.
83 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
84 do not close it down and have the child transport start it up again on
85 the passed-on TCP connection. Instead, proxy the child (and any
86 subsequent ones) for TLS via a unix-domain socket channel. Logging is
87 affected: the continued delivery log lines do not have any DNSSEC, TLS
88 Certificate or OCSP information. TLS cipher information is still logged.
90 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
91 identical IP addresses on different listening ports. Will also affect
94 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
95 add noisy ifdef guards to special-case this sillyness.
96 Patch from Bernd Kuhls.
98 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
99 than 255 are no longer allowed.
101 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
102 Disable the session-cache too, which might reduce our load. Since we
103 currrectly use a new context for every connection, both as server and
104 client, there is no benefit for these.
105 GnuTLS appears to not support tickets server-side by default (we don't
106 call gnutls_session_ticket_enable_server()) but client side is enabled
107 by default on recent versions (3.1.3 +) unless the PFS priority string
110 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
111 <https://reproducible-builds.org/specs/source-date-epoch/>.
113 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
114 the check for any unsuccessful recipients did not notice the limit, and
115 erroneously found still-pending ones.
117 JH/08 Pipeline CHUNKING command and data together, on kernels that support
118 MSG_MORE. Only in-clear (not on TLS connections).
120 JH/09 Avoid using a temporary file during transport using dkim. Unless a
121 transport-filter is involved we can buffer the headers in memory for
122 creating the signature, and read the spool data file once for the
123 signature and again for transmission.
125 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
126 4.77 as the kernel support then wasn't solid, having issues in 64bit
127 mode. Now, it's been long enough. Add support for FreeBSD also.
129 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
130 case where the routing stage had gathered several addresses to send to
131 a host before calling the transport for the first, we previously failed
132 to close down TLS in the old transport process before passing the TCP
133 connection to the new process. The new one sent a STARTTLS command
134 which naturally failed, giving a failed delivery and bloating the retry
135 database. Investigation and fix prototype from Wolfgang Breyha.
137 JH/12 Fix check on SMTP command input synchronisation. Previously there were
138 false-negatives in the check that the sender had not preempted a response
139 or prompt from Exim (running as a server), due to that code's lack of
140 awareness of the SMTP input buffering.
142 PP/04 Add commandline_checks_require_admin option.
143 Exim drops privileges sanely, various checks such as -be aren't a
144 security problem, as long as you trust local users with access to their
145 own account. When invoked by services which pass untrusted data to
146 Exim, this might be an issue. Set this option in main configuration
147 AND make fixes to the calling application, such as using `--` to stop
150 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
151 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
152 responses to those, into a single TLS record each way (this usually means
153 a single packet). As a side issue, smtp_enforce_sync now works on TLS
156 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
157 affects you only if you're dancing at the edge of the param size limits.
158 If you are, and this message makes sense to you, then: raise the
159 configured limit or use OpenSSL 1.1. Nothing we can do for older
162 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
163 cmdline element to get the documented default one. Previously it was
166 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
167 are unsuable for non-message-specific reasons - eg. connection timeout,
168 banner-time rejection.
170 JH/16 Fix logging of delivery remote port, when specified by router, under
173 PP/06 Repair manualroute's ability to take options in any order, even if one
174 is the name of a transport.
177 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
179 JH/17 Change the list-building routines interface to use the expanding-string
180 triplet model, for better allocation and copying behaviour.
182 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
183 Previously it was constructed the first time a possibly-matching string
184 was met in the configuration file input during startup; now it is done
187 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
188 compatible one, to avoid the (poorly documented) possibility of a config
189 file in the working directory redirecting the DB files, possibly correpting
190 some existing file. CVE-2017-10140 assigned for BDB.
192 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
193 cache-hot. Previously, although the result was properly cached, the
194 initial verify call returned a defer.
196 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
197 the main verify for receipient in uncached-mode.
199 JH/22 Retire historical build files to an "unsupported" subdir. These are
200 defined as "ones for which we have no current evidence of testing".
202 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
203 if present. Previously it was ignored.
205 JH/24 Start using specified-initialisers in C structure init coding. This is
206 a C99 feature (it's 2017, so now considered safe).
208 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
209 if was a fixed-sized field and bitmask ops via macros; it is now more
212 PP/07 GitHub PR 56: Apply MariaDB build fix.
213 Patch provided by Jaroslav Škarvada.
215 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
216 during Coverity cleanups [4.87 JH/47]
217 Diagnosis and fix provided by Michael Fischer v. Mollard.
219 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
220 the right size to place the terminating semicolon on its own folded
221 line, the header hash was calculated to an incorrect value thanks to
222 the (relaxed) space the fold became.
224 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
225 and confused the parent.
227 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
228 which could crash as a result. This could lead to undeliverable messages.
230 JH/28 Logging: "next input sent too soon" now shows where input was truncated
233 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
234 matters on fast-turnover and PID-randomising systems, which were getting
235 out-of-order delivery.
237 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
238 a possibly-overlapping copy. The symptom was that "Remote host closed
239 connection in response to HELO" was logged instead of the actual 4xx
242 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
243 Previously only that bufferd was discarded, resulting in SYMTP command
246 JH/32 DKIM: when a message has multiple signatures matching an identity given
247 in dkim_verify_signers, run the dkim acl once for each. Previously only
248 one run was done. Bug 2189.
250 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
251 "panic the current process" to "deliberately defer". The panic log is
252 still written with the problem list name; the mail and reject logs now
253 get a temp-reject line for the message that was being handled, saying
254 something like "domains check lookup or other defer". The SMTP 451
255 message is still "Temporary local problem".
257 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
258 A crafted sequence of BDAT commands could result in in-use memory beeing
259 freed. CVE-2017-16943.
261 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
262 from SMTP input. Previously it was always done; now only done for DATA
263 and not BDAT commands. CVE-2017-16944.
265 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
266 to the message (such as an overlong header line). Previously this was
267 not done and we did not exit BDAT mode. Followon from the previous item
268 though a different problem.
274 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
275 than -2003 did; needs libidn2 in addition to libidn.
277 JH/02 The path option on a pipe transport is now expanded before use.
279 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
280 Patch provided by "Björn", documentation fix added too.
282 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
283 missing a wire-to-host endian conversion.
285 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
286 close after a BDAT command line could be taken as a following command,
287 giving a synch failure. Fix by only checking for synch immediately
288 before acknowledging the chunk.
290 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
291 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
292 macro. Patches provided by Josh Soref.
294 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
295 Previously we did not; the RFC seems ambiguous and VRFY is not listed
296 by IANA as a service extension. However, John Klensin suggests that we
299 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
300 the dkim code may be unix-mode line endings rather than smtp wire-format
301 CRLF, so prepend a CR to any bare LF.
303 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
304 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
306 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
307 the first were themselves being wrongly included in the feed into dkim
308 processing; with most chunk sizes in use this resulted in an incorrect
309 body hash calculated value.
311 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
312 DKIM signature block, for verification. Although advised against by
313 standards it is specifically not ruled illegal.
315 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
317 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
318 missing a body hash (the bh= tag).
320 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
321 It seems that HAProxy sends the Proxy Protocol information in clear and
322 only then does a TLS startup, so do the same.
324 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
325 TCP connections (such as for Spamd) unless the daemon successfully set
326 Fast Open mode on its listening sockets. This fixes breakage seen on
327 too-old kernels or those not configured for Fast Open, at the cost of
328 requiring both directions being enabled for TFO, and TFO never being used
329 by non-daemon-related Exim processes.
331 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
332 endings, at least on the first header line. Try to canonify any that get
333 past that check, despite the cost.
335 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
336 now limited to an arbitrary five deep, while parsing addresses with the
337 strip_excess_angle_brackets option enabled.
339 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
340 instead leave the unprompted TLS handshake in socket buffer for the
341 TLS library to consume.
343 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
345 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
347 JH/16 Drop variables when they go out of scope. Memory management drops a whole
348 region in one operation, for speed, and this leaves assigned pointers
349 dangling. Add checks run only under the testsuite which checks all
350 variables at a store-reset and panics on a dangling pointer; add code
351 explicitly nulling out all the variables discovered. Fixes one known
352 bug: a transport crash, where a dangling pointer for $sending_ip_address
353 originally assigned in a verify callout, is re-used.
355 PP/06 Drop '.' from @INC in various Perl scripts.
357 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
359 PP/08 Reduce a number of compilation warnings under clang; building with
360 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
361 should be warning-free.
363 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
365 HS/01 Fix portability problems introduced by PP/08 for platforms where
366 realloc(NULL) is not equivalent to malloc() [SunOS et al].
368 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
369 chunk. This allows us to accept broken chunked messages. We need a more
370 general solution here.
372 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
373 already-broken messages in the queue.
375 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
377 JH/19 Fix reference counting bug in routing-generated-address tracking.
383 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
384 supports it and a size is available (ie. the sending peer gave us one).
386 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
387 years of being deprecated). The replacements are the ACLs
388 acl_smtp_mime and acl_not_smtp_mime.
390 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
391 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
392 or even an in-clear connection were permitted. Now, if the host lookup
393 was dnssec and dane was requested then the host is only used if the
394 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
395 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
396 if one fails this test.
397 This means that a poorly-configured remote DNS will make it incommunicado;
398 but it protects against a DNS-interception attack on it.
400 JH/04 Bug 1810: make continued-use of an open smtp transport connection
401 non-noisy when a race steals the message being considered.
403 JH/05 If main configuration option tls_certificate is unset, generate a
404 self-signed certificate for inbound TLS connections.
406 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
407 in rewrites and routers.
409 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
410 and logged a warning sing 4.83; now they are a configuration file error.
412 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
413 (lacking @domain). Apply the same qualification processing as RCPT.
415 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
417 JH/10 Support ${sha256:} applied to a string (as well as the previous
420 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
421 a cutthrough deliver is pending, as we always want to make a connection.
422 This also avoids re-routing the message when later placing the cutthrough
423 connection after a verify cache hit.
424 Do not update it with the verify result either.
426 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
427 when routing results in more than one destination address.
429 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
430 signing (which inhibits the cutthrough capability). Previously only
431 the presence of an option was tested; now an expansion evaluating as
432 empty is permissible (obviously it should depend only on data available
433 when the cutthrough connection is made).
435 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
436 the relevant preceding SMTP command did not note the pipelining mode.
438 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
439 Previously they were not counted.
441 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
442 as one having no matching records. Previously we deferred the message
443 that needed the lookup.
445 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
446 distinguished as "(=".
448 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
449 for missing MX records. Previously it only worked for missing A records.
451 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
453 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
454 after the data-go-ahead and data-ack. Patch from Jason Betts.
456 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
457 even for a "none" policy. Patch from Tony Meyer.
459 JH/22 Fix continued use of a connection for further deliveries. If a port was
460 specified by a router, it must also match for the delivery to be
463 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
464 When one of the recipients of a message was unsuitable for the connection
465 (has no matching addresses), we lost track of needing to mark it
466 deferred. As a result mail would be lost.
468 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
470 JH/25 Decoding ACL controls is now done using a binary search; the source code
471 takes up less space and should be simpler to maintain. Merge the ACL
472 condition decode tables also, with similar effect.
474 JH/26 Fix problem with one_time used on a redirect router which returned the
475 parent address unchanged. A retry would see the parent address marked as
476 delivered, so not attempt the (identical) child. As a result mail would
479 JH/27 Fix a possible security hole, wherein a process operating with the Exim
480 UID can gain a root shell. Credit to http://www.halfdog.net/ for
481 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
484 JH/28 Enable {spool,log} filesystem space and inode checks as default.
485 Main config options check_{log,spool}_{inodes,space} are now
486 100 inodes, 10MB unless set otherwise in the configuration.
488 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
489 Previously it only applied to the main-section connection policy
492 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
494 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
495 by me. Added RFC7919 DH primes as an alternative.
497 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
498 are not in the system include path.
500 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
501 GnuTLS, when a session startup failed (eg because the client disconnected)
502 Exim did stdio operations after fclose. This was exposed by a recent
503 change which nulled out the file handle after the fclose.
505 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
506 signed directly by the cert-signing cert, rather than an intermediate
507 OCSP-signing cert. This is the model used by LetsEncrypt.
509 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
511 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
512 an incoming connection.
514 HS/02 Bug 1802: Do not half-close the connection after sending a request
517 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
518 fallback to "prime256v1".
520 JH/34 SECURITY: Use proper copy of DATA command in error message.
521 Could leak key material. Remotely exploitable. CVE-2016-9963.
527 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
528 and 3.4.4 - once the server is enabled to respond to an OCSP request
529 it does even when not requested, resulting in a stapling non-aware
530 client dropping the TLS connection.
532 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
533 support variable-length bit vectors. No functional change.
535 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
536 The I= interface field on outgoing lines is now after the H= remote
537 host field, same as incoming lines. There is a separate
538 outgoing_interface log selector which allows you to disable the
541 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
542 If not running log_selector +smtp_connection the mainlog would be held
543 open indefinitely after a "too many connections" event, including to a
544 deleted file after a log rotate. Leave the per net connection logging
545 leaving it open for efficiency as that will be quickly detected by the
546 check on the next write.
548 HS/01 Bug 1671: Fix post transport crash.
549 Processing the wait-<transport> messages could crash the delivery
550 process if the message IDs didn't exist for some reason. When
551 using 'split_spool_directory=yes' the construction of the spool
552 file name failed already, exposing the same netto behaviour.
554 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
555 mime_regex ACL conditions.
557 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
558 to DSN fail messages (bounces): remote IP, remote greeting, remote response
559 to HELO, local diagnostic string.
561 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
562 log line to debug. Even when configured with a tls authenticator many
563 client connections are expected to not authenticate in this way, so
564 an authenticate fail is not an error.
566 HS/02 Add the Exim version string to the process info. This way exiwhat
567 gives some more detail about the running daemon.
569 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
570 matter for fast-change records such as DNSBLs.
572 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
573 retry record, even if constant. There may be multiple transports with
574 different interface settings and the retry behaviour needs to be kept
577 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
579 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
581 JH/10 Bug 840: fix log_defer_output option of pipe transport
583 JH/11 Bug 830: use same host for all RCPTS of a message, even under
584 hosts_randomize. This matters a lot when combined with mua_wrapper.
586 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
587 ${quote_pgsql:<string>} operator.
589 JH/13 Bug 1708: avoid misaligned access in cached lookup.
591 JH/14 Change header file name for freeradius-client. Relevant if compiling
592 with Radius support; from the Gentoo tree and checked under Fedora.
594 JH/15 Bug 1712: Introduce $prdr_requested flag variable
596 JH/16 Bug 1714: Permit an empty string as expansion result for transport
597 option transport_filter, meaning no filtering.
599 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
601 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
602 defaults to "*" (all hosts). The variable is now available when not built
603 with TLS, default unset, mainly to enable keeping the testsuite sane.
604 If a server certificate is not supplied (via tls_certificate) an error is
605 logged, and clients will find TLS connections fail on startup. Presumably
606 they will retry in-clear.
607 Packagers of Exim are strongly encouraged to create a server certificate
608 at installation time.
610 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
611 with the $config_file variable.
613 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
614 in transport context, after the attempt, and per-recipient. The latter type
615 is per host attempted. The event data is the error message, and the errno
616 information encodes the lookup type (A vs. MX) used for the (first) host,
617 and the trailing two digits of the smtp 4xx response.
619 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
620 to write to mainlog (or rejectlog, paniclog) in the window between file
621 creation and permissions/ownership being changed. Particularly affects
622 installations where exicyclog is run as root, rather than exim user;
623 result is that the running daemon panics and dies.
625 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
627 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
628 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
629 "pri" and "weight". Note that the previous implicit priority given by the
630 list order is no longer honoured.
632 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
635 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
636 by defining SUPPORT_SOCKS.
638 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
639 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
640 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
641 variables are renamed to proxy_{local,external}_{address,port}.
643 JH/27 Move Internationalisation support from Experimental to mainline, enabled
644 for a build by defining SUPPORT_I18N
646 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
647 of the query string, and make ${quote_redis:} do that quoting.
649 JH/29 Move Events support from Experimental to mainline, enabled by default
650 and removable for a build by defining DISABLE_EVENT.
652 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
654 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
655 cached by the daemon.
657 JH/32 Move Redis support from Experimental to mainline, enabled for a build
658 by defining LOOKUP_REDIS. The libhiredis library is required.
660 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
661 keys are given for lookup.
663 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
664 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
665 only supported when built with TLS support. The PolarSSL SHA routines
666 are still used when the TLS library is too old for convenient support.
668 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
669 openssl_options), for security. OpenSSL forces this from version 1.1.0
670 server-side so match that on older versions.
672 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
673 allocation for $value could be released as the expansion processing
674 concluded, but leaving the global pointer active for it.
676 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
677 and to use the domains and local_parts ACL conditions.
679 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
680 incorrectly not doubled on cutthrough transmission, hence seen as a
681 body-termination at the receiving system - resulting in truncated mails.
682 Commonly the sender saw a TCP-level error, and retransmitted the message
683 via the normal store-and-forward channel. This could result in duplicates
684 received - but deduplicating mailstores were liable to retain only the
685 initial truncated version.
687 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
689 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
691 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
692 we're in there, support oversigning also; bug 1309.
694 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
696 HS/04 Add support for keep_environment and add_environment options.
698 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
699 either intentional arithmetic overflow during PRNG, or testing config-
702 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
703 delivery resulted in actual delivery. Cancel cutthrough before DATA
706 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
707 rejects a recipient: pass the reject to the originator.
709 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
710 Many were false-positives and ignorable, but it's worth fixing the
713 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
714 for the new environment-manipulation done at startup. Move the routines
715 from being local to tls.c to being global via the os.c file.
717 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
718 an extract embedded as result-arg for a map, the first arg for extract
719 is unavailable so we cannot tell if this is a numbered or keyed
720 extraction. Accept either.
726 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
729 JH/02 The smtp transport option "multi_domain" is now expanded.
731 JH/03 The smtp transport now requests PRDR by default, if the server offers
734 JH/04 Certificate name checking on server certificates, when exim is a client,
735 is now done by default. The transport option tls_verify_cert_hostnames
736 can be used to disable this per-host. The build option
737 EXPERIMENTAL_CERTNAMES is withdrawn.
739 JH/05 The value of the tls_verify_certificates smtp transport and main options
740 default to the word "system" to access the system default CA bundle.
741 For GnuTLS, only version 3.0.20 or later.
743 JH/06 Verification of the server certificate for a TLS connection is now tried
744 (but not required) by default. The verification status is now logged by
745 default, for both outbound TLS and client-certificate supplying inbound
748 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
751 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
752 Status Notification (bounce) messages are now MIME format per RFC 3464.
753 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
754 under the control of the dsn_advertise_hosts option, and routers may
755 have a dsn_lasthop option.
757 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
758 default, modifiable by a malware= option. The list separator for
759 the options can now be changed in the usual way. Bug 68.
761 JH/10 The smtp_receive_timeout main option is now expanded before use.
763 JH/11 The incoming_interface log option now also enables logging of the
764 local interface on delivery outgoing connections.
766 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
767 if the interface and destination host and port all match.
769 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
772 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
773 Patch from Andrew Lewis.
775 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
776 now supports optional time-restrictions, weighting, and priority
777 modifiers per server. Patch originally by <rommer@active.by>.
779 JH/16 The spamd_address main option now supports a mixed list of local
780 and remote servers. Remote servers can be IPv6 addresses, and
781 specify a port-range.
783 JH/17 Bug 68: The spamd_address main option now supports an optional
784 timeout value per server.
786 JH/18 Bug 1581: Router and transport options headers_add/remove can
787 now have the list separator specified.
789 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
792 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
795 JH/21 Support for the A6 type of dns record is withdrawn.
797 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
798 rather than the verbs used.
800 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
801 from 255 to 1024 chars.
803 JH/24 Verification callouts now attempt to use TLS by default.
805 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
806 are generic router options now. The defaults didn't change.
808 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
809 Original patch from Alexander Shikoff, worked over by JH.
811 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
812 files if ZCAT_COMMAND is not executable.
814 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
816 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
818 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
819 Normally benign, it bites when the pair was led to by a CNAME;
820 modern usage is to not canonicalize the domain to a CNAME target
821 (and we were inconsistent anyway for A-only vs AAAA+A).
823 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
825 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
826 when evaluating $sender_host_dnssec.
828 JH/31 Check the HELO verification lookup for DNSSEC, adding new
829 $sender_helo_dnssec variable.
831 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
833 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
835 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
837 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
838 documented as working, but never had. Support all but $spam_report.
840 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
841 added for tls authenticator.
843 HS/03 Add perl_taintmode main config option
849 TL/01 When running the test suite, the README says that variables such as
850 no_msglog_check are global and can be placed anywhere in a specific
851 test's script, however it was observed that placement needed to be near
852 the beginning for it to behave that way. Changed the runtest perl
853 script to read through the entire script once to detect and set these
854 variables, reset to the beginning of the script, and then run through
855 the script parsing/test process like normal.
857 TL/02 The BSD's have an arc4random API. One of the functions to induce
858 adding randomness was arc4random_stir(), but it has been removed in
859 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
860 function when detected.
862 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
863 cause callback expansion.
865 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
866 syntax errors in an expansion can be treated as a string instead of
867 logging or causing an error, due to the internal use of bool_lax
868 instead of bool when processing it.
870 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
871 server certificates when making smtp deliveries.
873 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
875 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
877 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
879 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
880 Merged patch from Sebastian Wiedenroth.
882 JH/05 Fix results-pipe from transport process. Several recipients, combined
883 with certificate use, exposed issues where response data items split
884 over buffer boundaries were not parsed properly. This eventually
885 resulted in duplicates being sent. This issue only became common enough
886 to notice due to the introduction of connection certificate information,
887 the item size being so much larger. Found and fixed by Wolfgang Breyha.
889 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
890 size buffer was used, resulting in syntax errors when an expansion
893 JH/07 Add support for directories of certificates when compiled with a GnuTLS
894 version 3.3.6 or later.
896 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
897 is EXPERIMENTAL_EVENT, the main-configuration and transport options
898 both become "event_action", the variables become $event_name, $event_data
899 and $event_defer_errno. There is a new variable $verify_mode, usable in
900 routers, transports and related events. The tls:cert event is now also
901 raised for inbound connections, if the main configuration event_action
904 TL/06 In test suite, disable OCSP for old versions of openssl which contained
905 early OCSP support, but no stapling (appears to be less than 1.0.0).
907 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
908 server certificate names available under the smtp transport option
909 "tls_verify_cert_hostname" now do not permit multi-component wildcard
912 JH/10 Time-related extraction expansions from certificates now use the main
913 option "timezone" setting for output formatting, and are consistent
914 between OpenSSL and GnuTLS compilations. Bug 1541.
916 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
917 encoded parameter in the incoming message. Bug 1558.
919 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
920 include certificate info, eximon was claiming there were spoolfile
923 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
925 JH/14 Log delivery-related information more consistently, using the sequence
926 "H=<name> [<ip>]" wherever possible.
928 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
929 are problematic for Debian distribution, omit them from the release
932 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
934 JH/16 Fix string representation of time values on 64bit time_t architectures.
937 JH/17 Fix a null-indirection in certextract expansions when a nondefault
938 output list separator was used.
943 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
944 checkers that were complaining about end of non-void function with no
947 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
948 This was a regression introduced in 4.83 by another bugfix.
950 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
952 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
953 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
959 TF/01 Correctly close the server side of TLS when forking for delivery.
961 When a message was received over SMTP with TLS, Exim failed to clear up
962 the incoming connection properly after forking off the child process to
963 deliver the message. In some situations the subsequent outgoing
964 delivery connection happened to have the same fd number as the incoming
965 connection previously had. Exim would try to use TLS and fail, logging
966 a "Bad file descriptor" error.
968 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
969 utilities have not been installed.
971 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
972 temporary space as the ACL may create new global variables.
974 TL/01 LDAP support uses per connection or global context settings, depending
975 upon the detected version of the libraries at build time.
977 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
978 to extract and use the src ip:port in logging and expansions as if it
979 were a direct connection from the outside internet. PPv2 support was
980 updated based on HAProxy spec change in May 2014.
982 JH/02 Add ${listextract {number}{list}{success}{fail}}.
984 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
985 Properly escape header and check for NULL return.
987 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
990 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
992 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
993 characters in header names, implemented as a verify condition.
994 Contributed by Michael Fischer v. Mollard.
996 TL/05 Rename SPF condition results err_perm and err_temp to standardized
997 results permerror and temperror. Previous values are deprecated but
998 still accepted. In a future release, err_perm and err_temp will be
999 completely removed, which will be a backward incompatibility if the
1000 ACL tests for either of these two old results. Patch contributed by
1001 user bes-internal on the mailing list.
1003 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1005 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1006 selectors, in both main and reject logs.
1008 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1011 JH/07 Add malware type "sock" for talking to simple daemon.
1013 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1015 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1016 routers/transports under cutthrough routing.
1018 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1019 numbers. Touch up "bool" conditional to keep the same definition.
1021 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1023 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1025 JH/12 Expand items in router/transport headers_add or headers_remove lists
1026 individually rather than the list as a whole. Bug 1452.
1028 Required for reasonable handling of multiple headers_ options when
1029 they may be empty; requires that headers_remove items with embedded
1030 colons must have them doubled (or the list-separator changed).
1032 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1033 view the policy declared in the DMARC record. Currently, $dmarc_status
1034 is a combined value of both the record presence and the result of the
1037 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1039 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1040 dnslookup router and the smtp transport (applying to the forward
1043 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1044 of ldap servers used for a specific lookup. Patch provided by Heiko
1047 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1048 New variable $lookup_dnssec_authenticated for observability.
1050 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1051 Patch submitted by Lars Timman.
1053 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1055 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1056 Requires trusted mode and valid format message id, aborts otherwise.
1057 Patch contributed by Heiko Schlichting.
1059 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1060 certextract with support for various fields. Bug 1358.
1062 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1063 is requested by default, modifiable by smtp transport option
1066 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1067 operate on certificate variables to give certificate fingerprints
1068 Also new ${sha256:cert_variable}.
1070 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1072 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1073 Christian Aistleitner.
1075 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1077 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1078 file. Patch from Wolfgang Breyha.
1080 JH/25 Expand the coverage of the delivery $host and $host_address to
1081 client authenticators run in verify callout. Bug 1476.
1083 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1084 align with daemon_smtp_ports. Bug 72.
1086 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1087 support and error reporting did not work properly.
1089 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1090 and is readable. Patch from Andrew Colin Kissa.
1092 TL/14 Enhance documentation of ${run expansion and how it parses the
1093 commandline after expansion, particularly in the case when an
1094 unquoted variable expansion results in an empty value.
1096 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1098 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1099 by renaming away from it. Addresses GH issue 15, reported by
1102 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1104 TL/15 SECURITY: prevent double expansion in math comparison functions
1105 (can expand unsanitized data). Not remotely exploitable.
1112 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1114 PP/02 Make -n do something, by making it not do something.
1115 When combined with -bP, the name of an option is not output.
1117 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1120 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1121 $sender_host_name and config options to manage this, and basic check
1124 PP/05 DSCP support for outbound connections and control modifier for inbound.
1126 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1127 (Only plugin which currently uses this is kerberos4, which nobody should
1128 be using, but we should make it available and other future plugins might
1129 conceivably use it, even though it would break NAT; stuff *should* be
1130 using channel bindings instead).
1132 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1133 name; added for Sendmail compatibility; requires admin caller.
1134 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1135 just ignore it); requires trusted caller.
1136 Also parse but ignore: -Ac -Am -X<logfile>
1139 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1141 TL/02 Add +smtp_confirmation as a default logging option.
1143 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1144 Patch by Magnus Holmgren from 2007-02-20.
1146 TL/04 Bugzilla 1281 - Spec typo.
1147 Bugzilla 1283 - Spec typo.
1148 Bugzilla 1290 - Spec grammar fixes.
1150 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1152 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1154 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1155 mailing list by by Dmitry Isaikin.
1157 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1159 JH/02 Support "G" suffix to numbers in ${if comparisons.
1161 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1163 NM/01 Bugzilla 1197 - Spec typo
1164 Bugzilla 1196 - Spec examples corrections
1166 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1168 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1169 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1172 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1173 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1175 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1176 "acl {{name}{arg}...}", and optional args on acl condition
1179 JH/05 Permit multiple router/transport headers_add/remove lines.
1181 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1183 JH/07 Avoid using a waiting database for a single-message-only transport.
1184 Performance patch from Paul Fisher. Bugzilla 1262.
1186 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1189 JH/09 Add $headers_added variable, with content from use of ACL modifier
1190 add_header (but not yet added to the message). Bugzilla 199.
1192 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1193 Pulled from Bugzilla 817 by Wolfgang Breyha.
1195 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1197 (nb: this is the same fix as in Exim 4.80.1)
1199 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1202 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1205 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1207 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1208 advertise SMTP AUTH mechanism to us, instead of a generic
1209 protocol violation error. Also, make Exim more robust to bad
1210 data from the Dovecot auth socket.
1212 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1214 When a queue runner is handling a message, Exim first routes the
1215 recipient addresses, during which it prunes them based on the retry
1216 hints database. After that it attempts to deliver the message to
1217 any remaining recipients. It then updates the hints database using
1220 So if a recipient address works intermittently, it can get repeatedly
1221 deferred at routing time. The retry hints record remains fresh so the
1222 address never reaches the final cutoff time.
1224 This is a fairly common occurrence when a user is bumping up against
1225 their storage quota. Exim had some logic in its local delivery code
1226 to deal with this. However it did not apply to per-recipient defers
1227 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1229 This change adds a proper retry rule check during routing so that the
1230 final cutoff time is checked against the message's age. We only do
1231 this check if there is an address retry record and there is not a
1232 domain retry record; this implies that previous attempts to handle
1233 the address had the retry_use_local_parts option turned on. We use
1234 this as an approximation for the destination being like a local
1235 delivery, as in LMTP.
1237 I suspect this new check makes the old local delivery cutoff check
1238 redundant, but I have not verified this so I left the code in place.
1240 TF/02 Correct gecos expansion when From: is a prefix of the username.
1242 Test 0254 submits a message to Exim with the header
1246 When I ran the test suite under the user fanf2, Exim expanded
1247 the header to contain my full name, whereas it should have added
1248 a Resent-Sender: header. It erroneously treats any prefix of the
1249 username as equal to the username.
1251 This change corrects that bug.
1253 GF/01 DCC debug and logging tidyup
1254 Error conditions log to paniclog rather than rejectlog.
1255 Debug lines prefixed by "DCC: " to remove any ambiguity.
1257 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1259 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1260 Bug spotted by Jeremy Harris; was flawed since initial commit.
1261 Would have resulted in OCSP responses post-SNI triggering an Exim
1262 NULL dereference and crash.
1264 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1266 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1267 Bug detection, analysis and fix by Samuel Thibault.
1268 Bugzilla 1331, Debian bug #698092.
1270 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1272 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1273 Server implementation by Todd Lyons, client by JH.
1274 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1275 config variable "prdr_enable" controls whether the server
1276 advertises the facility. If the client requests PRDR a new
1277 acl_data_smtp_prdr ACL is called once for each recipient, after
1278 the body content is received and before the acl_smtp_data ACL.
1279 The client is controlled by both of: a hosts_try_prdr option
1280 on the smtp transport, and the server advertisement.
1281 Default client logging of deliveries and rejections involving
1282 PRDR are flagged with the string "PRDR".
1284 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1285 fclose(). Diagnosis by Todd Lyons.
1287 PP/17 Update configure.default to handle IPv6 localhost better.
1288 Patch by Alain Williams (plus minor tweaks).
1291 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1292 This is now consistent with GnuTLS, and is now documented: the
1293 previous undocumented portable approach to treating the option as
1294 unset was to force an expansion failure. That still works, and
1295 an empty string is now equivalent.
1297 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1298 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1299 not performing validation itself.
1301 PP/20 Added force_command boolean option to pipe transport.
1302 Patch from Nick Koston, of cPanel Inc.
1304 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1307 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1309 PP/21 Fix eximon continuous updating with timestamped log-files.
1310 Broken in a format-string cleanup in 4.80, missed when I repaired the
1311 other false fix of the same issue.
1312 Report and fix from Heiko Schlichting.
1315 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1316 Report from Prashanth Katuri.
1318 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1319 It's SecureTransport, so affects any MacOS clients which use the
1320 system-integrated TLS libraries, including email clients.
1322 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1323 using a MIME ACL for non-SMTP local injection.
1324 Report and assistance in diagnosis by Warren Baker.
1326 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1328 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1330 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1331 last id that failed so it may be referenced in subsequent ACL's.
1333 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1336 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1337 ldap library initialization, allowing self-signed CA's to be
1338 used. Also properly sets require_cert option later in code by
1339 using NULL (global ldap config) instead of ldap handle (per
1340 session). Bug diagnosis and testing by alxgomz.
1342 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1343 the src/util/ subdirectory.
1345 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1346 renamed to Transport Post Delivery Action by Jeremy Harris, as
1349 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1350 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1351 redis_servers = needs to be configured which will be used by the redis
1352 lookup. Patch from Warren Baker, of The Packet Hub.
1354 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1356 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1357 hostname or reverse DNS when processing a host list. Used suggestions
1358 from multiple comments on this bug.
1360 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1362 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1363 Missed a few lines, added it to make the runtest require no keyboard
1366 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1367 contains upper case chars. Make router use caseful_local_part.
1369 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1370 support when GnuTLS has been built with p11-kit.
1376 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1378 This, or similar/improved, will also be change PP/11 of 4.82.
1384 PP/01 Handle short writes when writing local log-files.
1385 In practice, only affects FreeBSD (8 onwards).
1386 Bugzilla 1053, with thanks to Dmitry Isaikin.
1388 NM/01 Bugzilla 949 - Documentation tweak
1390 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1393 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1395 PP/02 Implemented gsasl authenticator.
1397 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1399 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1400 `pkg-config foo` for cflags/libs.
1402 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1403 with rest of GSASL and with heimdal_gssapi.
1405 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1406 `pkg-config foo` for cflags/libs for the TLS implementation.
1408 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1409 properties get this fed in as external SSF. A number of robustness
1410 and debugging improvements to the cyrus_sasl authenticator.
1412 PP/08 cyrus_sasl server now expands the server_realm option.
1414 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1415 Patch by Jeremy Harris.
1417 PP/10 Added dbmjz lookup type.
1419 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1421 PP/12 MAIL args handles TAB as well as SP, for better interop with
1422 non-compliant senders.
1423 Analysis and variant patch by Todd Lyons.
1425 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1426 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1427 Patch from Dirk Mueller <dmueller@suse.com>
1429 PP/13 tls_peerdn now print-escaped for spool files.
1430 Observed some $tls_peerdn in wild which contained \n, which resulted
1431 in spool file corruption.
1433 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1434 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1435 or write after TLS renegotiation, which otherwise led to messages
1438 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1439 as a tracking header (ie: a signed header comes before the signature).
1440 Patch from Wolfgang Breyha.
1442 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1443 comma-sep list; embedded commas doubled.
1445 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1447 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1449 Report and patch from Dmitry Banschikov.
1451 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1452 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1453 failures. We appear to now support TLS1.1+ with Exim.
1455 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1456 lets Exim select keys and certificates based upon TLS SNI from client.
1457 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1458 before an outbound SMTP session. New log_selector, +tls_sni.
1460 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1461 NULL dereference. Report and patch from Alun Jones.
1463 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1464 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1465 Not seeing resolver debug output on NetBSD, but suspect this is a
1466 resolver implementation change.
1468 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1469 Left warnings. Added "eximon gdb" invocation mode.
1471 PP/21 Defaulting "accept_8bitmime" to true, not false.
1473 PP/22 Added -bw for inetd wait mode support.
1475 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1476 locate the relevant includes and libraries. Made this the default.
1478 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1479 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1481 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1482 This may cause build issues on older platforms.
1484 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1485 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1486 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1487 Added SNI support via GnuTLS too.
1488 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1490 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1492 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1493 Applied second patch from Janne, implementing suggestion to default
1494 multiple-strings-in-record handling to match SPF spec.
1496 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1498 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1499 read-only, out of scope).
1500 Patch from Wolfgang Breyha, report from Stuart Northfield.
1502 PP/29 Fix three issues highlighted by clang analyser static analysis.
1503 Only crash-plausible issue would require the Cambridge-specific
1504 iplookup router and a misconfiguration.
1505 Report from Marcin Mirosław.
1507 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1509 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1510 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1511 As part of this, removing so much warning spew let me fix some minor
1512 real issues in debug logging.
1514 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1515 assignment on my part. Fixed.
1517 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1518 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1519 Janne Snabb (who went above and beyond: thank you).
1521 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1522 string otherwise requires a connection and a bunch more work and it's
1523 relatively easy to get wrong. Should also expose TLS library linkage
1526 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1527 64-bit ${eval} (JH/03).
1529 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1530 GNU libc to support some of the 64-bit stuff, should not lead to
1531 conflicts. Defined before os.h is pulled in, so if a given platform
1532 needs to override this, it can.
1534 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1535 protection layer was required, which is not implemented.
1536 Bugzilla 1254, patch from Wolfgang Breyha.
1538 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1539 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1540 tls_dhparam take prime identifiers. Also unbreak combination of
1541 OpenSSL+DH_params+TLSSNI.
1543 PP/39 Disable SSLv2 by default in OpenSSL support.
1549 PP/01 Solaris build fix for Oracle's LDAP libraries.
1550 Bugzilla 1109, patch from Stephen Usher.
1552 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1554 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1557 TF/02 Fix a couple more cases where we did not log the error message
1558 when unlink() failed. See also change 4.74-TF/03.
1560 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1561 lock up or crash if it happened to be inside a call to libc when it
1562 got a SIGUSR1 from exiwhat.
1564 The SIGUSR1 handler appends the current process status to the process
1565 log which is later printed by exiwhat. It used to use the general
1566 purpose logging code to do this, but several functions it calls are
1567 not safe for signals.
1569 The new output code in the SIGUSR1 handler is specific to the process
1570 log, and simple enough that it's easy to inspect for signal safety.
1571 Removing some special cases also simplifies the general logging code.
1572 Removing the spurious timestamps from the process log simplifies
1575 TF/04 Improved ratelimit ACL condition.
1577 The /noupdate option has been deprecated in favour of /readonly which
1578 has clearer semantics. The /leaky, /strict, and /readonly update modes
1579 are mutually exclusive. The update mode is no longer included in the
1580 database key; it just determines when the database is updated. (This
1581 means that when you upgrade Exim will forget old rate measurements.)
1583 Exim now checks that the per_* options are used with an update mode that
1584 makes sense for the current ACL. For example, when Exim is processing a
1585 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1586 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1587 must specify per_mail/readonly. If you omit the update mode it defaults to
1588 /leaky where that makes sense (as before) or /readonly where required.
1590 The /noupdate option is now undocumented but still supported for
1591 backwards compatibility. It is equivalent to /readonly except that in
1592 ACLs where /readonly is required you may specify /leaky/noupdate or
1593 /strict/noupdate which are treated the same as /readonly.
1595 A useful new feature is the /count= option. This is a generalization
1596 of the per_byte option, so that you can measure the throughput of other
1597 aggregate values. For example, the per_byte option is now equivalent
1598 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1600 The per_rcpt option has been generalized using the /count= mechanism
1601 (though it's more complicated than the per_byte equivalence). When it is
1602 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1603 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1604 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1605 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1606 Note that using per_rcpt with a non-readonly update mode in more than
1607 one ACL will cause the recipients to be double-counted. (The per_mail
1608 and per_byte options don't have this problem.)
1610 The handling of very low rates has changed slightly. If the computed rate
1611 is less than the event's count (usually one) then this event is the first
1612 after a long gap. In this case the rate is set to the same as this event's
1613 count, so that the first message of a spam run is counted properly.
1615 The major new feature is a mechanism for counting the rate of unique
1616 events. The new per_addr option counts the number of different
1617 recipients that someone has sent messages to in the last time period. It
1618 behaves like per_rcpt if all the recipient addresses are different, but
1619 duplicate recipient addresses do not increase the measured rate. Like
1620 the /count= option this is a general mechanism, so the per_addr option
1621 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1622 example, measure the rate that a client uses different sender addresses
1623 with the options per_mail/unique=$sender_address. There are further
1624 details in the main documentation.
1626 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1628 TF/06 Removed a few PCRE remnants.
1630 TF/07 Automatically extract Exim's version number from tags in the git
1631 repository when doing development or release builds.
1633 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1634 Bugzilla 879. Patch from Paul Fisher.
1636 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1637 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1640 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1642 PP/05 Variable $av_failed, true if the AV scanner deferred.
1643 Bugzilla 1078. Patch from John Horne.
1645 PP/06 Stop make process more reliably on build failure.
1646 Bugzilla 1087. Patch from Heiko Schlittermann.
1648 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1649 Bugzilla 1089. Patch from Heiko Schlittermann.
1651 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1652 Bugzilla 1131. Patch from Holger Weiß.
1654 PP/09 Handle IPv6 addresses with SPF.
1655 Bugzilla 860. Patch from Wolfgang Breyha.
1657 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1659 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1662 PP/11 match_* no longer expand right-hand-side by default.
1663 New compile-time build option, EXPAND_LISTMATCH_RHS.
1664 New expansion conditions, "inlist", "inlisti".
1666 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1668 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1670 PP/14 fix log_write() format string regression from TF/03.
1671 Bugzilla 1152. Patch from Dmitry Isaikin.
1677 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1679 PP/02 Harmonised TLS library version reporting; only show if debugging.
1680 Layout now matches that introduced for other libraries in 4.74 PP/03.
1682 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1684 PP/04 New "dns_use_edns0" global option.
1686 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1689 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1690 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1692 TK/01 Updated PolarSSL code to 0.14.2.
1693 Bugzilla 1097. Patch from Andreas Metzler.
1695 PP/07 Catch divide-by-zero in ${eval:...}.
1696 Fixes bugzilla 1102.
1698 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1701 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1702 format-string attack -- SECURITY: remote arbitrary code execution.
1704 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1705 time unintentionally subject to list matching rules, letting the header
1706 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1707 arbitrary string expansion). This allowed for information disclosure.
1709 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1710 INT_MIN/-1 -- value coerced to INT_MAX.
1716 NM/01 Workaround for PCRE version dependency in version reporting
1719 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1720 This fixes portability to compilers other than gcc, notably
1721 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1723 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1724 makefiles for portability to HP-UX and POSIX correctness.
1726 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1727 Also via indented variable definition in the Makefile.
1728 (Debugging by Oliver Heesakkers).
1730 PP/02 Restore caching of spamd results with expanded spamd_address.
1731 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1733 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1734 Improves build reliability. Fix from: Frank Elsner
1736 NM/02 Fix wide character breakage in the rfc2047 coding
1737 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1739 NM/03 Allow underscore in dnslist lookups
1740 Fixes bug 1026. Patch from Graeme Fowler
1742 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1743 Code patches from Adam Ciarcinski of NetBSD.
1745 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1748 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1749 is logged, to avoid truncation. Patch from John Horne.
1751 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1752 Patch from Jakob Hirsch.
1754 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1755 SQL string expansion failure details.
1756 Patch from Andrey Oktyabrski.
1758 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1759 Patch from Simon Arlott.
1761 PP/09 New lookups functionality failed to compile on old gcc which rejects
1762 extern declarations in function scope.
1763 Patch from Oliver Fleischmann
1765 PP/10 Use sig_atomic_t for flags set from signal handlers.
1766 Check getgroups() return and improve debugging.
1767 Fixed developed for diagnosis in bug 927 (which turned out to be
1770 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1771 Patch from Mark Zealey.
1773 PP/12 Bugzilla 1056: Improved spamd server selection.
1774 Patch from Mark Zealey.
1776 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1777 Based on patch from Heiko Schlittermann.
1779 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1780 Patch from Uwe Doering, sign-off by Michael Haardt.
1782 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1783 variable declaration deep within a block. Bug and patch from
1786 PP/15 lookups-Makefile IRIX compatibility coercion.
1788 PP/16 Make DISABLE_DKIM build knob functional.
1790 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1791 Patch by Simon Arlott
1793 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1794 variable argument macros. Our copy now differs from upstream.
1800 TF/01 Failure to get a lock on a hints database can have serious
1801 consequences so log it to the panic log.
1803 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1804 controlled using the smtp_confirmation log selector.
1806 TF/03 Include the error message when we fail to unlink a spool file.
1808 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1809 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1810 for maintaining out-of-tree patches for some time.
1812 PP/01 Bugzilla 139: Documentation and portability issues.
1813 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1814 Handle per-OS dynamic-module compilation flags.
1816 PP/02 Let /dev/null have normal permissions.
1817 The 4.73 fixes were a little too stringent and complained about the
1818 permissions on /dev/null. Exempt it from some checks.
1819 Reported by Andreas M. Kirchwitz.
1821 PP/03 Report version information for many libraries, including
1822 Exim version information for dynamically loaded libraries. Created
1823 version.h, now support a version extension string for distributors
1824 who patch heavily. Dynamic module ABI change.
1826 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1827 privilege escalation vulnerability whereby the Exim run-time user
1828 can cause root to append content of the attacker's choosing to
1831 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1834 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1835 If dropping privileges for untrusted macros, we disabled normal logging
1836 on the basis that it would fail; for the Exim run-time user, this is not
1837 the case, and it resulted in successful deliveries going unlogged.
1838 Fixed. Reported by Andreas Metzler.
1844 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1845 only prepend for the Resent-* case. Fixes regression introduced in
1846 Exim 4.70 by NM/22 for Bugzilla 607.
1848 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1849 increasing numbers of administrators be bitten by this.
1851 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1853 PP/03 Bugzilla 994: added openssl_options main configuration option.
1855 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1857 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1859 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1861 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1862 without a peer certificate, leading to a segfault because of an
1863 assumption that peers always have certificates. Be a little more
1864 paranoid. Problem reported by Martin Tscholak.
1866 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1867 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1868 NB: ClamAV planning to remove STREAM in "middle of 2010".
1869 CL also introduces -bmalware, various -d+acl logging additions and
1870 more caution in buffer sizes.
1872 PP/09 Implemented reverse_ip expansion operator.
1874 PP/10 Bugzilla 937: provide a "debug" ACL control.
1876 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1878 PP/12 Bugzilla 973: Implement --version.
1880 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1882 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1884 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1886 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1887 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1888 ignore trailing whitespace.
1890 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1892 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1895 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1896 Notification from Dr Andrew Aitchison.
1898 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1899 ExtendedDetectionInfo response format.
1900 Notification from John Horne.
1902 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1905 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1906 XSL and documented dependency on system catalogs, with examples of how
1909 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1912 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1913 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1914 configuration file which is writeable by the Exim user or group.
1916 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1917 of configuration files to cover files specified with the -C option if
1918 they are going to be used with root privileges, not just the default
1921 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1922 option (effectively making it always true).
1924 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1925 files to be used while preserving root privileges.
1927 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1928 that rogue child processes cannot use them.
1930 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1931 run-time user, instead of root.
1933 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1934 Exim run-time user without dropping privileges.
1936 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1937 result string, instead of calling string_vformat() twice with the same
1940 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1941 for other users. Others should always drop root privileges if they use
1942 -C on the command line, even for a whitelisted configure file.
1944 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1946 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1952 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1953 $data_path, and $header_path variables; fixed documentation bugs and
1956 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1957 exipick to access non-standard spools, including the "frozen" queue
1960 NM/01 Bugzilla 965: Support mysql stored procedures.
1961 Patch from Alain Williams
1963 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1965 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1966 Patch from Andreas Metzler
1968 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1969 Patch from Kirill Miazine
1971 NM/05 Bugzilla 671: Added umask to procmail example.
1973 JJ/03 installed exipick 20100323.0, fixing doc bug
1975 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1976 directory. Notification and patch from Dan Rosenberg.
1978 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1980 TK/02 Improve log output when DKIM signing operation fails.
1982 MH/01 Treat the transport option dkim_domain as a colon separated
1983 list, not as a single string, and sign the message with each element,
1984 omitting multiple occurences of the same signer.
1986 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1987 Bugzilla 985, 986. Patch by Simon Arlott
1989 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1990 Patch by Simon Arlott
1992 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1993 MBX locking. Notification from Dan Rosenberg.
1999 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2001 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2003 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2005 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2007 NM/04 Bugzilla 915: Flag broken perl installation during build.
2013 TK/01 Added patch by Johannes Berg that expands the main option
2014 "spamd_address" if it starts with a dollar sign.
2016 TK/02 Write list of recipients to X-Envelope-Sender header when building
2017 the mbox-format spool file for content scanning (suggested by Jakob
2020 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2021 (http://www.dcc-servers.net/) support via dccifd. Activated by
2022 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2024 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2025 by Mark Daniel Reidel <mr@df.eu>.
2027 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2028 When building exim an external PCRE library is now needed -
2029 PCRE is a system library on the majority of modern systems.
2030 See entry on PCRE_LIBS in EDITME file.
2032 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2033 conversation. Added nologin parameter to request.
2034 Patch contributed by Kirill Miazine.
2036 TF/01 Do not log submission mode rewrites if they do not change the address.
2038 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2040 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2041 log files in place. Contributed by Roberto Lima.
2043 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2045 TF/03 Bugzilla 615: When checking the local_parts router precondition
2046 after a local_part_suffix or local_part_prefix option, Exim now
2047 does not use the address's named list lookup cache, since this
2048 contains cached lookups for the whole local part.
2050 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2051 Robert Millan. Documentation is in experimental-spec.txt.
2053 TF/04 Bugzilla 668: Fix parallel build (make -j).
2055 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2057 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2058 Patch provided by Jan Srzednicki.
2060 TF/05 Leading white space used to be stripped from $spam_report which
2061 wrecked the formatting. Now it is preserved.
2063 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2064 that they are available at delivery time.
2066 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2068 TF/08 TLS error reporting now respects the incoming_interface and
2069 incoming_port log selectors.
2071 TF/09 Produce a more useful error message if an SMTP transport's hosts
2072 setting expands to an empty string.
2074 NM/06 Bugzilla 744: EXPN did not work under TLS.
2075 Patch provided by Phil Pennock.
2077 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2078 Patch provided by Richard Godbee.
2080 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2081 acl_smtp_notquit, added index entry.
2083 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2084 Patch provided by Eugene Bujak.
2086 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2087 accept(). Patch provided by Maxim Dounin.
2089 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2090 Patch provided by Phil Pennock.
2092 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2094 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2095 Patch provided by Brad "anomie" Jorsch.
2097 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2098 Patch provided by Dean Brooks.
2100 TK/05 Add native DKIM support (does not depend on external libraries).
2102 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2103 Patch provided by Graeme Fowler.
2105 NM/16 Bugzilla 851: Documentation example syntax fix.
2107 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2109 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2112 NM/19 Bugzilla 745: TLS version reporting.
2113 Patch provided by Phil Pennock.
2115 NM/20 Bugzilla 167: bool: condition support.
2116 Patch provided by Phil Pennock.
2118 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2119 clients. Patch provided by Phil Pennock.
2121 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2122 Patch provided by Brad "anomie" Jorsch.
2124 NM/23 Bugzilla 687: Fix misparses in eximstats.
2125 Patch provided by Heiko Schlittermann.
2127 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2128 Patch provided by Heiko Schlittermann.
2130 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2131 plus update to original patch.
2133 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2135 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2136 Patch provided by David Brownlee.
2138 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2140 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2142 NM/30 Bugzilla 888: TLS documentation bugfixes.
2144 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2146 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2147 Unlike the original bugzilla I have changed all shell scripts in src tree.
2149 NM/33 Bugzilla 898: Transport filter timeout fix.
2150 Patch by Todd Rinaldo.
2152 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2153 Patch by Serge Demonchaux.
2155 NM/35 Bugzilla 39: Base64 decode bug fixes.
2156 Patch by Jakob Hirsch.
2158 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2160 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2162 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2164 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2170 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2171 ALT-N's libdkim that I have put here:
2172 http://duncanthrax.net/exim-experimental/
2174 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2175 were called 'true' and it seems that C99 defines that as a reserved
2176 keyword to be used with 'bool' variable types. That means you could
2177 not include C99-style headers which use bools without triggering
2178 build errors in sieve.c.
2180 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2181 as mailq or other aliases. Changed the --help handling significantly
2182 to do whats expected. exim_usage() emits usage/help information.
2184 SC/01 Added the -bylocaldomain option to eximstats.
2186 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2188 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2190 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2196 PH/01 Another patch from the Sieve maintainer.
2198 PH/02 When an IPv6 address is converted to a string for single-key lookup
2199 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2200 dots are used instead of colons so that keys in lsearch files need not
2201 contain colons. This was done some time before quoting was made available
2202 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2203 (notated using the quote facility) so as to distinguish them from IPv4
2204 keys. This meant that lookups for IP addresses in host lists did not work
2205 for iplsearch lookups.
2207 This has been fixed by arranging for IPv6 addresses to be expressed with
2208 colons if the lookup type is iplsearch. This is not incompatible, because
2209 previously such lookups could never work.
2211 The situation is now rather anomalous, since one *can* have colons in
2212 ordinary lsearch keys. However, making the change in all cases is
2213 incompatible and would probably break a number of configurations.
2215 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2218 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2219 conversion specification without a maximum field width, thereby enabling
2220 a rogue spamd server to cause a buffer overflow. While nobody in their
2221 right mind would setup Exim to query an untrusted spamd server, an
2222 attacker that gains access to a server running spamd could potentially
2223 exploit this vulnerability to run arbitrary code as the Exim user.
2225 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2226 $primary_hostname instead of what libspf2 thinks the hosts name is.
2228 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2229 a directory entry by the name of the lookup key. Previously, if a
2230 symlink pointed to a non-existing file or a file in a directory that
2231 Exim lacked permissions to read, a lookup for a key matching that
2232 symlink would fail. Now it is enough that a matching directory entry
2233 exists, symlink or not. (Bugzilla 503.)
2235 PH/03 The body_linecount and body_zerocount variables are now exported in the
2238 PH/04 Added the $dnslist_matched variable.
2240 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2241 This means they are set thereafter only if the connection becomes
2244 PH/06 Added the client_condition to authenticators so that some can be skipped
2245 by clients under certain conditions.
2247 PH/07 The error message for a badly-placed control=no_multiline_responses left
2248 "_responses" off the end of the name.
2250 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2252 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2253 (without spaces) instead of just copying the configuration text.
2255 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2257 PH/11 Added $max_received_linelength.
2259 PH/12 Added +ignore_defer and +include_defer to host lists.
2261 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2262 way in which PCRE > 7.0 is built.
2264 PH/14 Implemented queue_only_load_latch.
2266 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2267 MAIL command. The effect was to mangle the value on 64-bit systems.
2269 PH/16 Another patch from the Sieve maintainer.
2271 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2273 PH/18 If a system quota error occurred while trying to create the file for
2274 a maildir delivery, the message "Mailbox is full" was not appended to the
2275 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2276 only to a quota excession during the actual writing of the file.
2278 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2279 characters?) which causes problems in log lines. The DN values are now
2280 passed through string_printing() before being added to log lines.
2282 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2283 and InterBase are left for another time.)
2285 PH/21 Added message_body_newlines option.
2287 PH/22 Guard against possible overflow in moan_check_errorcopy().
2289 PH/23 POSIX allows open() to be a macro; guard against that.
2291 PH/24 If the recipient of an error message contained an @ in the local part
2292 (suitably quoted, of course), incorrect values were put in $domain and
2293 $local_part during the evaluation of errors_copy.
2299 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2300 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2303 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2304 issue a MAIL command.
2306 PH/02 In an ACL statement such as
2308 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2310 if a client was not listed at all, or was listed with a value other than
2311 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2312 the condition was not true (as it should be), so access was not denied.
2313 The bug was that the ! inversion was incorrectly passed on to the second
2314 item. This has been fixed.
2316 PH/03 Added additional dnslists conditions == and =& which are different from
2317 = and & when the dns lookup returns more than one IP address.
2319 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2320 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2322 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2323 FSYNC, which compiles an option called disable_fsync that allows for
2324 bypassing fsync(). The documentation is heavily laced with warnings.
2326 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2328 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2329 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2330 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2331 including adding "make clean"; (3) Added -fPIC when compiling the test
2332 dynamically loaded module, to get rid of a warning.
2334 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2335 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2336 = 0s. The bug is otherwise harmless.
2338 PH/07 There was a bug in the dovecot authenticator such that the value of
2339 $auth1 could be overwritten, and so not correctly preserved, after a
2340 successful authentication. This usually meant that the value preserved by
2341 the server_setid option was incorrect.
2343 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2345 PH/09 Installed PCRE release 7.0.
2347 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2348 run for batched SMTP input. It is now run at the start of every message
2349 in the batch. While fixing this I discovered that the process information
2350 (output by running exiwhat) was not always getting set for -bs and -bS
2351 input. This is fixed, and it now also says "batched" for BSMTP.
2353 PH/11 Added control=no_pipelining.
2355 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2356 patch, slightly modified), and move the expansion of helo_data till after
2357 the connection is made in the smtp transport (so it can use these
2360 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2362 PH/14 Added log_selector = +pid.
2364 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2366 PH/16 Add ${if forany and ${if forall.
2368 PH/17 Added dsn_from option to vary the From: line in DSNs.
2370 PH/18 Flush SMTP output before performing a callout, unless control =
2371 no_callout_flush is set.
2373 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2374 was true (the default) a successful delivery failed to delete the retry
2375 item, thus causing premature timeout of the address. The bug is now
2378 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2380 PH/21 Long custom messages for fakedefer and fakereject are now split up
2381 into multiline responses in the same way that messages for "deny" and
2382 other ACL rejections are.
2384 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2385 with slight modification.
2387 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2388 draft, changing the syntax and factoring some duplicate code".
2390 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2391 for deliveries of the second and subsequent messages over the same SMTP
2394 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2395 ${reduce, with only minor "tidies".
2397 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2399 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2400 expansion side effects.
2402 PH/27 When a message times out after an over-quota error from an Exim-imposed
2403 quota, the bounce message says "mailbox is full". This message was not
2404 being given when it was a system quota that was exceeded. It now should
2407 MH/03 Made $recipients available in local_scan(). local_scan() already has
2408 better access to the recipient list through recipients_list[], but
2409 $recipients can be useful in postmaster-provided expansion strings.
2411 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2412 in the case of a MAIL command with additional options following the
2413 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2414 were accidentally chopped off.
2416 PH/29 SMTP synchronization checks are implemented when a command is read -
2417 there is a check that no more input is waiting when there shouldn't be
2418 any. However, for some commands, a delay in an ACL can mean that it is
2419 some time before the response is written. In this time, more input might
2420 arrive, invalidly. So now there are extra checks after an ACL has run for
2421 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2422 pipelining has not been advertised.
2424 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2426 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2427 returned in the SMTP response when smtp_return_error_details was false.
2428 This has been fixed.
2430 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2431 instead of the C I/O that was originally supplied, because problems were
2432 reported on Solaris.
2434 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2435 Exim which did not show up earlier: it was assuming that a call to
2436 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2437 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2438 was a macro that became an assignment, so it seemed to work. This has
2439 changed to a proper function call with a void return, hence the compile
2440 error. Exim's code has been fixed.
2442 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2445 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2447 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2449 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2450 because it turns out to be too big or there is a timeout) there is no
2451 "Completed" line in the log. When some messages of this type were
2452 selected by exigrep, they were listed as "not completed". Others were
2453 picked up by some special patterns. I have improved the selection
2454 criteria to be more general.
2456 PH/38 The host_find_failed option in the manualroute router can now be set
2457 to "ignore", to completely ignore a host whose IP address cannot be
2458 found. If all hosts are ignored, the behaviour is controlled by the new
2459 host_all_ignored option.
2461 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2462 homing or because of multiple MX records with /mx) generated more than
2463 one IP address, and the following item turned out to be the local host,
2464 all the secondary addresses of the first item were incorrectly removed
2465 from the list, along with the local host and any following hosts (which
2466 is what is supposed to happen).
2468 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2469 whoever called Exim into the -H file. In the case of the daemon it was
2470 behaving confusingly. When first started, it used values for whoever
2471 started the daemon, but after a SIGHUP it used the Exim user (because it
2472 calls itself on a restart). I have changed the code so that it now always
2475 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2476 message are rejected with the same error (e.g. no authentication or bad
2477 sender address), and a DATA command is nevertheless sent (as can happen
2478 with PIPELINING or a stupid MUA), the error message that was given to the
2479 RCPT commands is included in the rejection of the DATA command. This is
2480 intended to be helpful for MUAs that show only the final error to their
2483 PH/42 Another patch from the Sieve maintainer.
2485 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2486 Eximstats - Fixed some broken HTML links and added missing column headers
2488 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2489 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2491 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2493 PH/43 Yet another patch from the Sieve maintainer.
2495 PH/44 I found a way to check for a TCP/IP connection going away before sending
2496 the response to the final '.' that terminates a message, but only in the
2497 case where the client has not sent further data following the '.'
2498 (unfortunately, this is allowed). However, in many cases there won't be
2499 any further data because there won't be any more messages to send. A call
2500 to select() can be used: if it shows that the input is "ready", there is
2501 either input waiting, or the socket has been closed. An attempt to read
2502 the next input character can distinguish the two cases. Previously, Exim
2503 would have sent an OK response which the client would never have see.
2504 This could lead to message repetition. This fix should cure that, at
2505 least in a lot of common cases.
2507 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2508 advertised in response to EHLO.
2514 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2515 fixed by 4.65/MH/01 (is this a record?) are fixed:
2517 (i) An empty string was always treated as zero by the numeric comparison
2518 operators. This behaviour has been restored.
2520 (ii) It is documented that the numeric comparison operators always treat
2521 their arguments as decimal numbers. This was broken in that numbers
2522 starting with 0 were being interpreted as octal.
2524 While fixing these problems I realized that there was another issue that
2525 hadn't been noticed. Values of message_size_limit (both the global option
2526 and the transport option) were treated as octal if they started with 0.
2527 The documentation was vague. These values are now always treated as
2528 decimal, and I will make that clear in the documentation.
2534 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2535 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2538 MH/01 Don't check that the operands of numeric comparison operators are
2539 integers when their expansion is in "skipping" mode (fixes bug
2540 introduced by 4.64-PH/07).
2542 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2543 child addresses, Exim now panics and dies. Previously, because the count
2544 is held in a short int, deliveries were likely to be lost. As such a
2545 large number of recipients for a single message is ridiculous
2546 (performance will be very, very poor), I have chosen to impose a limit
2547 rather than extend the field.
2553 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2554 leftover -K file (the existence of which was triggered by #402).
2555 While we were at it, introduced process PID as part of the -K
2556 filename. This should rule out race conditions when creating
2559 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2560 processing considerably. Previous code took too long for large mails,
2561 triggering a timeout which in turn triggers #401.
2563 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2564 in the DK code in transports.c. sendfile() is not really portable,
2565 hence the _LINUX specificness.
2567 TF/01 In the add_headers option to the mail command in an Exim filter,
2568 there was a bug that Exim would claim a syntax error in any
2569 header after the first one which had an odd number of characters
2572 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2573 callout verification, Exim cached a "reject" for the entire domain. This
2574 is correct for most verifications, but it is not correct for a recipient
2575 verification with use_sender or use_postmaster set, because in that case
2576 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2577 case of MAIL FROM:<> rejection from other early rejections (e.g.
2578 rejection of HELO). When verifying a recipient using a non-null MAIL
2579 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2580 Whatever the result of the callout, the value of the domain cache is
2581 left unchanged (for any other kind of callout, getting as far as trying
2582 RCPT means that the domain itself is ok).
2584 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2587 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2588 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2589 session, the daemon ignores ECONNECT errors and logs others; it now
2590 ignores EPIPE as well.
2592 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2593 (quoted-printable decoding).
2595 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2596 later the small subsequent patch to fix an introduced bug.
2598 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2600 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2602 PH/08 An error is now given if message_size_limit is specified negative.
2604 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2605 to be given (somewhat) arbitrary names.
2607 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2610 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2611 miscellaneous code fixes
2613 PH/10 Added the log_reject_target ACL modifier to specify where to log
2616 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2617 hostname. This is wrong, because it relates to the incoming message (and
2618 probably the interface on which it is arriving) and not to the outgoing
2619 callout (which could be using a different interface). This has been
2620 changed to use the value of the helo_data option from the smtp transport
2621 instead - this is what is used when a message is actually being sent. If
2622 there is no remote transport (possible with a router that sets up host
2623 addresses), $smtp_active_hostname is used.
2625 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2626 tweaks were necessary in order to get it to work (see also 21 below):
2627 (a) The code assumed that strncpy() returns a negative number on buffer
2628 overflow, which isn't the case. Replaced with Exim's string_format()
2630 (b) There were several signed/unsigned issues. I just did the minimum
2631 hacking in of casts. There is scope for a larger refactoring.
2632 (c) The code used strcasecmp() which is not a standard C function.
2633 Replaced with Exim's strcmpic() function.
2634 (d) The code set only $1; it now sets $auth1 as well.
2635 (e) A simple test gave the error "authentication client didn't specify
2636 service in request". It would seem that Dovecot has changed its
2637 interface. Fortunately there's a specification; I followed it and
2638 changed what the client sends and it appears to be working now.
2640 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2643 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2644 address A is aliased to B and C, where B exists and C does not. Without
2645 -v the output is "A verified" because verification stops after a
2646 successful redirection if more than one address is generated. However,
2647 with -v the child addresses are also verified. Exim was outputting "A
2648 failed to verify" and then showing the successful verification for C,
2649 with its parentage. It now outputs "B failed to verify", showing B's
2650 parentage before showing the successful verification of C.
2652 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2653 look up a TXT record in a specific list after matching in a combined
2656 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2657 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2658 they consult the DNS. I had assumed they would set it the way they
2659 wanted; and indeed my experiments on Linux seem to show that in some
2660 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2661 To be on the safe side, however, I have now made the interface to
2662 host_find_byname() similar to host_find_bydns(), with an argument
2663 containing the DNS resolver options. The host_find_byname() function now
2664 sets these options at its start, just as host_find_bydns() does. The smtp
2665 transport options dns_qualify_single and dns_search_parents are passed to
2666 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2667 of host_find_byname() use the default settings of RES_DEFNAMES
2668 (qualify_single) but not RES_DNSRCH (search_parents).
2670 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2671 spool_read_header() do less string testing, by means of a preliminary
2672 switch on the second character of optional "-foo" lines. (This is
2673 overdue, caused by the large number of possibilities that now exist.
2674 Originally there were few.) While I was there, I also converted the
2675 str(n)cmp tests so they don't re-test the leading "-" and the first
2676 character, in the hope this might squeeze out yet more improvement.
2678 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2679 flag allowing group syntax was set by the header_syntax check but not
2680 turned off, possible causing trouble later; (2) The flag was not being
2681 set at all for the header_verify test, causing "group"-style headers to
2682 be rejected. I have now set it in this case, and also caused header_
2683 verify to ignore an empty address taken from a group. While doing this, I
2684 came across some other cases where the code for allowing group syntax
2685 while scanning a header line wasn't quite right (mostly, not resetting
2686 the flag correctly in the right place). These bugs could have caused
2687 trouble for malformed header lines. I hope it is now all correct.
2689 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2690 with the "reply" argument non-NULL. The code, however (which originally
2691 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2692 but it didn't always do it. This confused somebody who was copying the
2693 code for some other use. I have removed all the tests.
2695 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2696 feature that was used to support insecure browsers during the U.S. crypto
2697 embargo. It requires special client support, and Exim is probably the
2698 only MTA that supported it -- and would never use it because real RSA is
2699 always available. This code has been removed, because it had the bad
2700 effect of slowing Exim down by computing (never used) parameters for the
2701 RSA_EXPORT functionality.
2703 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2704 authenticator to fail if there's a tab character in the incoming data
2705 (there should never be unless someone is messing about, as it's supposed
2706 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2707 if the connection is using TLS or if the remote IP is the same as the
2708 local IP, and the "valid-client-cert option" if a client certificate has
2711 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2712 authenticators. This can be used for authorization after authentication
2713 succeeds. (In the case of plaintext, it servers for both authentication
2716 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2717 if any retry times were supplied.
2719 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2720 connection, where there is no client IP address to check. In this
2721 situation, the verify now always succeeds.
2723 PH/25 Applied John Jetmore's -Mset patch.
2725 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2727 PH/27 In a string expansion for a processed (not raw) header when multiple
2728 headers of the same name were present, leading whitespace was being
2729 removed from all of them, but trailing whitespace was being removed only
2730 from the last one. Now trailing whitespace is removed from each header
2731 before concatenation. Completely empty headers in a concatenation (as
2732 before) are ignored.
2734 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2735 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2737 PH/29 [Removed. This was a change that I later backed out, and forgot to
2738 correct the ChangeLog entry (that I had efficiently created) before
2739 committing the later change.]
2741 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2742 address errors (4xx response to RCPT) over the same connection as other
2743 messages routed to the same hosts. Such deliveries are always "forced",
2744 so retry times are not inspected. This resulted in far too many retries
2745 for the affected addresses. The effect occurred only when there were more
2746 hosts than the hosts_max_try setting in the smtp transport when it had
2747 the 4xx errors. Those hosts that it had tried were not added to the list
2748 of hosts for which the message was waiting, so if all were tried, there
2749 was no problem. Two fixes have been applied:
2751 (i) If there are any address or message errors in an SMTP delivery, none
2752 of the hosts (tried or untried) are now added to the list of hosts
2753 for which the message is waiting, so the message should not be a
2754 candidate for sending over the same connection that was used for a
2755 successful delivery of some other message. This seems entirely
2756 reasonable: after all the message is NOT "waiting for some host".
2757 This is so "obvious" that I'm not sure why it wasn't done
2758 previously. Hope I haven't missed anything, but it can't do any
2759 harm, as the worst effect is to miss an optimization.
2761 (ii) If, despite (i), such a delivery is accidentally attempted, the
2762 routing retry time is respected, so at least it doesn't keep
2763 hammering the server.
2765 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2766 in ${readsocket because some servers need this prod.
2768 PH/32 Added some extra debug output when updating a wait-xxx database.
2770 PH/33 The hint "could be header name not terminated by colon", which has been
2771 given for certain expansion errors for a long time, was not being given
2772 for the ${if def:h_colon_omitted{... case.
2774 PH/34 The spec says: "With one important exception, whenever a domain list is
2775 being scanned, $domain contains the subject domain." There was at least
2776 one case where this was not true.
2778 PH/35 The error "getsockname() failed: connection reset by peer" was being
2779 written to the panic log as well as the main log, but it isn't really
2780 panic-worthy as it just means the connection died rather early on. I have
2781 removed the panic log writing for the ECONNRESET error when getsockname()
2784 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2785 runs only) independently of the message's sender address. This meant
2786 that, if the 4xx error was in fact related to the sender, a different
2787 message to the same recipient with a different sender could confuse
2788 things. In particular, this can happen when sending to a greylisting
2789 server, but other circumstances could also provoke similar problems.
2790 I have changed the default so that the retry time for these errors is now
2791 based a combination of the sender and recipient addresses. This change
2792 can be overridden by setting address_retry_include_sender=false in the
2795 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2796 remote server are returned as part of bounce messages. This was not
2797 happening for LMTP over a pipe (the lmtp transport), but now it is the
2798 same for both kinds of LMTP.
2800 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2801 in header lines that were in fact CNAMEs. This is no longer the case.
2803 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2804 and queue runs started by the daemon processed all messages. This has
2805 been fixed so that -R and -S can now usefully be given with -q<time>.
2807 PH/40 Import PCRE release 6.7 (fixes some bugs).
2809 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2811 PH/42 Give an error if -q is specified more than once.
2813 PH/43 Renamed the variables $interface_address and $interface_port as
2814 $received_ip_address and $received_port, to make it clear that these
2815 values apply to message reception, and not to the outgoing interface when
2816 a message is delivered. (The old names remain recognized, of course.)
2818 PH/44 There was no timeout on the connect() call when using a Unix domain
2819 socket in the ${readsocket expansion. There now is.
2821 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2822 be meaningful with "accept".
2824 SC/01 Eximstats V1.43
2825 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2827 SC/02 Eximstats V1.44
2828 Use a glob alias rather than an array ref in the generated
2829 parser. This improves both readability and performance.
2831 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2832 Collect SpamAssassin and rejection statistics.
2833 Don't display local sender or destination tables unless
2834 there is data to show.
2835 Added average volumes into the top table text output.
2837 SC/04 Eximstats V1.46
2838 Collect data on the number of addresses (recipients)
2839 as well as the number of messages.
2841 SC/05 Eximstats V1.47
2842 Added 'Message too big' to the list of mail rejection
2843 reasons (thanks to Marco Gaiarin).
2845 SC/06 Eximstats V1.48
2846 Mainlog lines which have GMT offsets and are too short to
2847 have a flag are now skipped.
2849 SC/07 Eximstats V1.49 (Alain Williams)
2850 Added the -emptyok flag.
2852 SC/08 Eximstats V1.50
2853 Fixes for obtaining the IP address from reject messages.
2855 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2856 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2857 whitespace changes from 4.64-PH/27
2859 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2862 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2863 are found, allow negative numbers in numeric criteria)
2865 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2867 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2868 to match changes made in 4.64-PH/43
2870 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2872 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2873 "domain" is actually an IP address, and force a failure. This locks out
2874 those revolvers/nameservers that support "A-for-A" lookups, in
2875 contravention of the specifications.
2877 PH/48 When a host name was looked up from an IP address, and the subsequent
2878 forward lookup of the name timed out, the host name was left in
2879 $sender_host_name, contrary to the specification.
2881 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2882 restricted to single-key lookups, Exim was not diagnosing an error if
2883 * or *@ was used with a query-style lookup.
2885 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2887 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2888 long ago, but noone interested enough thought of it. Let's just say that
2889 the "1.1" means that there are some new functions that weren't there at
2890 some point in the past.
2892 PH/51 Error processing for expansion failure of helo_data from an smtp
2893 transport during callout processing was broken.
2895 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2896 tested/used via the -bh/-bhc/-bs options.
2898 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2899 bug, fixed in subsequent PCRE releases).
2901 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2902 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2904 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2910 SC/01 Use a glob alias rather than an array ref in eximstats generated
2911 parser. This improves both readability and performance.
2913 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2914 Don't display local sender or destination tables in eximstats unless
2915 there is data to show.
2916 Added average volumes into the eximstats top table text output.
2918 SC/03 Collect data on the number of addresses (recipients) as well
2919 as the number of messages in eximstats.
2921 TF/01 Correct an error in the documentation for the redirect router. Exim
2922 does (usually) call initgroups() when daemonizing.
2924 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2925 with consistent privilege compared to when running as a daemon.
2927 TF/03 Note in the spec that $authenticated_id is not set for local
2928 submissions from trusted users.
2930 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2931 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2933 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2934 by adding some example configuration directives to the default
2935 configuration file. A little bit of work is required to uncomment the
2936 directives and define how usernames and passwords are checked, but
2937 there is now a framework to start from.
2939 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2940 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2941 without this. I don't know how relevant this is to other LDAP libraries.
2943 PH/02 Add the verb name to the "unknown ACL verb" error.
2945 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2947 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2949 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2950 directory not expanded when it should be if an expanded home directory
2951 was set for the address (which is overridden by the transport).
2953 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2956 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2957 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2958 because it is too late at that time, and has no effect.
2960 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2961 security issue with \' (bugzilla #107). I could not use the
2962 PQescapeStringConn() function, because it needs a PGconn value as one of
2965 PH/08 When testing addresses using -bt, indicate those final addresses that
2966 are duplicates that would not cause an additional delivery. At least one
2967 person was confused, thinking that -bt output corresponded to deliveries.
2968 (Suppressing duplicates isn't a good idea as you lose the information
2969 about possibly different redirections that led to the duplicates.)
2971 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2972 systems where poll() doesn't work, in particular OS X.
2974 PH/10 Added more information to debugging output for retry time not reached.
2976 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2977 operations in malware.c.
2979 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2982 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2983 syslog_duplication set false, log lines that would normally be written
2984 both the the main log and to the reject log were not written to syslog at
2987 PH/14 In the default configuration, change the use of "message" in ACL warn
2988 statements to "add_header".
2990 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2991 not followed by a command (e.g. "seen endif").
2993 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2994 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2997 PH/17 Added extra conditions to the default value of delay_warning_condition
3001 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3002 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3003 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3006 The Auto-Submitted: and various List- headers are standardised, whereas I
3007 don't think Precedence: ever was.
3009 PH/18 Refactored debugging code in route_finduser() to show more information,
3010 in particular, the error code if getpwnam() issues one.
3012 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3013 This is apparently needed in addition to the PH/07 change above to avoid
3014 any possible encoding problems.
3016 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3017 but not after initializing Perl.
3019 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3020 output them only if debugging. By default they are written stderr,
3021 apparently, which is not desirable.
3023 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3026 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3029 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3031 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3032 authenticated or an ident call has been made. Suppress the default
3033 values for $authenticated_id and $authenticated_sender (but permit -oMai
3034 and -oMas) when testing with -bh.
3036 PH/24 Re-jigged the order of the tests in the default configuration so that the
3037 tests for valid domains and recipients precede the DNS black list and CSA
3038 tests, on the grounds that those ones are more expensive.
3040 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3041 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3042 This bug exists in every version of Exim that I still have, right back to
3045 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3046 However, an attempt to turn on case-sensitivity in a regex key by
3047 including (?-i) didn't work because the subject string was already
3048 lowercased, and the effects were non-intuitive. It turns out that a
3049 one-line patch can be used to allow (?-i) to work as expected.
3055 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3056 other effects) broke the use of negated acl sub-conditions.
3058 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3061 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3062 "Deny" causes Exim to reject the incoming connection with a 554 error.
3063 Unfortunately, if there is a major crisis, such as a disk failure,
3064 tcp-wrappers gives "deny", whereas what one would like would be some
3065 kind of temporary error. A kludge has been added to help with this.
3066 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3067 554 error is used if errno is still zero or contains ENOENT (which occurs
3068 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3071 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3073 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3074 errors. Otherwise a message that provokes a temporary error (when other
3075 messages do not) can cause a whole host to time out.
3077 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3078 addresses were routed directly to files or pipes from a redirect router.
3079 File deliveries just didn't batch; pipe deliveries might have suffered
3082 PH/06 A failure to get a lock for a hints database would erroneously always say
3083 "Failed to get write lock", even when it was really a read lock.
3085 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3086 of 0600. This has been changed to use the value of the lockfile_mode
3087 option (which defaults to 0600).
3089 PH/08 Applied small patch from the Sieve maintainer.
3091 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3092 folder from quota calculations, a direct delivery into this folder messed
3093 up the contents of the maildirsize file. This was because the regex was
3094 used only to exclude .Trash (or whatever) when the size of the mailbox
3095 was calculated. There was no check that a delivery was happening into an
3096 excluded directory. This bug has been fixed by ignoring all quota
3097 processing for deliveries into excluded directories.
3099 PH/10 Added the maildirfolder_create_regex option to appendfile.
3105 PH/01 The code for finding all the local interface addresses on a FreeBSD
3106 system running IPv6 was broken. This may well have applied to all BSD
3107 systems, as well as to others that have similar system calls. The broken
3108 code found IPv4 interfaces correctly, but gave incorrect values for the
3109 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3110 that it would not match correctly against @[] and not recognize the IPv6
3113 PH/02 The ipliteral router was not recognizing addresses of the form user@
3114 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3116 PH/03 Added disable_ipv6.
3118 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3119 decoded form, because it is most often used to construct To: headers
3120 lines in autoreplies, and the decoded form may well be syntactically
3121 invalid. However, $reply_address has leading white space removed, and all
3122 newlines turned into spaces so that the autoreply transport does not
3125 PH/05 If group was specified without a user on a router, and no group or user
3126 was specified on a transport, the group from the router was ignored.
3128 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3129 for visible compile-time settings that can be used to change these
3130 numbers, for those that want even more. Backwards compatibility with old
3131 spool files has been maintained. However, going back to a previous Exim
3132 release will lost any variables that are in spool files.
3134 PH/07 Two small changes when running in the test harness: increase delay when
3135 passing a TCP/IP connection to a new process, in case the original
3136 process has to generate a bounce, and remove special handling of
3137 127.0.0.2 (sic), which is no longer necessary.
3139 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3140 be the same on different OS.
3142 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3145 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3146 whether --show-vars was specified or not
3148 JJ/02 exipick: Added support for new ACL variable spool format introduced
3151 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3152 syntactically invalid From: or Reply-to: line, and a filter used this to
3153 generate an autoreply, and therefore failed to obtain an address for the
3154 autoreply, Exim could try to deliver to a non-existent relative file
3155 name, causing unrelated and misleading errors. What now happens is that
3156 it logs this as a hard delivery error, but does not attempt to create a
3159 PH/11 The exinext utility has a -C option for testing purposes, but although
3160 the given file was scanned by exinext itself; it wasn't being passed on
3161 when Exim was called.
3163 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3164 an end-of-file indication when reading a command response.
3166 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3167 compiled. In many other places in Exim, IPv6 addresses are always
3168 recognized, so I have changed this. It also means that IPv4 domain
3169 literals of the form [IPV4:n.n.n.n] are now always recognized.
3171 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3172 used if the router is not running as root, for example, when verifying at
3173 ACL time, or when using -bh. The debugging output from this situation was
3174 non-existent - all you got was a failure to exec. I have made two
3177 (a) Failures to set uid/gid, the current directory, or a process leader
3178 in a subprocess such as that created by queryprogram now generate
3179 suitable debugging output when -d is set.
3181 (b) The queryprogram router detects when it is not running as root,
3182 outputs suitable debugging information if -d is set, and then runs
3183 the subprocess without attempting to change uid/gid.
3185 PH/15 Minor change to Makefile for building test_host (undocumented testing
3188 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3189 additional section of a DNS packet that returns MX or SRV records.
3190 Instead, it always explicitly searches for A/AAAA records. This avoids
3191 major problems that occur when a DNS server includes only records of one
3192 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3193 fixed another bug: if SRV records were looked up and the corresponding
3194 address records were *not* found in the additional section, the port
3195 values from the SRV records were lost.
3197 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3198 using the correct key (the original address) when searching the retry
3199 rules in order to find which one to use for generating the retry hint.
3201 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3202 adding the default one. Similarly, if it contains a Reply-To: header, the
3203 errors_reply_to option, if set, is not used.
3205 PH/19 When calculating a retry time, Exim used to measure the "time since
3206 failure" by looking at the "first failed" field in the retry record. Now
3207 it does not use this if it is later than than the arrival time of the
3208 message. Instead it uses the arrival time. This makes for better
3209 behaviour in cases where some deliveries succeed, thus re-setting the
3210 "first failed" field. An example is a quota failure for a huge message
3211 when small messages continue to be delivered. Without this change, the
3212 "time since failure" will always be short, possible causing more frequent
3213 delivery attempts for the huge message than are intended.
3214 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3216 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3217 $1, $2, $3) because the numerical variables can be reset during some
3218 expansion items (e.g. "match"), thereby losing the authentication data.
3220 PH/21 Make -bV show the size of off_t variables so that the test suite can
3221 decide whether to run tests for quotas > 2G.
3223 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3224 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3225 filecount value is greater than 2G or if a quota value is greater than 2G
3226 on a system where the size of off_t is not greater than 4, a panic error
3229 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3230 never match. The debug and -bh output now contains an explicit error
3231 message indicating a malformed IPv4 address or mask.
3233 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3234 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3235 PH/23 above applies.
3237 PH/25 Do not write to syslog when running in the test harness. The only
3238 occasion when this arises is a failure to open the main or panic logs
3239 (for which there is an explicit test).
3241 PH/26 Added the /no_tell option to "control=freeze".
3243 PH/27 If a host name lookup failed very early in a connection, for example, if
3244 the IP address matched host_lookup and the reverse lookup yielded a name
3245 that did not have a forward lookup, an error message of the form "no IP
3246 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3247 could be logged. Now it outputs the IP address instead of "NULL".
3249 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3250 allows the sender and the authenticated sender to be set when
3251 submitting a message from within Exim. Since child_open_exim() is
3252 documented for local_scan(), the new function should be too.
3254 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3255 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3256 results in an empty string is now treated as unset.
3258 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3260 PH/31 Added sender_verify_fail logging option.
3262 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3263 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3264 tidied the source and removed it altogether.
3266 PH/33 When a queue run was abandoned because the load average was too high, a
3267 log line was always written; now it is written only if the queue_run log
3268 selector is set. In addition, the log line for abandonment now contains
3269 information about the queue run such as the pid. This is always present
3270 in "start" and "stop" lines but was omitted from the "abandon" line.
3272 PH/34 Omit spaces between a header name and the colon in the error message that
3273 is given when verify = headers_syntax fails (if there are lots of them,
3274 the message gets confusing).
3276 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3277 names, as there are now some PTR records that contain slashes. This check
3278 is only to protect against broken name servers that fall over on strange
3279 characters, so the fact that it applies to all lookups doesn't matter.
3281 PH/36 Now that the new test suite is complete, we can remove some of the
3282 special code in Exim that was needed for the old test suite. For example,
3283 sorting DNS records because real resolvers return them in an arbitrary
3284 order. The new test suite's fake resolver always returns records in the
3287 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3288 bounces) except when queue_only is set, to avoid logging races between
3289 the different processes.
3291 PH/38 Panic-die if .include specifies a non-absolute path.
3293 PH/39 A tweak to the "H" retry rule from its user.
3295 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3296 a label. They prevented compilation on older perls.
3298 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3299 a warning to be raised on newish perls.
3301 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3302 on queue. Changes to match documented behaviour of showing count of
3303 messages matching specified criteria.
3305 PH/40 Changed the default ident timeout from 30s to 5s.
3307 PH/41 Added support for the use of login_cap features, on those BSD systems
3308 that have them, for controlling the resources used by pipe deliveries.
3310 PH/42 The content-scanning code uses fopen() to create files in which to put
3311 message data. Previously it was not paying any attention to the mode of
3312 the files. Exim runs with umask(0) because the rest of the code creates
3313 files with open(), and sets the required mode explicitly. Thus, these
3314 files were ending up world-writeable. This was not a big issue, because,
3315 being within the spool directory, they were not world-accessible. I have
3316 created a function called modefopen, which takes an additional mode
3317 argument. It sets umask(777), creates the file, chmods it to the required
3318 mode, then resets the umask. All the relevant calls to fopen() in the
3319 content scanning code have been changed to use this function.
3321 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3322 to 24 hours. This avoids potential overflow problems when processing G
3323 and H retry rules. I suspect nobody ever tinkers with this value.
3325 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3327 PH/45 When the plaintext authenticator is running as a client, the server's
3328 challenges are checked to ensure they are valid base64 strings. By
3329 default, the authentication attempt is cancelled if an invalid string is
3330 received. Setting client_ignore_invalid_base64 true ignores these errors.
3331 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3332 they are received. Thus, the responses can be made to depend on the
3333 challenges. If an invalid string is ignored, an empty string is placed in
3336 PH/46 Messages that are created by the autoreply transport now contains a
3337 References: header, in accordance with RFCs 2822 and 3834.
3339 PH/47 Added authenticated_sender_force to the smtp transport.
3341 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3343 PH/49 Installed latest patch from the Sieve maintainer.
3345 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3346 was also set, the appendfile transport was unnecessarily scanning a
3347 directory of message files (e.g. for maildir delivery) to find the count
3348 of files (along with the size), even though it did not need this
3349 information. It now does the scan only if it needs to find either the
3350 size of the count of files.
3352 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3354 PH/52 Two bugs concerned with error handling when the smtp transport is
3357 (i) Exim was not creating retry information for temporary errors given
3358 for individual recipients after the DATA command when the smtp transport
3359 was used in LMTP mode. This meant that they could be retried too
3360 frequently, and not timed out correctly.
3362 (ii) Exim was setting the flag that allows error details to be returned
3363 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3364 recipients that were returned after the DATA command.
3366 PH/53 This is related to PH/52, but is more general: for any failing address,
3367 when detailed error information was permitted to be returned to the
3368 sender, but the error was temporary, then after the final timeout, only
3369 "retry timeout exceeded" was returned. Now it returns the full error as
3370 well as "retry timeout exceeded".
3372 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3373 do this, and (what is worse) MTAs that accept it.
3375 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3376 will now be deprecated.
3378 PH/56 New os.c-cygwin from the Cygwin maintainer.
3380 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3381 formats (previously only available in exim formats via -bpr, -bpru,
3382 and -bpra. Now also available in native and exiqgrep formats)
3384 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3385 with very large, slow to parse queues
3387 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3389 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3391 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3392 responses to authentication challenges, though it was showing the
3393 challenges; (ii) I've removed the CR characters from the debug output for
3396 PH/58 Allow for the insertion of a newline as well as a space when a string
3397 is turned into more than one encoded-word during RFC 2047 encoding. The
3398 Sieve code now uses this.
3400 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3401 data_4xx, lost_connection, tls_required.
3403 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3404 message was being sent as an SMTP response.
3406 PH/61 Add -l and -k options to exicyclog.
3408 PH/62 When verifying, if an address was redirected to one new address, so that
3409 verification continued, and the new address failed or deferred after
3410 having set something in $address_data, the value of $address_data was not
3411 passed back to the ACL. This was different to the case when no
3412 redirection occurred. The value is now passed back in both cases.
3414 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3415 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3416 use login_cap.h, so on its own it isn't the distinguishing feature. The
3417 new name refers directly to the setclassresources() function.
3419 PH/65 Added configuration files for NetBSD3.
3421 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3423 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3424 is preferred over IPv4.
3426 PH/68 The bounce_return_message and bounce_return_body options were not being
3427 honoured for bounces generated during the reception of non-SMTP messages.
3428 In particular, this applied to messages rejected by the ACL. This bug has
3429 been fixed. However, if bounce_return_message is true and bounce_return_
3430 body is false, the headers that are returned for a non-SMTP message
3431 include only those that have been read before the error was detected.
3432 (In the case of an ACL rejection, they have all been read.)
3434 PH/69 The HTML version of the specification is now built in a directory called
3435 spec_html instead of spec.html, because the latter looks like a path with
3436 a MIME-type, and this confuses some software.
3438 PH/70 Catch two compiler warnings in sieve.c.
3440 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3441 function verify_get_ident() calls ip_connect() to connect a socket, but
3442 if the "connect()" function timed out, ip_connect() used to close the
3443 socket. However, verify_get_ident() also closes the socket later, and in
3444 between Exim writes to the log, which may get opened at this point. When
3445 the socket was closed in ip_connect(), the log could get the same file
3446 descriptor number as the socket. This naturally causes chaos. The fix is
3447 not to close the socket in ip_connect(); the socket should be closed by
3448 the function that creates it. There was only one place in the code where
3449 this was missing, in the iplookup router, which I don't think anybody now
3450 uses, but I've fixed it anyway.
3452 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3453 well as to direct DNS lookups. Otherwise the handling of names in host
3454 lists is inconsistent and therefore confusing.
3460 PH/01 Two changes to the default runtime configuration:
3462 (1) Move the checks for relay_from_hosts and authenticated clients from
3463 after to before the (commented out) DNS black list checks.
3465 (2) Add control=submission to the relay_from_hosts and authenticated
3466 clients checks, on the grounds that messages accepted by these
3467 statements are most likely to be submissions.
3469 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3471 (1) Generate an error if the third argument for the ${prvs expansion is
3474 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3477 (3) Reset the variables that are obtained from the first argument of
3478 ${prvscheck and used in the second argument before leaving the code,
3479 because their memory is reclaimed, so using them afterwards may do
3482 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3483 one (it's much easier than Tom thought :-).
3485 (5) Because of (4), we can now allow for the use of $prvscheck_result
3486 inside the third argument.
3488 PH/03 For some reason, the default setting of PATH when running a command from
3489 a pipe transport was just "/usr/bin". I have changed it to
3492 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3493 anything to be listed in the output from -bV.
3495 PH/05 When a filter generated an autoreply, the entire To: header line was
3496 quoted in the delivery log line, like this:
3498 => >A.N.Other <ano@some.domain> <original@ddress> ...
3500 This has been changed so that it extracts the operative address. There
3501 may be more than one such address. If so, they are comma-separated, like
3504 => >ano@some.domain,ona@other.domain <original@ddress> ...
3506 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3507 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3508 not being looked up in the rDNS to get a host name, Exim was showing the
3509 IP address twice in Received: lines, even though the IP addresses were
3510 identical. For example:
3512 Received: from [1.2.3.4] (helo=[1.2.3.4])
3514 However, if the real host name was known, it was omitting the HELO data
3515 if it matched the actual IP address. This has been tidied up so that it
3516 doesn't show the same IP address twice.
3518 PH/07 When both +timestamp and +memory debugging was on, the value given by
3519 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3520 called by the debug printing, thereby overwriting the timestamp buffer.
3521 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3523 PH/08 When the original message was included in an autoreply transport, it
3524 always said "this is a copy of the message, including all the headers",
3525 even if body_only or headers_only was set. It now gives an appropriate
3528 PH/09 Applied a patch from the Sieve maintainer which:
3530 o fixes some comments
3531 o adds the (disabled) notify extension core
3532 o adds some debug output for the result of if/elsif tests
3533 o points to the current vacation draft in the documentation
3534 and documents the missing references header update
3538 o fixes a bug in processing the envelope test (when testing
3539 multiple envelope elements, the last element determined the
3542 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3543 Electronic Mail") by including:
3545 Auto-submitted: auto-generated
3547 in the messages that it generates (bounce messages and others, such as
3548 warnings). In the case of bounce messages for non-SMTP messages, there was
3549 also a typo: it was using "Auto_submitted" (underscore instead of
3550 hyphen). Since every message generated by Exim is necessarily in response
3551 to another message, thes have all been changed to:
3553 Auto-Submitted: auto-replied
3555 in accordance with these statements in the RFC:
3557 The auto-replied keyword:
3559 - SHOULD be used on messages sent in direct response to another
3560 message by an automatic process,
3562 - MUST NOT be used on manually-generated messages,
3564 - MAY be used on Delivery Status Notifications (DSNs) and Message
3565 Disposition Notifications (MDNs),
3567 - MUST NOT be used on messages generated by automatic or periodic
3568 processes, except for messages which are automatic responses to
3571 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3572 to the default Received: header definition.
3574 PH/12 Added log selector acl_warn_skipped (default on).
3576 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3577 variables because (a) they are in the wrong storage pool and (b) even if
3578 they were copied, it wouldn't work properly because of the caching.
3580 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3581 checking when decoding. Apparently there are clients that generate
3582 overlong encoded strings. Why am I not surprised?
3584 PH/15 If the first argument of "${if match_address" was not empty, but did not
3585 contain an "@" character, Exim crashed. Now it writes a panic log message
3586 and treats the condition as false.
3588 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3590 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3591 extension "envelope-auth". The code is finished and in agreement with
3592 other implementations, but there is no documentation so far and in fact,
3593 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3594 not changing the active code.
3596 Print executed "if" and "elsif" statements when debugging is used. This
3597 helps a great deal to understand what a filter does.
3599 Document more things not specified clearly in RFC3028. I had all this
3600 sorted out, when out of a sudden new issues came to my mind. Oops."
3602 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3605 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3606 canonical form (as documented). However, after a host name lookup from
3607 the IP address, check_host() was doing a simple string comparison with
3608 addresses acquired from the DNS when checking that the found name did
3609 have the original IP as one of its addresses. Since any found IPv6
3610 addresses are likely to be in abbreviated form, the comparison could
3611 fail. Luckily, there already exists a function for doing the comparison
3612 by converting both addresses to binary, so now that is used instead of
3613 the text comparison.
3615 PH/20 There was another similar case to PH/19, when a complete host name was
3616 given in a host list; looking up its IP address could give an abbreviated
3617 form, whereas the current host's name might or might not be abbreviated.
3618 The same fix has been applied.
3624 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3625 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3628 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3629 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3631 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3633 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3634 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3635 Also, for "Auto-Submitted", treat anything other than "no" as
3636 non-personal, in accordance with RFC 3834. (Previously it treated
3637 anything starting "auto-" as non-personal.)
3639 TF/01 The control=submission/name=... option had a problem with syntax
3640 errors if the name included a slash character. The /name= option
3641 now slurps the rest of the string, so it can include any characters
3642 but it must come last in the list of options (after /sender_retain
3645 PH/05 Some modifications to the interface to the fake nameserver for the new
3653 TK/01 Added the "success_on_redirect" address verification option. See
3654 NewStuff for rationale and an example.
3656 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3658 PH/02 Patch to exigrep to allow it to work on syslog lines.
3660 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3661 fread() to skip over the body file's header line, because in Cygwin the
3662 header line is locked and is inaccessible.
3664 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3665 co-exist for some time) to make it clear that it is the Exim ID that is
3666 referenced, not the Message-ID: header line.
3668 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3669 string_format() function, because snprintf() does not exist on all
3672 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3673 ${stat: expansion item.
3675 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3676 protocol synchronization error", to keep the pedants happy.
3678 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3679 well as for IRIX systems, when gcc is being used. See the host.c source
3682 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3684 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3687 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3688 transport and to the smtp transport in LMTP mode.
3690 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3692 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3693 run a filter in a subprocess. This could lead to confusion in subsequent
3694 lookups in the parent process. There should also be a search_tidyup() at
3695 the end of the subprocess.
3697 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3698 only if the host matched helo_try_verify_hosts, which caused the
3699 verification to occur when the EHLO/HELO command was issued. The ACL just
3700 tested the remembered result. Now, if a previous verification attempt has
3701 not happened, "verify = helo" does it there and then.
3703 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3705 TK/03 Fix log output including CR from clamd.
3707 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3708 exist provoked a memory error which could cause a segfault.
3710 PH/15 Installed PCRE 6.2
3712 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3714 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3715 of the problem. Specifically, suggested +O2 rather than +O1 for the
3718 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3720 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3721 record, the port was not correctly logged when the outgoing_port log
3722 selector was set (it logged the transort's default port).
3724 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3725 fallback_hosts, and "hosts" in the smtp transport.
3727 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3728 host errors such as "Connection refused".
3730 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3731 authentication with radiusclient 0.4.9:
3733 - Error returned from rc_read_config was caught wrongly
3734 - Username/password not passed on to radius server due to wrong length.
3736 The presumption is that some radiusclient API changes for 4.51/PH/17
3737 were not taken care of correctly. The code is still untested by me (my
3738 Linux distribution still has 0.3.2 of radiusclient), but it was
3739 contributed by a Radius user.
3741 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3742 expanding the "port" option of the smtp transport.
3744 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3745 while reading a MIME header. Thanks to Tom Hughes for a patch.
3747 PH/24 Include config.h inside local_scan.h so that configuration settings are
3750 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3751 that in an ACL for RCPT (for example), you can examine exactly what was
3754 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3755 commands, but it was not correctly comparing the address with the actual
3756 client host address. Thus, it would show the EHLO address in Received:
3757 header lines when this was not necessary.
3759 PH/27 Added the % operator to ${eval:}.
3761 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3762 it should be ignoring failures (because with -C, for example, it has lost
3763 privilege). It wasn't ignoring creation failures other than "already
3766 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3769 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3770 because an input line was too long, either on its own, or by virtue of
3771 too many continuations, the temporary file was not being removed, and the
3772 return code was incorrect.
3774 PH/31 Missing "BOOL" in function definition in filtertest.c.
3776 PH/32 Applied Sieve patches from the maintainer.
3778 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3780 PH/33 Added "verify = not_blind".
3782 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3783 Local/Makefile (with some defaults set). These are used in built scripts
3784 such as exicyclog, but they have never been used in the exim_install
3785 script (though there are many overriding facilities there). I have
3786 arranged that the exim_install script now takes note of these two
3789 PH/35 Installed configuration files for Dragonfly.
3791 PH/36 When a locally submitted message by a trusted user did not contain a
3792 From: header, and the sender address was obtained from -f or from an SMTP
3793 MAIL command, and the trusted user did not use -F to supply a sender
3794 name, $originator_name was incorrectly used when constructing a From:
3795 header. Furthermore, $originator_name was used for submission mode
3796 messages from external hosts without From: headers in a similar way,
3797 which is clearly wrong.
3799 PH/37 Added control=suppress_local_fixups.
3801 PH/38 When log_selector = +received_sender was set, and the addition of the
3802 sender made the log line's construction buffer exactly full, or one byte
3803 less than full, an overflow happened when the terminating "\n" was
3806 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3807 when the result of a list match is failure because a DNS lookup failed.
3809 PH/40 RM_COMMAND is now used in the building process.
3811 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3812 the "build-* directories that it finds.
3814 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3815 address in a domain literal was a prefix of an interface address.
3817 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3818 when verifying a sender address, unless rewrite_headers is false.
3820 PH/44 Wrote a long comment about why errors_to addresses are verified as
3821 recipients, not senders.
3823 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3824 the ratelimit ACL was added.
3826 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3828 PH/46 Added extra information about PostgreSQL errors to the error string.
3830 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3831 avoiding the need to install special zones in a real server. This is
3832 backwards compatible; if it can't find the fake resolver, it drops back.
3833 Thus, both old and new test suites can be run.
3835 TF/02 Added util/ratelimit.pl
3837 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3838 clock is set back in time.
3840 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3841 Candler <B.Candler@pobox.com>.
3843 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3844 OK for addresses that are the result of redirections.
3846 PH/48 A number of further additions for the benefit of the new test suite,
3847 including a fake gethostbyname() that interfaces to the fake DNS resolver
3850 TF/06 The fix for widen_domains has also been applied to qualify_single and
3851 search_parents which are the other dnslookup options that can cause
3854 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3857 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3859 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3860 service" if it encountered an MX record with an empty target hostname.
3861 The message is now "an MX or SRV record indicated no SMTP service".
3863 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3864 if the DNS of the sending site is misconfigured. This is quite a
3865 common situation. This change restores the behaviour of treating a
3866 helo verification defer as a failure.
3868 PH/51 If self=fail was set on a router, the bounce message did not include the
3869 actual error message.
3875 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3877 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3878 command itself ended in error, the underlying message about the transport
3879 filter timeout was being overwritten with the pipe command error. Now the
3880 underlying error message should be appended to the second error message.
3882 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3884 PH/02 Reduce the amount of output that "make" produces by default. Full output
3885 can still be requested.
3887 PH/03 The warning log line about a condition test deferring for a "warn" verb
3888 was being output only once per connection, rather than after each
3889 occurrence (because it was using the same function as for successful
3890 "warn" verbs). This seems wrong, so I have changed it.
3892 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3893 it should not have, which might have caused a crash in the right
3894 circumstances, but probably never did.
3896 PH/04 Installed a modified version of Tony Finch's patch to make submission
3897 mode fix the return path as well as the Sender: header line, and to
3898 add a /name= option so that you can make the user's friendly name appear
3901 TF/03 Added the control = fakedefer ACL modifier.
3903 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3904 Mark Lowes for thorough testing.
3906 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3908 TK/03 Merged latest SRS patch from Miles Wilton.
3910 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3911 with the definition in sysexits.h (which is #included earlier).
3912 Fortunately, Exim does not actually use EX_OK. The code used to try to
3913 preserve the sysexits.h value, by assuming that macro definitions were
3914 scanned for macro replacements. I have been disabused of this notion,
3915 so now the code just undefines EX_OK before #including unistd.h.
3917 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3918 in the smtp transport. When a block could not be written in a single
3919 write() function, the timeout was being re-applied to each part-write.
3920 This seems wrong - if the receiver was accepting one byte at a time it
3921 would take for ever. The timeout is now adjusted when this happens. It
3922 doesn't have to be particularly precise.
3924 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3925 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3927 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3928 without a domain if the check to <postmaster@domain> fails.
3930 SC/01 Eximstats: added -xls and the ability to specify output files
3931 (patch written by Frank Heydlauf).
3933 SC/02 Eximstats: use FileHandles for outputting results.
3935 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3937 SC/04 Eximstats: fixed display of large numbers with -nvr option
3939 SC/05 Eximstats: fixed merging of reports with empty tables.
3941 SC/06 Eximstats: added the -include_original_destination flag
3943 SC/07 Eximstats: removed tabs and trailing whitespace.
3945 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3947 TK/06 MBOX spool code: Add real "From " MBOX separator line
3948 so the .eml file is really in mbox format (even though
3949 most programs do not really care). Patch from Alex Miller.
3951 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3952 The latter is generated from $received_to and is only set if the
3953 message has one envelope recipient. SA can use these headers,
3954 obviously out-of-the-box. Patch from Alex Miller.
3956 PH/08 The ${def test on a variable was returning false if the variable's
3957 value was "0", contrary to what the specification has always said!
3958 The result should be true unless the variable is empty.
3960 PH/09 The syntax error of a character other than { following "${if
3961 def:variable_name" (after optional whitespace) was not being diagnosed.
3962 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3963 accidental colon was present, for example, could give incorrect results.
3965 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3966 result is used (not including appendfile, where other changes are about
3969 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3970 This involved changing a lot of size variables from int to off_t. It
3971 should work with maildirs and everything.
3973 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3974 spamd dying while we are connected to it.
3976 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3979 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3980 function for BDB 4.3.
3982 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3984 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3985 the sources for PCRE have been re-arranged and more files are now
3988 PH/15 The code I had for printing potentially long long variables in PH/11
3989 above was not the best (it lost precision). The length of off_t variables
3990 is now inspected at build time, and an appropriate printing format (%ld
3991 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3992 to be "long long int" or "long int". This is needed for the internal
3993 formatting function string_vformat().
3995 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3996 the configuration file to be ":syslog", then the script "guesses" where
3997 the logs files are, rather than using the compiled in default. In our
3998 case the guess is not the same as the compiled default, so the script
3999 suddenly stopped working when I started to use syslog. The patch checks
4000 to see if log_file_path is "". If so, it attempts to read it from exim
4001 with no configuration file to get the compiled in version, before it
4002 falls back to the previous guessing code."
4004 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4005 implementing BATV in an Exim configuration. See NewStuff for the gory
4008 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4009 Makefile that are specific to HP-UX.
4011 PH/18 If the "use_postmaster" option was set for a recipient callout together
4012 with the "random" option, the postmaster address was used as the MAIL
4013 FROM address for the random test, but not for the subsequent recipient
4014 test. It is now used for both.
4016 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4017 patch removes a few documentation additions to RFC 3028, because the
4018 latest draft now contains them. It adds the new en;ascii-case comparator
4019 and a new error check for 8bit text in MIME parts. Comparator and
4020 require names are now matched exactly. I enabled the subaddress
4021 extension, but it is not well tested yet (read: it works for me)."
4023 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4024 rework some of the code of TK/09 above to avoid the hardwired use of
4025 "%lld" and "long long". Replaced the call to snprintf() with a call to
4028 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4029 records point to non-existent hosts", "retry timeout exceeded", and
4030 "retry time not reached for any host after a long failure period".
4032 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4033 experimental DomainKeys support:
4035 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4036 (2) On an error such as an illegally used "control", the wrong name for
4037 the control was given.
4039 These problems did NOT occur unless DomainKeys support was compiled.
4041 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4043 PH/24 Added ${if match_ip condition.
4045 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4046 databases so that it will be absolutely obvious if a crash occurs in the
4047 DB library. This is a regular occurrence (often caused by mis-matched
4050 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4051 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4052 picked up on a user's system that detects such things. There doesn't seem
4053 to be a gcc warning option for this - only an attribute that has to be
4054 put on the function's prototype. It seems that in Fedora Core 4 they have
4055 set this on a number of new functions. No doubt there will be more in due
4058 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4059 specify a transport. However, if an address that was verified by such a
4060 router was the subject of a callout, Exim crashed because it tried to
4061 read the rcpt_include_affixes from the non-existent transport. Now it
4062 just assumes that the setting of that option is false. This bug was
4063 introduced by 4.51/PH/31.
4065 PH/28 Changed -d+all to exclude +memory, because that information is very
4066 rarely of interest, but it makes the output a lot bigger. People tend to
4067 do -d+all out of habit.
4069 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4070 code in os-type was giving problems when libc.so lives in lib64, like on
4073 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4074 aren't the modern standard, and it seems that some systems' include files
4075 don't always have them. Exim was already checking for some of the newer
4076 ones like T_AAAA, and defining it itself. I've added checks for all the
4077 record types that Exim uses.
4079 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4080 not automatically generating a new one, as it is supposed to. This
4081 prevented TLS from working. If the file did exist, but contained invalid
4082 data, a new version was generated, as expected. It was only the case of a
4083 non-existent file that was broken.
4085 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4086 with a change in libdomainkeys > 0.64.
4088 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4089 from DNS. If the selector record carries the flag, it now has
4090 precedence over the domain-wide flag.
4092 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4094 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4095 the use of a port name that isn't defined in /etc/services) occurred, the
4096 message was deferred as in a normal delivery, and thus remained on the
4097 spool, instead of being failed because of the mua_wrapper setting. This
4098 is now fixed, and I tidied up some of the mua_wrapper messages at the
4101 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4102 the messages in a hash of arrays rather than using individual hashes.
4103 This is a bit cleaner and results in dramatic memory savings, albeit
4104 at a slight CPU cost.
4106 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4107 as requested by Marc Sherman.
4109 SC/10 Eximstats: added histograms for user specified patterns as requested
4112 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4114 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4115 fopen() in the content-scanning modules that did not already have it.
4121 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4122 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4124 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4126 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4128 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4129 to test Sieve filters that use "vacation".
4131 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4132 that changes the way the GnuTLS parameters are stored in the cache file.
4133 The new format can be generated externally. For backward compatibility,
4134 if the data in the cache doesn't make sense, Exim assumes it has read an
4135 old-format file, and it generates new data and writes a new file. This
4136 means that you can't go back to an older release without removing the
4139 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4140 work if there are any delivery delays because "one_time" forces the
4141 parent to be marked "delivered", so its unseen clone is never tried
4142 again. For this reason, Exim now forbids the simultaneous setting of
4145 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4146 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4147 entry if you want to know the details. The fix, however, neglected to
4148 consider the case when local delivery batching is involved. The test for
4149 "previously delivered" was not happening when checking to see if an
4150 address could be batched with a previous (undelivered) one; under
4151 certain circumstances this could lead to multiple deliveries to the same
4154 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4155 in its include files, and this causes problems building Exim.
4157 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4158 header_syntax) but Exim was just ignoring anything given after a slash.
4159 In particular, this caused confusion with an attempt to use "verify =
4160 reverse_host_lookup/defer_ok". An error is now given when options are
4161 supplied for verify items that do not have them. (Maybe reverse_host_
4162 lookup should have a defer_ok option, but that's a different point.)
4164 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4165 defined by RFC 821) to 2048, because there were problems with some AUTH
4166 commands, and RFC 1869 says the size should be increased for extended
4167 SMTP commands that take arguments.
4169 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4172 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4173 "unknown" error; now it says that the functionality isn't in the binary.
4175 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4176 an address' error message when a string expansion fails (syntax or
4177 whatever). Otherwise the password may appear in the log. Following change
4178 PH/42 below, there is no longer a chance of it appearing in a bounce
4181 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4183 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4184 of its addresses was ever tried. (Bugzilla bug #2.)
4186 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4187 the result incorrectly in the debug output. (It correctly added a newline
4188 to what was transported.)
4190 TF/01 Added $received_time.
4192 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4193 commented out examples of how to interface to a virus scanner and to
4194 SpamAssassin. Also added commented examples of av_scanner and
4195 spamd_address settings.
4197 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4198 and controls are allowed in which ACLs. There were a couple of minor
4199 errors. Some of the entries in the conditions table (which is a table of
4200 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4201 negation of where the condition IS allowed.
4203 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4205 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4206 header file does not have a version number, so I've had to invent a new
4207 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4208 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4209 radiusclient), but it was contributed by a Radius user.
4211 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4212 files or directories, for OpenSSL.
4214 PH/19 When an Exim process that is running as root has to create an Exim log
4215 file, it does so in a subprocess that runs as exim:exim so as to get the
4216 ownership right at creation (otherwise, other Exim processes might see
4217 the file with the wrong ownership). There was no test for failure of this
4218 fork() call, which would lead to the process getting stuck as it waited
4219 for a non-existent subprocess. Forks do occasionally fail when resources
4220 run out. I reviewed all the other calls to fork(); they all seem to check
4223 PH/20 When checking for unexpected SMTP input at connect time (before writing
4224 the banner), Exim was not dealing correctly with a non-positive return
4225 from the read() function. If the client had disconnected by this time,
4226 the result was a log entry for a synchronization error with an empty
4227 string after "input=" when read() returned zero. If read() returned -1
4228 (an event I could not check), uninitialized data bytes were printed.
4229 There were reports of junk text (parts of files, etc) appearing after
4232 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4234 PH/22 Added support for macro redefinition, and (re)definition in between
4235 driver and ACL definitions.
4237 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4238 forgetting to use the resulting value; it was using the unexpanded value.
4240 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4241 hadn't been configured. The fix is from Juergen Kreileder, who
4242 understands it better than I do:
4244 "Here's what I see happening with three configured cyrus_sasl
4245 authenticators configured (plain, login, cram-md5):
4247 On startup auth_cyrus_sasl_init() gets called for each of these.
4248 This means three calls to sasl_listmech() without a specified mech_list.
4249 => SASL tests which mechs of all available mechs actually work
4250 => three warnings about OTP not working
4251 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4253 With the patch, sasl_listmech() also gets called three times. But now
4254 SASL's mech_list option is set to the server_mech specified in the the
4255 authenticator. Or in other words, the answer from sasl_listmech()
4256 gets limited to just the mech you're testing for (which is different
4258 => the return list contains just 'plain' or 'login', 'cram-md5' or
4259 nothing depending on the value of ob->server_mech.
4261 I've just tested the patch: Authentication still works fine,
4262 unavailable mechs specified in the exim configuration are still
4263 caught, and the auth.log warnings about OTP are gone."
4265 PH/25 When debugging is enabled, the contents of the command line are added
4266 to the debugging output, even when log_selector=+arguments is not
4269 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4270 answer is "GNU", and only if the return is "GNU/something" is the answer
4273 PH/27 $acl_verify_message is now set immediately after the failure of a
4274 verification in an ACL, and so is available in subsequent modifiers. In
4275 particular, the message can be preserved by coding like this:
4277 warn !verify = sender
4278 set acl_m0 = $acl_verify_message
4280 Previously, $acl_verify_message was set only while expanding "message"
4281 and "log_message" when a very denied access.
4283 PH/28 Modified OS/os.c-Linux with
4285 -#ifndef OS_LOAD_AVERAGE
4286 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4288 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4289 nomenclature these days.)
4291 PH/29 Installed patch from the Sieve maintainer that adds the options
4292 sieve_useraddress and sieve_subaddress to the redirect router.
4294 PH/30 In these circumstances:
4295 . Two addresses routed to the same list of hosts;
4296 . First host does not offer TLS;
4297 . First host accepts first address;
4298 . First host gives temporary error to second address;
4299 . Second host offers TLS and a TLS session is established;
4300 . Second host accepts second address.
4301 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4302 and peerdn, if requested) that were in fact used only for the second
4305 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4306 attention to any local part prefix or suffix that was matched by the
4307 router that accepted the address. It now behaves in the same way as it
4308 does for delivery: the affixes are removed from the local part unless
4309 rcpt_include_affixes is set on the transport.
4311 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4312 timeout during the DATA phase of an incoming message.
4314 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4315 applied a patch sanctioned by the Sieve maintainer.
4317 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4318 the uid or gid is negative. A case of a negative gid caused this to be
4319 noticed. The fix allows for either to be negative.
4321 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4322 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4325 PH/36 The change PH/12 above was broken. Fixed it.
4327 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4328 the grounds that routing the same address twice would always produce the
4329 same answer. This might have been true once, but it is certainly no
4330 longer true now. Routing a child address may depend on the previous
4331 routing that produced that child. Some complicated redirection strategies
4332 went wrong when messages had multiple recipients, and made Exim's
4333 behaviour dependent on the order in which the addresses were given.
4335 I have moved the duplicate checking until after the routing is complete.
4336 Exim scans the addresses that are assigned to local and remote
4337 transports, and removes any duplicates. This means that more work will be
4338 done, as duplicates will always all be routed, but duplicates are
4339 presumably rare, so I don't expect this is of any significance.
4341 For deliveries to pipes, files, and autoreplies, the duplicate checking
4342 still happens during the routing process, since they are not going to be
4345 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4346 It corrects a timeout issue with spamd. This is Ian's comment: "The
4347 background is that sometimes spamd either never reads data from a
4348 connection it has accepted, or it never writes response data. The exiscan
4349 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4350 blindly assumes that writes won't block so it may never time out."
4352 PH/39 Allow G after quota size as well as K and M.
4354 PH/40 The value set for $authenticated_id in an authenticator may not contain
4355 binary zeroes or newlines because the value is written to log lines and
4356 to spool files. There was no check on this. Now the value is run through
4357 the string_printing() function so that such characters are converted to
4358 printable escape sequences.
4360 PH/41 $message_linecount is a new variable that contains the total number of
4361 lines in the message. Compare $body_linecount, which is the count for the
4364 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4365 bounce and delay warning messages, except in certain special cases, which
4368 (a) An SMTP error message from a remote host;
4369 (b) A message specified in a :fail: redirection;
4370 (c) A message specified in a "fail" command in a system filter;
4371 (d) A message specified in a FAIL return from the queryprogram router;
4372 (e) A message specified by the cannot_route_message router option.
4374 In these cases only, Exim does include the error details in bounce and
4375 warning messages. There are also a few cases where bland messages such
4376 as "unrouteable address" or "local delivery error" are given.
4378 PH/43 $value is now also set for the "else" part of a ${run expansion.
4380 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4381 being worked on, but at least Exim now implements the latest version to
4384 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4385 process to complete was treated as a delivery failure, a timeout while
4386 writing the message to the pipe was logged, but erroneously treated as a
4387 successful delivery. Such timeouts include transport filter timeouts. For
4388 consistency with the overall process timeout, these timeouts are now
4389 treated as errors, giving rise to delivery failures by default. However,
4390 there is now a new Boolean option for the pipe transport called
4391 timeout_defer, which, if set TRUE, converts the failures into defers for
4392 both kinds of timeout. A transport filter timeout is now identified in
4395 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4396 systems where "make" and "gmake" are different, calling "gmake" at top
4397 level broke things. I've arranged for the value of $(MAKE) to be passed
4398 from the Makefile to this script so that it can call the same version of
4402 A note about Exim versions 4.44 and 4.50
4403 ----------------------------------------
4405 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4406 changes of various kinds. As a consequence, a big documentation update was
4407 needed. This delayed the release for rather longer than seemed good, especially
4408 in the light of a couple of (minor) security issues. Therefore, the changes
4409 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4410 release. So 4.44 and 4.50 are in effect two different branches that both start
4413 I have left the 4.50 change log unchanged; it contains all the changes since
4414 4.43. The change log for 4.44 is below; many of its items are identical to
4415 those for 4.50. This seems to be the most sensible way to preserve the
4416 historical information.
4422 1. Minor wording change to the doc/README.SIEVE file.
4424 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4425 computation of the current number of files was incorrect.
4427 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4428 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4431 4. Give more explanation in the error message when the command for a transport
4432 filter fails to execute.
4434 5. There are several places where Exim runs a non-Exim command in a
4435 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4436 was being done only for the command run by the queryprogram router. It is
4437 now done for all such subprocesses. The other cases are: ${run, transport
4438 filters, and the commands run by the lmtp and pipe transports.
4440 6. Added CONFIGURE_GROUP build-time option.
4442 7. Some older OS have a limit of 256 on the maximum number of file
4443 descriptors. Exim was using setrlimit() to set 1000 as a large value
4444 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4445 systems. I've change it so that if it can't get 1000, it tries for 256.
4447 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4448 was an oversight, and furthermore, ever since the addition of extra
4449 controls (e.g. 4.43/32), the checks on when to allow different forms of
4450 "control" were broken. There should now be diagnostics for all cases when a
4451 control that does not make sense is encountered.
4453 9. Added the /retain_sender option to "control=submission".
4455 10. $recipients is now available in the predata ACL (oversight).
4457 11. Tidy the search cache before the fork to do a delivery from a message
4458 received from the command line. Otherwise the child will trigger a lookup
4459 failure and thereby defer the delivery if it tries to use (for example) a
4460 cached ldap connection that the parent has called unbind on.
4462 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4463 of $address_data from the recipient verification was clobbered by the
4464 sender verification.
4466 13. The value of address_data from a sender verification is now available in
4467 $sender_address_data in subsequent conditions in the ACL statement.
4469 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4471 15. Added a new option "connect=<time>" to callout options, to set a different
4474 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4475 was its contents. (It was OK if the option was not defined at all.)
4477 17. A "Completed" log line is now written for messages that are removed from
4478 the spool by the -Mrm option.
4480 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4481 information about exactly what failed.
4483 19. Added -dd to debug only the daemon process.
4485 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4486 handles timeouts, both on the server side and network timeouts. Renamed the
4487 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4489 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4490 It is now set to "smtps".
4492 22. $host_address is now set to the target address during the checking of
4493 ignore_target_hosts.
4495 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4496 being passed; this would have caused $sender_host_name to have been used if
4497 matching the list had actually called for a host name (not very likely,
4498 since this list is usually IP addresses). A host name is now passed as
4501 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4502 code that specifies a non-restarting handler (typically sigaction() in
4503 modern systems) in an attempt to fix a rare and obscure crash bug.
4505 25. Narrowed the window for a race in the daemon that could cause it to ignore
4506 SIGCHLD signals. This is not a major problem, because they are used only to
4507 wake it up if nothing else does.
4509 26. A malformed maildirsize file could cause Exim to calculate negative values
4510 for the mailbox size or file count. Odd effects could occur as a result.
4511 The maildirsize information is now recalculated if the size or filecount
4514 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4515 support for a long time. Removed HAVE_SYS_VFS_H.
4517 28. Installed the latest version of exipick from John Jetmore.
4519 29. In an address list, if the pattern was not a regular expression, an empty
4520 subject address (from a bounce message) matched only if the pattern was an
4521 empty string. Non-empty patterns were not even tested. This was the wrong
4522 because it is perfectly reasonable to use an empty address as part of a
4523 database query. An empty address is now tested by patterns that are
4524 lookups. However, all the other forms of pattern expect the subject to
4525 contain a local part and a domain, and therefore, for them, an empty
4526 address still always fails if the pattern is not itself empty.
4528 30. Exim went into a mad DNS loop when attempting to do a callout where the
4529 host was specified on an smtp transport, and looking it up yielded more
4530 than one IP address.
4532 31. Re-factored the code for checking spool and log partition space into a
4533 function that finds that data and another that does the check. The former
4534 is then used to implement four new variables: $spool_space, $log_space,
4535 $spool_inodes, and $log_inodes.
4537 32. The RFC2047 encoding function was originally intended for short strings
4538 such as real names; it was not keeping to the 75-character limit for
4539 encoded words that the RFC imposes. It now respects the limit, and
4540 generates multiple encoded words if necessary. To be on the safe side, I
4541 have increased the buffer size for the ${rfc2047: expansion operator from
4544 33. It is now permitted to omit both strings after an "if" condition; if the
4545 condition is true, the result is "true". As before, when the second string
4546 is omitted, a false condition yields an empty string. This makes it less
4547 cumbersome to write custom ACL and router conditions.
4549 34. Failure to deliver a bounce message always caused it to be frozen, even if
4550 there was an errors_to setting on the router. The errors_to setting is now
4553 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4554 canonical form (fully expanded) before being placed in
4555 $sender_host_address.
4557 36. The table in the code that translates DNS record types into text (T_A to
4558 "A" for instance) was missing entries for NS and CNAME. It is just possible
4559 that this could have caused confusion if both these types were looked up
4560 for the same domain, because the text type is used as part of Exim's
4561 per-process caching. But the chance of anyone hitting this buglet seems
4564 37. The dnsdb lookup has been extended in a number of ways.
4566 (1) There is a new type, "zns", which walks up the domain tree until it
4567 finds some nameserver records. It should be used with care.
4569 (2) There is a new type, "mxh", which is like "mx" except that it returns
4570 just the host names, not the priorities.
4572 (3) It is now possible to give a list of domains (or IP addresses) to be
4573 looked up. The behaviour when one of the lookups defers can be
4574 controlled by a keyword.
4576 (4) It is now possible to specify the separator character for use when
4577 multiple records are returned.
4579 38. The dnslists ACL condition has been extended: it is now possible to supply
4580 a list of IP addresses and/or domains to be looked up in a particular DNS
4583 39. Added log_selector=+queue_time_overall.
4585 40. When running the queue in the test harness, wait just a tad after forking a
4586 delivery process, to get repeatability of debugging output.
4588 41. Include certificate and key file names in error message when GnuTLS fails
4589 to set them up, because the GnuTLS error message doesn't include the name
4590 of the failing file when there is a problem reading it.
4592 42. Allow both -bf and -bF in the same test run.
4594 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4596 44. The "Exiscan patch" is now merged into the mainline Exim source.
4598 45. Sometimes the final signoff response after QUIT could fail to get
4599 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4600 < 0 before doing a fflush(). This bug looks as though it goes back to the
4601 introduction of TLS in release 3.20, but "sometimes" must have been rare
4602 because the tests only now provoked it.
4604 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4605 (this can affect the format of dates).
4607 47. exim_tidydb, when checking for the continued existence of a message for
4608 which it has found a message-specific retry record, was not finding
4609 messages that were in split spool directories. Consequently, it was
4610 deleting retry records that should have stayed in existence.
4612 48. Steve fixed some bugs in eximstats.
4614 49. The SPA authentication driver was not abandoning authentication and moving
4615 on to the next authenticator when an expansion was forced to fail,
4616 contradicting the general specification for all authenticators. Instead it
4617 was generating a temporary error. It now behaves as specified.
4619 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4620 (the order specifies the preference for clients). The order is now AES256,
4621 AES128, 3DES, ARCFOUR128.
4623 51. Small patch to Sieve code - explicitly set From: when generating an
4626 52. Exim crashed if a remote delivery caused a very long error message to be
4627 recorded - for instance if somebody sent an entire SpamAssassin report back
4628 as a large number of 550 error lines. This bug was coincidentally fixed by
4629 increasing the size of one of Exim's internal buffers (big_buffer) that
4630 happened as part of the Exiscan merge. However, to be on the safe side, I
4631 have made the code more robust (and fixed the comments that describe what
4634 53. Now that there can be additional text after "Completed" in log lines (if
4635 the queue_time_overall log selector is set), a one-byte patch to exigrep
4636 was needed to allow it to recognize "Completed" as not the last thing in
4639 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4640 patch that reportedly fixes this has been added. I am not expert enough to
4641 create a test for it. This is what the patch creator wrote:
4643 "I found a little strange behaviour of ldap code when working with
4644 Windows 2003 AD Domain, where users was placed in more than one
4645 Organization Units. When I tried to give exim partial DN, the exit code
4646 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4647 But simultaneously result of request was absolutely normal ldap result,
4648 so I produce this patch..."
4650 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4651 REFERENCE, so I have modified the code to exclude the patch when that macro
4654 55. Some experimental protocols are using DNS PTR records for new purposes. The
4655 keys for these records are domain names, not reversed IP addresses. The
4656 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4657 leaves it alone. Component reversal etc. now happens only for IP addresses.
4660 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4662 57. Double the size of the debug message buffer (to 2048) so that more of very
4663 long debug lines gets shown.
4665 58. The exicyclog utility now does better if the number of log files to keep
4666 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4668 59. Two changes related to the smtp_active_hostname option:
4670 (1) $smtp_active_hostname is now available as a variable.
4671 (2) The default for smtp_banner uses $smtp_active_hostname instead
4672 of $primary_hostname.
4674 60. The host_aton() function is supposed to be passed a string that is known
4675 to be a valid IP address. However, in the case of IPv6 addresses, it was
4676 not checking this. This is a hostage to fortune. Exim now panics and dies
4677 if the condition is not met. A case was found where this could be provoked
4678 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4679 components; fortuitously, this particular loophole had already been fixed
4680 by change 4.50/55 above.
4682 If there are any other similar loopholes, the new check in host_aton()
4683 itself should stop them being exploited. The report I received stated that
4684 data on the command line could provoke the exploit when Exim was running as
4685 exim, but did not say which command line option was involved. All I could
4686 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4687 running as the user.
4690 61. There was a buffer overflow vulnerability in the SPA authentication code
4691 (which came originally from the Samba project). I have added a test to the
4692 spa_base64_to_bits() function which I hope fixes it.
4695 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4696 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4698 63. The daemon start-up calls getloadavg() while still root for those OS that
4699 need the first call to be done as root, but it missed one case: when
4700 deliver_queue_load_max is set with deliver_drop_privilege. This is
4701 necessary for the benefit of the queue runner, because there is no re-exec
4702 when deliver_drop_privilege is set.
4704 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4705 This has been fixed.
4707 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4708 was in use, was not putting the data itself into the right store pool;
4709 consequently, it could be overwritten for a subsequent message in the same
4710 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4713 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4715 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4716 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4717 the function were treating the return as a boolean value, which happened to
4718 work because 0=false and not-0=true, but is not correct code.
4720 68. The host_aton() function was not handling scoped IPv6 addresses (those
4721 with, for example, "%eth0" on the end) correctly.
4723 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4724 negated items (that is, ~something) in unsigned ints. Some compilers
4725 apparently mutter when there is no cast.
4727 70. If an address verification called from an ACL failed, and did not produce a
4728 user-specific message (i.e. there was only a "system" message), nothing was
4729 put in $acl_verify_message. In this situation, it now puts the system
4732 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4733 session; change 4.31/43 added the unwanted input to the log line - except
4734 that it did not do this in the start of session case. It now does.
4736 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4737 This is wrong and can cause the other end to generate a synchronization
4738 error if it is another Exim or anything else that does the synchronization
4739 check. A QUIT command is no longer sent after a timeout.
4741 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4742 during host lookups.
4744 74. The defer_ok option of callout verification was not working if it was used
4745 when verifying addresses in header lines, that is, for this case:
4747 verify = header_sender/callout=defer_ok
4749 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4750 those file descriptors could be used for SMTP connections. If anything
4751 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4752 could be sent to the SMTP client, causing chaos. The daemon now opens
4753 stdin, stdout, and stderr to /dev/null when it puts itself into the
4756 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4757 log by default. The user can override this with suitable Perl magic.
4759 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4760 the log message when discard triggers, was not working for the DATA ACL or
4761 for the non-SMTP ACL.
4763 78. Error message wording change in sieve.c.
4765 79. If smtp_accept_max_per_host was set, the number of connections could be
4766 restricted to fewer than expected, because the daemon was trying to set up
4767 a new connection before checking whether the processes handling previous
4768 connections had finished. The check for completed processes is now done
4769 earlier. On busy systems, this bug wouldn't be noticed because something
4770 else would have woken the daemon, and it would have reaped the completed
4773 80. If a message was submitted locally by a user whose login name contained one
4774 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4775 It caused a spool format error. I have fixed the spool reading code. A
4776 related problem was that the "from" clause in the Received: line became
4777 illegal because of the space(s). It is now covered by ${quote_local_part.
4779 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4782 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4784 83. Patch from Sieve maintainer to fix unterminated string problem in
4785 "vacation" handling.
4787 84. Some minor changes to the Linux configuration files to help with other
4788 OS variants using glibc.
4790 85. One more patch for Sieve to update vacation handling to latest spec.
4793 ----------------------------------------------------
4794 See the note above about the 4.44 and 4.50 releases.
4795 ----------------------------------------------------
4801 1. Change 4.43/35 introduced a bug that caused file counts to be
4802 incorrectly computed when quota_filecount was set in an appendfile
4805 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4806 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4809 3. Give more explanation in the error message when the command for a transport
4810 filter fails to execute.
4812 4. There are several places where Exim runs a non-Exim command in a
4813 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4814 was being done only for the command run by the queryprogram router. It is
4815 now done for all such subprocesses. The other cases are: ${run, transport
4816 filters, and the commands run by the lmtp and pipe transports.
4818 5. Some older OS have a limit of 256 on the maximum number of file
4819 descriptors. Exim was using setrlimit() to set 1000 as a large value
4820 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4821 systems. I've change it so that if it can't get 1000, it tries for 256.
4823 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4824 was an oversight, and furthermore, ever since the addition of extra
4825 controls (e.g. 4.43/32), the checks on when to allow different forms of
4826 "control" were broken. There should now be diagnostics for all cases when a
4827 control that does not make sense is encountered.
4829 7. $recipients is now available in the predata ACL (oversight).
4831 8. Tidy the search cache before the fork to do a delivery from a message
4832 received from the command line. Otherwise the child will trigger a lookup
4833 failure and thereby defer the delivery if it tries to use (for example) a
4834 cached ldap connection that the parent has called unbind on.
4836 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4837 of $address_data from the recipient verification was clobbered by the
4838 sender verification.
4840 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4841 was its contents. (It was OK if the option was not defined at all.)
4843 11. A "Completed" log line is now written for messages that are removed from
4844 the spool by the -Mrm option.
4846 12. $host_address is now set to the target address during the checking of
4847 ignore_target_hosts.
4849 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4850 being passed; this would have caused $sender_host_name to have been used if
4851 matching the list had actually called for a host name (not very likely,
4852 since this list is usually IP addresses). A host name is now passed as
4855 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4856 code that specifies a non-restarting handler (typically sigaction() in
4857 modern systems) in an attempt to fix a rare and obscure crash bug.
4859 15. Narrowed the window for a race in the daemon that could cause it to ignore
4860 SIGCHLD signals. This is not a major problem, because they are used only to
4861 wake it up if nothing else does.
4863 16. A malformed maildirsize file could cause Exim to calculate negative values
4864 for the mailbox size or file count. Odd effects could occur as a result.
4865 The maildirsize information is now recalculated if the size or filecount
4868 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4869 support for a long time. Removed HAVE_SYS_VFS_H.
4871 18. Updated exipick to current release from John Jetmore.
4873 19. Allow an empty sender to be matched against a lookup in an address list.
4874 Previously the only cases considered were a regular expression, or an
4877 20. Exim went into a mad DNS lookup loop when doing a callout where the
4878 host was specified on the transport, if the DNS lookup yielded more than
4881 21. The RFC2047 encoding function was originally intended for short strings
4882 such as real names; it was not keeping to the 75-character limit for
4883 encoded words that the RFC imposes. It now respects the limit, and
4884 generates multiple encoded words if necessary. To be on the safe side, I
4885 have increased the buffer size for the ${rfc2047: expansion operator from
4888 22. Failure to deliver a bounce message always caused it to be frozen, even if
4889 there was an errors_to setting on the router. The errors_to setting is now
4892 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4893 canonical form (fully expanded) before being placed in
4894 $sender_host_address.
4896 24. Updated eximstats to version 1.33
4898 25. Include certificate and key file names in error message when GnuTLS fails
4899 to set them up, because the GnuTLS error message doesn't include the name
4900 of the failing file when there is a problem reading it.
4902 26. Expand error message when OpenSSL has problems setting up cert/key files.
4905 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4906 (this can affect the format of dates).
4908 28. exim_tidydb, when checking for the continued existence of a message for
4909 which it has found a message-specific retry record, was not finding
4910 messages that were in split spool directories. Consequently, it was
4911 deleting retry records that should have stayed in existence.
4913 29. eximstats updated to version 1.35
4914 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4915 1.35 - bugfix such that pie charts by volume are generated correctly
4917 30. The SPA authentication driver was not abandoning authentication and moving
4918 on to the next authenticator when an expansion was forced to fail,
4919 contradicting the general specification for all authenticators. Instead it
4920 was generating a temporary error. It now behaves as specified.
4922 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4923 (the order specifies the preference for clients). The order is now AES256,
4924 AES128, 3DES, ARCFOUR128.
4926 31. Small patch to Sieve code - explicitly set From: when generating an
4929 32. Exim crashed if a remote delivery caused a very long error message to be
4930 recorded - for instance if somebody sent an entire SpamAssassin report back
4931 as a large number of 550 error lines. This bug was coincidentally fixed by
4932 increasing the size of one of Exim's internal buffers (big_buffer) that
4933 happened as part of the Exiscan merge. However, to be on the safe side, I
4934 have made the code more robust (and fixed the comments that describe what
4937 33. Some experimental protocols are using DNS PTR records for new purposes. The
4938 keys for these records are domain names, not reversed IP addresses. The
4939 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4940 leaves it alone. Component reversal etc. now happens only for IP addresses.
4943 34. The host_aton() function is supposed to be passed a string that is known
4944 to be a valid IP address. However, in the case of IPv6 addresses, it was
4945 not checking this. This is a hostage to fortune. Exim now panics and dies
4946 if the condition is not met. A case was found where this could be provoked
4947 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4948 components; fortuitously, this particular loophole had already been fixed
4949 by change 4.50/55 or 4.44/33 above.
4951 If there are any other similar loopholes, the new check in host_aton()
4952 itself should stop them being exploited. The report I received stated that
4953 data on the command line could provoke the exploit when Exim was running as
4954 exim, but did not say which command line option was involved. All I could
4955 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4956 running as the user.
4959 35. There was a buffer overflow vulnerability in the SPA authentication code
4960 (which came originally from the Samba project). I have added a test to the
4961 spa_base64_to_bits() function which I hope fixes it.
4964 36. The daemon start-up calls getloadavg() while still root for those OS that
4965 need the first call to be done as root, but it missed one case: when
4966 deliver_queue_load_max is set with deliver_drop_privilege. This is
4967 necessary for the benefit of the queue runner, because there is no re-exec
4968 when deliver_drop_privilege is set.
4970 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4971 was in use, was not putting the data itself into the right store pool;
4972 consequently, it could be overwritten for a subsequent message in the same
4973 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4976 38. Sometimes the final signoff response after QUIT could fail to get
4977 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4978 < 0 before doing a fflush(). This bug looks as though it goes back to the
4979 introduction of TLS in release 3.20, but "sometimes" must have been rare
4980 because the tests only now provoked it.
4986 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4987 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4988 Now it returns an integer. A number of calls were still expecting a T/F
4989 return. Fortuitously, in all cases, the tests worked in OK situations,
4990 which is the norm. However, things would have gone wrong on any write
4991 failures on the smtp file descriptor. This function is used when sending
4992 messages over SMTP and also when doing verify callouts.
4994 2. When Exim is called to do synchronous delivery of a locally submitted
4995 message (the -odf or -odi options), it no longer closes stderr before doing
4998 3. Implemented the mua_wrapper option.
5000 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5002 5. Implemented the functions header_remove(), header_testname(),
5003 header_add_at_position(), and receive_remove_recipient(), and exported them
5006 6. If an ACL "warn" statement specified the addition of headers, Exim already
5007 inserted X-ACL-Warn: at the start if there was no header name. However, it
5008 was not making this test for the second and subsequent header lines if
5009 there were newlines in the string. This meant that an invalid header could
5010 be inserted if Exim was badly configured.
5012 7. Allow an ACL "warn" statement to add header lines at the start or after all
5013 the Received: headers, as well as at the end.
5015 8. Added the rcpt_4xx retry error code.
5017 9. Added postmaster_mailfrom=xxx to callout verification option.
5019 10. Added mailfrom=xxxx to the callout verification option, for verify=
5022 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5023 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5025 12. Inserted some casts to stop certain compilers warning when using pointer
5026 differences as field lengths or precisions in printf-type calls (mostly
5027 affecting debugging statements).
5029 13. Added optional readline() support for -be (dynamically loaded).
5031 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5032 same clock tick as a message's arrival, so that its received time was the
5033 same as the "first fail" time on the retry record, and that message
5034 remained on the queue past the ultimate address timeout, every queue runner
5035 would try a delivery (because it was past the ultimate address timeout) but
5036 after another failure, the ultimate address timeout, which should have then
5037 bounced the address, did not kick in. This was a "< instead of <=" error;
5038 in most cases the first failure would have been in the next clock tick
5039 after the received time, and all would be well.
5041 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5042 being recognized when the domain list was tested by the match_domain
5043 condition in an expansion string.
5045 16. Added the ${str2b64: operator.
5047 17. Exim was always calling setrlimit() to set a large limit for the number of
5048 processes, without checking whether the existing limit was already
5049 adequate. (It did check for the limit on file descriptors.) Furthermore,
5050 errors from getrlimit() and setrlimit() were being ignored. Now they are
5051 logged to the main and panic logs, but Exim does carry on, to try to do its
5052 job under whatever limits there are.
5054 18. Imported PCRE 5.0.
5056 19. Trivial typo in log message " temporarily refused connection" (the leading
5059 20. If the log selector return_path_on_delivery was set and an address was
5060 redirected to /dev/null, the delivery process crashed because it assumed
5061 that a return path would always be set for a "successful" delivery. In this
5062 case, the whole delivery is bypassed as an optimization, and therefore no
5065 21. Internal re-arrangement: the function for sending a challenge and reading
5066 a response while authentication was assuming a zero-terminated challenge
5067 string. It's now changed to take a pointer and a length, to allow for
5068 binary data in such strings.
5070 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5072 23. Exim was not respecting finduser_retries when seeking the login of the
5073 uid under which it was called; it was always trying 10 times. (The default
5074 setting of finduser_retries is zero.) Also, it was sleeping after the final
5075 failure, which is pointless.
5077 24. Implemented tls_on_connect_ports.
5079 25. Implemented acl_smtp_predata.
5081 26. If the domain in control=submission is set empty, Exim assumes that the
5082 authenticated id is a complete email address when it generates From: or
5083 Sender: header lines.
5085 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5086 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5087 chown and chgrp in /bin and hostname in /usr/bin.
5089 28. Exim was keeping the "process log" file open after each use, just as it
5090 does for the main log. This opens the possibility of it remaining open for
5091 long periods when the USR1 signal hits a daemon. Occasional processlog
5092 errors were reported, that could have been caused by this. Anyway, it seems
5093 much more sensible not to leave this file open at all, so that is what now
5096 29. The long-running daemon process does not normally write to the log once it
5097 has entered its main loop, and it closes the log before doing so. This is
5098 so that log files can straightforwardly be renamed and moved. However,
5099 there are a couple of unusual error situations where the daemon does write
5100 log entries, and I had neglected to close the log afterwards.
5102 30. The text of an SMTP error response that was received during a remote
5103 delivery was being truncated at 512 bytes. This is too short for some of
5104 the long messages that one sometimes sees. I've increased the limit to
5107 31. It is now possible to make retry rules that apply only when a message has a
5108 specific sender, in particular, an empty sender.
5110 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5111 it possible to be selective about when SMTP synchronization is enforced.
5113 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5115 32. Implemented hosts_connection_nolog.
5117 33. Added an ACL for QUIT.
5119 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5122 35. Added mailbox_size and mailbox_filecount to appendfile.
5124 36. Added control = no_multiline_responses to ACLs.
5126 37. There was a bug in the logic of the code that waits for the clock to tick
5127 in the case where the clock went backwards by a substantial amount such
5128 that the microsecond fraction of "now" was more than the microsecond
5129 fraction of "then" (but the whole seconds number was less).
5131 38. Added support for the libradius Radius client library this is found on
5132 FreeBSD (previously only the radiusclient library was supported).
5138 1. When certain lookups returned multiple values in the form name=value, the
5139 quoting of the values was not always being done properly. Specifically:
5140 (a) If the value started with a double quote, but contained no whitespace,
5142 (b) If the value contained whitespace other than a space character (i.e.
5143 tabs or newlines or carriage returns) it was not quoted.
5144 This fix has been applied to the mysql and pgsql lookups by writing a
5145 separate quoting function and calling it from the lookup code. The fix
5146 should probably also be applied to nisplus, ibase and oracle lookups, but
5147 since I cannot test any of those, I have not disturbed their existing code.
5149 2. A hit in the callout cache for a specific address caused a log line with no
5150 reason for rejecting RCPT. Now it says "Previous (cached) callout
5151 verification failure".
5153 3. There was an off-by-one bug in the queryprogram router. An over-long
5154 return line was truncated at 256 instead of 255 characters, thereby
5155 overflowing its buffer with the terminating zero. As well as fixing this, I
5156 have increased the buffer size to 1024 (and made a note to document this).
5158 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5159 when Exim is waiting for an SMTP response from a remote server, Exim
5160 restarts its select() call on the socket, thereby resetting its timeout.
5161 This is not a problem when such interrupts are rare. Somebody set up a cron
5162 job to run exiwhat every 2 minutes, which is less than the normal select()
5163 timeout (5 or 10 minutes). This meant that the select() timeout never
5164 kicked in because it was always reset. I have fixed this by comparing the
5165 time when an interrupt arrives with the time at the start of the first call
5166 to select(). If more time than the timeout has elapsed, the interrupt is
5167 treated as a timeout.
5169 5. Some internal re-factoring in preparation for the addition of Sieve
5170 extensions (by MH). In particular, the "personal" test is moved to a
5171 separate function, and given an option for scanning Cc: and Bcc: (which is
5172 not set for Exim filters).
5174 6. When Exim created an email address using the login of the caller as the
5175 local part (e.g. when creating a From: or Sender: header line), it was not
5176 quoting the local part when it contained special characters such as @.
5178 7. Installed new OpenBSD configuration files.
5180 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5181 try to make them clearer.
5183 9. Callout options, other than the timeout value, were being ignored when
5184 verifying sender addresses in header lines. For example, when using
5186 verify = header_sender/callout=no_cache
5188 the cache was (incorrectly) being used.
5190 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5191 only the Cygwin environment.
5193 11. When return_path_on_delivery was set as a log selector, if different remote
5194 addresses in the same message used different return paths and parallel
5195 remote delivery occurred, the wrong values would sometimes be logged.
5196 (Whenever a remote delivery process finished, the return path value from
5197 the most recently started remote delivery process was logged.)
5199 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5200 lines when AUTH and/or TLS are in use. This is the "received protocol"
5201 field. Exim used to use "asmtp" for authenticated SMTP, without any
5202 indication (in the protocol name) for TLS use. Now it follows the RFC and
5203 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5204 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5205 names appear in log lines as well as in Received: header lines.
5207 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5208 extensions, and comparison tests, and to fix some bugs.
5210 14. Changes to the "personal" filter test:
5212 (1) The test was buggy in that it was just doing the equivalent of
5213 "contains" tests on header lines. For example, if a user's address was
5214 anne@some.where, the "personal" test would incorrectly be true for
5216 To: susanne@some.where
5218 This test is now done by extracting each address from the header in turn,
5219 and checking the entire address. Other tests that are part of "personal"
5220 are now done using regular expressions (for example, to check local parts
5221 of addresses in From: header lines).
5223 (2) The list of non-personal local parts in From: addresses has been
5224 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5225 taken from the Sieve specification recommendations.
5227 (3) If the message contains any header line starting with "List-" it is
5228 treated as non-personal.
5230 (4) The test for "circular" in the Subject: header line has been removed
5231 because it now seems ill-conceived.
5233 15. Minor typos in src/EDITME comments corrected.
5235 16. Installed latest exipick from John Jetmore.
5237 17. If headers_add on a router specified a text string that was too long for
5238 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5239 of string_sprintf() is now avoided.
5241 18. $message_body_size was not set (it was always zero) when running the DATA
5242 ACL and the local_scan() function.
5244 19. For the "mail" command in an Exim filter, no default was being set for
5245 the once_repeat time, causing a random time value to be used if "once" was
5246 specified. (If the value happened to be <= 0, no repeat happened.) The
5247 default is now 0s, meaning "never repeat". The "vacation" command was OK
5248 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5249 (I found it when inspecting the code).
5251 20. There is now an overall timeout for performing a callout verification. It
5252 defaults to 4 times the callout timeout, which applies to individual SMTP
5253 commands during the callout. The overall timeout applies when there is more
5254 than one host that can be tried. The timeout is checked before trying the
5255 next host. This prevents very long delays if there are a large number of
5256 hosts and all are timing out (e.g. when the network connections are timing
5257 out). The value of the overall timeout can be changed by specifying an
5258 additional sub-option for "callout", called "maxwait". For example:
5260 verify = sender/callout=5s,maxwait=20s
5262 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5263 to the end before writing, but this should make it even safer).
5265 22. Exim was forgetting that it had advertised PIPELINING for the second and
5266 subsequent messages on an SMTP connection. It was also not resetting its
5267 memory on STARTTLS and an internal HELO.
5269 23. When Exim logs an SMTP synchronization error within a session, it now
5270 records whether PIPELINING has been advertised or not.
5272 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5273 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5274 rather than long int.
5276 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5278 26. Added the never_mail option to autoreply.
5284 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5285 crash if the getsockname() call failed; this can happen if a connection is
5286 closed very soon after it is established. The problem was simply in the
5287 order in which certain operations were done, causing Exim to try to write
5288 to the SMTP stream before it had set up the file descriptor. The bug has
5289 been fixed by making things happen in the correct order.
5295 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5296 before the connection was closed, thus losing the rejection response.
5298 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5299 some early Solaris releases, but causes trouble in current releases where
5300 socklen_t is defined.
5302 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5305 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5308 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5309 that releasing the top of it at the end releases what was used for sub-
5310 expansions (unless the block got too big). However, discard this block if
5311 the first thing is a variable or header, so that we can use its block when
5312 it is dynamic (useful for very large $message_headers, for example).
5314 6. Lookups now cache *every* query, not just the most recent. A new, separate
5315 store pool is used for this. It can be recovered when all lookup caches are
5316 flushed. Lookups now release memory at the end of their result strings.
5317 This has involved some general refactoring of the lookup sources.
5319 7. Some code has been added to the store_xxx() functions to reduce the amount
5320 of flapping under certain conditions.
5322 8. log_incoming_interface used to affect only the <= reception log lines. Now
5323 it causes the local interface and port to be added to several more SMTP log
5324 lines, for example "SMTP connection from", and rejection lines.
5326 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5328 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5330 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5331 could be overwritten at the end of the current message (or the start of a
5332 new message if it was set in a HELO ACL). The value is now preserved for
5333 the duration of the SMTP connection.
5335 12. If a transport had a headers_rewrite setting, and a matching header line
5336 contained an unqualified address, that address was qualified, even if it
5337 did not match any rewriting rules. The underlying bug was that the values
5338 of the flags that permit the existence of unqualified sender and recipient
5339 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5340 non-local messages, and by -bnq for local messages) were not being
5341 preserved with the message after it was received.
5343 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5344 "next input=" as part of the text comprising the host identity instead of
5345 the correct text. The code was using the same buffer for two different
5346 strings. However, depending on which order the printing function evaluated
5347 its arguments, the bug did not always show up. Under Linux, for example, my
5348 test suite worked just fine.
5350 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5351 doesn't work with some older versions of Perl. It has been changed to "my",
5352 which in any case is probably the better facility to use.
5354 15. A really picky compiler found some instances of statements for creating
5355 error messages that either had too many or two few arguments for the format
5358 16. The size of the buffer for calls to the DNS resolver has been increased
5359 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5360 for addresses that have a lot of PTR records. This alleviates a problem; it
5361 does not fully solve it.
5363 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5364 buffer now truncates the list and logs the incident, which is the same
5365 action as happens when Exim is looking up a host name and its aliases.
5366 Previously in this situation something unpredictable would happen;
5367 sometimes it was "internal error: store_reset failed".
5369 18. If a server dropped the connection unexpectedly when an Exim client was
5370 using GnuTLS and trying to read a response, the client delivery process
5371 crashed while trying to generate an error log message.
5373 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5374 string, for example:
5376 warn message = H1: something\nH2: something
5378 the text was added as a single header line from Exim's point of view
5379 though it ended up OK in the delivered message. However, searching for the
5380 second and subsequent header lines using $h_h2: did not work. This has been
5381 fixed. Similarly, if a system filter added multiple headers in this way,
5382 the routers could not see them.
5384 20. Expanded the error message when iplsearch is called with an invalid key to
5385 suggest using net-iplsearch in a host list.
5387 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5388 ACLs are no longer actually imposed (and a message to that effect is
5391 22. If a "gecos" field in a passwd entry contained escaped characters, in
5392 particular, if it contained a \" sequence, Exim got it wrong when building
5393 a From: or a Sender: header from that name. A second bug also caused
5394 incorrect handling when an unquoted " was present following a character
5395 that needed quoting.
5397 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5398 was not being matched caselessly.
5400 24. Arranged for all hyphens in the exim.8 source to be escaped with
5403 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5404 properly. Recipient callout cache records were still being keyed to include
5405 the sender, even when use_sender was set false. This led to far more
5406 callouts that were necessary. The sender is no longer included in the key
5407 when use_sender is false.
5409 26. Added "control = submission" modifier to ACLs.
5411 27. Added the ${base62d: operator to decode base 62 numbers.
5413 28. dnsdb lookups can now access SRV records.
5415 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5416 the configuration file.
5418 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5419 (-v) mode. This makes the output for a verbose queue run more intelligible.
5421 31. Added a use_postmaster feature to recipient callouts.
5423 32. Added the $body_zerocount variable, containing the number of binary zero
5424 bytes in the message body.
5426 33. The time of last modification of the "new" subdirectory is now used as the
5427 "mailbox time last read" when there is a quota error for a maildir
5430 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5432 35. Added +ignore_unknown as a special item in host lists.
5434 36. Code for decoding IPv6 addresses in host lists is now included, even if
5435 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5436 address was recognized as an IP address, but was then not correctly decoded
5437 into binary, causing unexpected and incorrect effects when compared with
5444 1. Very minor rewording of debugging text in manualroute to say "list of
5445 hosts" instead of "hostlist".
5447 2. If verify=header_syntax was set, and a header line with an unqualified
5448 address (no domain) and a large number of spaces between the end of the
5449 name and the colon was received, the reception process suffered a buffer
5450 overflow, and (when I tested it) crashed. This was caused by some obsolete
5451 code that should have been removed. The fix is to remove it!
5453 3. When running in the test harness, delay a bit after writing a bounce
5454 message to get a bit more predictability in the log output.
5456 4. Added a call to search_tidyup() just before forking a reception process. In
5457 theory, someone could use a lookup in the expansion of smtp_accept_max_
5458 per_host which, without the tidyup, could leave open a database connection.
5460 5. Added the variables $recipient_data and $sender_data which get set from a
5461 lookup success in an ACL "recipients" or "senders" condition, or a router
5462 "senders" option, similar to $domain_data and $local_part_data.
5464 6. Moved the writing of debug_print from before to after the "senders" test
5467 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5468 problems for message scanning, either using a data ACL, or using
5469 local_scan() because the Received: header was not generated till after they
5470 were called (in order to set the time as the time of reception completion).
5471 I have revised the way this works. The header is now generated after the
5472 body is received, but before the ACL or local_scan() are called. After they
5473 are run, the timestamp in the header is updated.
5479 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5480 before starting a queue runner without re-exec. This happened only when
5481 deliver_drop_privilege was set or when the Exim user was set to root. The
5482 effect of the bug was that timeouts during subsequent deliveries caused
5483 crashes instead of being properly handled. The handler is now left at its
5484 default (and expected) setting.
5486 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5487 message, again when deliver_drop_privilege is set or Exim is run as root.
5488 The bug described in (1) was not present in this case, but the tidying up
5489 of the other signals was missing. I have made the two cases consistent.
5491 3. The ignore_target_hosts setting on a manualroute router was being ignored
5492 for hosts that were looked up using the /MX notation.
5494 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5497 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5498 operated on the sender address. After changing the $sender_address to <>
5499 for the sender address verify, Exim was re-instated it as the original
5500 (before rewriting) address, but remembering that it had rewritten it, so it
5501 wasn't rewriting it again. This bug also had the effect of breaking the
5502 sender address verification caching when the sender address was rewritten.
5504 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5505 This has been changed so that if the ip literal address matches
5506 ignore_target_hosts, the router declines.
5508 7. Added expansion conditions match_domain, match_address, and match_local_
5509 part (NOT match_host).
5511 8. The placeholder for the Received: header didn't have a length field set.
5513 9. Added code to Exim itself and to exim_lock to test for a specific race
5514 condition that could lead to file corruption when using MBX delivery. The
5515 issue is with the lockfile that is created in /tmp. If this file is removed
5516 after a process has opened it but before that process has acquired a lock,
5517 there is the potential for a second process to recreate the file and also
5518 acquire a lock. This could lead to two Exim processes writing to the file
5519 at the same time. The added code performs the same test as UW imapd; it
5520 checks after acquiring the lock that its file descriptor still refers to
5521 the same named file.
5523 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5524 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5527 11. Added the smtp_active_hostname option. If used, this will typically be made
5528 to depend on the incoming interface address. Because $interface_address is
5529 not set up until the daemon has forked a reception process, error responses
5530 that can happen earlier (such as "too many connections") no longer contain
5533 12. If an expansion in a condition on a "warn" statement fails because a lookup
5534 defers, the "warn" statement is abandoned, and the next ACL statement is
5535 processed. Previously this caused the whole ACL to be aborted.
5537 13. Added the iplsearch lookup type.
5539 14. Added ident_timeout as a log selector.
5541 15. Added tls_certificate_verified as a log selector.
5543 16. Added a global option tls_require_ciphers (compare the smtp transport
5544 option of the same name). This controls incoming TLS connections.
5546 17. I finally figured out how to make tls_require_ciphers do a similar thing
5547 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5548 before starting the TLS session.
5550 18. Tabs are now shown as \t in -bP output.
5552 19. If the log selector return_path_on_delivery was set, Exim crashed when
5553 bouncing a message because it had too many Received: header lines.
5555 20. If two routers both had headers_remove settings, and the first one included
5556 a superfluous trailing colon, the final name in the first list and the
5557 first name in the second list were incorrectly joined into one item (with a
5558 colon in the middle).
5564 1. Added -C and -D options to the exinext utility, mainly to make it easier
5565 to include in the automated testing, but these could be helpful when
5566 multiple configurations are in use.
5568 2. The exinext utility was not formatting the output nicely when there was
5569 an alternate port involved in the retry record key, nor when there was a
5570 message id as well (for retries that were specific to a specific message
5571 and a specific host). It was also confused by IPv6 addresses, because of
5572 the additional colons they contain. I have fixed the IPv4 problem, and
5573 patched it up to do a reasonable job for IPv6.
5575 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5576 delivery, the log line now contains "pipelined" if PIPELINING was used.
5578 4. An SMTP transport process used to panic and die if the bind() call to set
5579 an explicit outgoing interface failed. This has been changed; it is now
5580 treated in the same way as a connect() failure.
5582 5. A reference to $sender_host_name in the part of a conditional expansion
5583 that was being skipped was still causing a DNS lookup. This no longer
5586 6. The def: expansion condition was not recognizing references to header lines
5587 that used bh_ and bheader_.
5589 7. Added the _cache feature to named lists.
5591 8. The code for checking quota_filecount in the appendfile transport was
5592 allowing one more file than it should have been.
5594 9. For compatibility with Sendmail, the command line option
5602 and sets the incoming protocol and host name (for trusted callers). The
5603 host name and its colon can be omitted when only the protocol is to be set.
5604 Note the Exim already has two private options, -pd and -ps, that refer to
5605 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5606 "s", but I don't think that's a major issue.
5608 10. A number of refactoring changes to the code, none of which should affect
5611 (a) The number of logging options was getting close to filling up the
5612 32-bit word that was used as a bit map. I have split them into two classes:
5613 those that are passed in the argument to log_write(), and those that are
5614 only ever tested independently outside of that function. These are now in
5615 separate 32-bit words, so there is plenty of room for expansion again.
5616 There is no change in the user interface or the logging behaviour.
5618 (b) When building, for example, log lines, the code previously used a
5619 macro that called string_cat() twice, in order to add two strings. This is
5620 not really sufficiently general. Furthermore, there was one instance where
5621 it was actually wrong because one of the argument was used twice, and in
5622 one call a function was used. (As it happened, calling the function twice
5623 did not affect the overall behaviour.) The macro has been replaced by a
5624 function that can join an arbitrary number of extra strings onto a growing
5627 (c) The code for expansion conditions now uses a table and a binary chop
5628 instead of a serial search (which was left over from when there were very
5629 few conditions). Also, it now recognizes conditions like "pam" even when
5630 the relevant support is not compiled in: a suitably worded error message is
5631 given if an attempt is made to use such a condition.
5633 11. Added ${time_interval:xxxxx}.
5635 12. A bug was causing one of the ddress fields not to be passed back correctly
5636 from remote delivery subprocesses. The field in question was not being
5637 subsequently used, so this caused to problems in practice.
5639 13. Added new log selectors queue_time and deliver_time.
5641 14. Might have fixed a bug in maildirsizefile handling that threw up
5642 "unexpected character" debug warnings, and recalculated the data
5643 unnecessarily. In any case, I expanded the warning message to give more
5646 15. Added the message "Restricted characters in address" to the statements in
5647 the default ACL that block characters like @ and % in local parts.
5649 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5650 Three changes have been made:
5652 (a) There was a serious bug; a negative response to MAIL caused the whole
5653 recipient domain to be cached as invalid, thereby blocking all messages
5654 to all local parts at the same domain, from all senders. This bug has
5655 been fixed. The domain is no longer cached after a negative response to
5656 MAIL if the sender used is not empty.
5658 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5661 (c) A new callout option, "use_sender" has been added for people who want
5662 the modified behaviour.
5668 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5671 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5672 indeed breaks things for older releases.
5674 3. Added additional logging to the case where there is a problem reading data
5675 from a filter that is running in a subprocess using a pipe, in order to
5676 try to track down a specific problem.
5678 4. Testing facility fudge: when running in the test harness and attempting
5679 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5680 getting "No route to host". Convert this to a timeout.
5682 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5685 6. Some OS don't have socklen_t but use size_t instead. This affects the
5686 fifth argument of getsockopt() amongst other things. This is now
5687 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5688 can be set for individual OS. I have set it for SunOS5, OSF1, and
5689 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5690 some earlier ones do not.
5692 7. Change 4.30/15 was not doing the test caselessly.
5694 8. The standard form for an IPv6 address literal was being rejected by address
5695 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5696 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5697 this, as well as the form without the "IPv6" on the front (but only when
5698 address literals are enabled, of course).
5700 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5702 10. Exim crashed if a message with an empty sender address specified by -f
5703 encountered a router with an errors_to setting. This could be provoked only
5704 by a command such as
5708 where an empty string was supplied; "<>" did not hit this bug.
5710 11. Installed PCRE release 4.5.
5712 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5713 remained set. It is now erased.
5715 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5716 times from message ids (which are base 36 rather than the normal 62).
5718 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5719 were being counted as actual protocol errors, and logged if the log
5720 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5721 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5722 and DATA following a set of rejected RCPTs do not count as protocol errors.
5723 In other words, Exim assumes they were pipelined, though this may not
5724 actually be the case. Of course, in all cases the client gets an
5725 appropriate error code.
5727 15. If a lookup fails in an ACL condition, a message about the failure may
5728 be available; it is used if testing the ACL cannot continue, because most
5729 such messages specify what the cause of the deferral is. However, some
5730 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5731 that caused an old message to be retained and used if a later statement
5732 caused a defer, replacing the real cause of the deferral.
5734 16. If an IP address had so many PTR records that the DNS lookup buffer
5735 was not large enough to hold them, Exim could crash while trying to process
5736 the truncated data. It now detects and logs this case.
5738 17. Further to 4.21/58, another change has been made: if (and only if) the
5739 first line of a message (the first header line) ends with CRLF, a bare LF
5740 in a subsequent header line has a space inserted after it, so as not to
5741 terminate the header.
5743 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5744 unnecessarily, used atoi() instead of strtol(), and didn't check the
5745 termination when getting file sizes from file names by regex.
5747 19. Completely re-implemented the support for maildirsize files, in the light
5748 of a number of problems with the previous contributed implementation
5749 (4.30/29). In particular:
5751 . If the quota is zero, the maildirsize file is maintained, but no quota is
5754 . If the maildir directory does not exist, it is created before any attempt
5755 to write a maildirsize file.
5757 . The quota value in the file is just a cache; if the quota is changed in
5758 the transport, the new value overrides.
5760 . A regular expression is available for excluding directories from the
5763 20. The autoreply transport checks the characters in options that define the
5764 message's headers; it allows continued headers, but it was checking with
5765 isspace() after an embedded newline instead of explicitly looking for a
5768 21. If all the "regular" hosts to which an address was routed had passed their
5769 expiry times, and had not reached their retry times, the address was
5770 bounced, even if fallback hosts were defined. Now Exim should go on to try
5773 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5774 equivalent code in the SMTP transport. Some hosts send humungous responses
5775 to HELO/EHLO, more than 1024 it seems.
5777 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5778 gives compiler warnings in some environments. I've now done it "properly",
5781 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5782 (because of problems with the built-in one) was declared to return uschar *
5783 instead of char *, causing compiler failure.
5785 25. Fixed a file descriptor leak when processing alias/forward files.
5787 26. Fixed a minor format string issue in dbfn.c.
5789 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5791 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5792 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5795 29. When a sender address is verified, it is cached, to save repeating the test
5796 when there is more than one recipient in a message. However, when the
5797 verification involves a callout, it is possible for different callout
5798 options to be set for different recipients. It is too complicated to keep
5799 track of this in the cache, so now Exim always runs a verification when a
5800 callout is required, relying on the callout cache for the optimization.
5801 The overhead is duplication of the address routing, but this should not be
5804 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5805 to be verified with callout=postmaster, and the main callout worked but the
5806 postmaster check failed, the verification correctly failed. However, if a
5807 subsequent RCPT command asked for sender verification *without* the
5808 postmaster check, incorrect caching caused this verification also to fail,
5811 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5812 it was not caching the DNS options (qualify_single, search_parents) that
5813 were used when the lookup failed. A subsequent lookup with different
5814 options therefore always gave the same answer, though there were cases
5815 where it should not have. (Example: a "domains = !$mx_any" option on a
5816 dnslookup router: the "domains" option is always processed without any
5817 widening, but the router might have qualify_single set.) Now Exim uses the
5818 cached value only when the same options are set.
5820 32. Added John Jetmore's "exipick" utility to the distribution.
5822 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5823 than a timeout (e.g. a certificate is required, and is not provided), an
5824 Exim server now closes the connection immediately. Previously it waited for
5825 the client to close - but if the client is SSL, it seems that they each
5826 wait for each other, leading to a delay before one of them times out.
5828 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5829 maintained 0.8.x compatibility because I don't think many are using it, and
5830 it is clearly obsolete.
5832 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5835 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5836 client certificate was expired. A simple patch fixes this, though I don't
5837 understand the full logic of why the verify callback is called multiple
5840 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5841 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5842 which causes problems with some clients (such as the Certicom SSL Plus
5843 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5844 disables the coutermeasure allowing Eudora to connect."
5846 38. Exim was not checking that a write() to a log file succeeded. This could
5847 lead to Bad Things if a log got too big, in particular if it hit a file
5848 size limit. Exim now panics and dies if it cannot write to a log file, just
5849 as it does if it cannot open a log file.
5851 39. Modified OS/Makefile-Linux so that it now contains
5853 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5855 The two -D definitions ensure that Exim is compiled with large file
5856 support, which makes it possible to handle log files that are bigger than
5859 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5860 instance) a domain was checked against a named list that involved a lookup,
5861 causing $domain_data to be set, then another domain was checked against the
5862 same list, then the first domain was re-checked, the value of $domain_data
5863 after the final check could be wrong. In particular, if the second check
5864 failed, it could be set empty. This bug probably also applied to
5867 41. The strip_trailing_dot option was not being applied to the address given
5868 with the -f command-line option.
5870 42. The code for reading a message's header from the spool was incrementing
5871 $received_count, but never initializing it. This meant that the value was
5872 incorrect (doubled) while delivering a message in the same process in which
5873 it was received. In the most common configuration of Exim, this never
5874 happens - a fresh exec is done - but it can happen when
5875 deliver_drop_privilege is set.
5877 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5878 it now includes up to 150 characters of the unexpected data in the log
5881 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5882 and building data strings. The size of both of these buffers was 10 000
5883 bytes - far larger than anybody would *ever* want, thought I. Needless to
5884 say, somebody hit the limit. I have increased the maximum line length to
5885 20 000 and the maximum data length of concatenated lines to 100 000. I have
5886 also fixed two bugs, because there was no checking on these buffers. Tsk,
5887 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5888 buffer is too small.
5890 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5891 lsearch lookups. Now it does.
5893 46. When parsing a route_list item in a manualroute router, a fixed-length
5894 buffer was used for the list of hosts. I made this 1024 bytes long,
5895 thinking that nobody would ever have a list of hosts that long. Wrong.
5896 Somebody had a whole pile of complicated expansion conditions, and the
5897 string was silently truncated, leading to an expansion error. It turns out
5898 that it is easier to change to an unlimited length (owing to other changes
5899 that have happened since this code was originally written) than to build
5900 structure for giving a limitation error. The length of the item that
5901 expands into the list of hosts is now unlimited.
5903 47. The lsearch lookup could not handle data where the length of text line was
5904 more than 4095 characters. Such lines were truncated, leading to shortened
5905 data being returned. It should now handle lines of any length.
5907 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5908 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5911 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5912 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5913 no longer happens in comment lines. A list of replacements is now placed
5914 at the head of all of the source files, except those whose only change is
5915 to replace PERL_COMMAND in the very first #! line.
5917 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5918 messages on the queue, with a bottom-up merge sort, using code contributed
5919 by Michael Haardt. This should make operations like -bp somewhat faster on
5920 large queues. It won't affect queue runners, except when queue_run_in_order
5923 51. Installed eximstats 1.31 in the distribution.
5925 52. Added support for SRV lookups to the dnslookup router.
5927 53. If an ACL referred to $message_body or $message_body_end, the value was not
5928 reset for any messages that followed in the same SMTP session.
5930 54. The store-handling optimization for building very long strings was not
5931 differentiating between the different store pools. I don't think this
5932 actually made any difference in practice, but I've tidied it.
5934 55. While running the routers to verify a sender address, $sender_address
5935 was still set to the sender address. This is wrong, because when routing to
5936 send a bounce to the sender, it would be empty. Therefore, I have changed
5937 it so that, while verifying a sender address, $sender_address is set to <>.
5938 (There is no change to what happens when verifying a recipient address.)
5940 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5941 target A or AAAA records (if not already returned) without resetting the
5942 qualify_single or search_parents options of the DNS resolver. These are
5943 inappropriate in this case because the targets of MX and SRV records must
5944 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5945 target that, when qualified, matched something in the local domain. These
5946 two options are now turned off when doing these lookups.
5948 57. It seems that at least some releases of Reiserfs (which does not have the
5949 concept of a fixed number of inodes) returns zero and not -1 for the
5950 number of available inodes. This interacted badly with check_spool_inodes,
5951 which assumed that -1 was the "no such thing" setting. What I have done is
5952 to check that the total number of inodes is greater than zero before doing
5953 the test of how many are available.
5955 58. When a "warn" ACL statement has a log_message modifier, the message is
5956 remembered, and not repeated. This is to avoid a lot of repetition when a
5957 message has many recipients that cause the same warning to be written.
5958 However, Exim was preserving the list of already written lines for an
5959 entire SMTP session, which doesn't seem right. The memory is now reset if a
5960 new message is started.
5962 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5963 debugging output unless log_rewrite was also set.
5965 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5966 of (int)(handle) when we know that handle contains (void *)(-1).
5968 61. The Exim daemon panic-logs an error return when it closes the incoming
5969 connection. However "connection reset by peer" seems to be common, and
5970 isn't really an error worthy of noting specially, so that particular error
5973 62. When Exim is trying to find all the local interfaces, it used to panic and
5974 die if the ioctl to get the interface flags failed. However, it seems that
5975 on at least one OS (Solaris 9) it is possible to have an interface that is
5976 included in the list of interfaces, but for which you get a failure error
5977 for this call. This happens when the interface is not "plumbed" into a
5978 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5979 failure of the "get flags" call assumes that the interface is down.
5981 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5982 makes life easier for people who are doing arithmetic on fields extracted
5983 from dates, where you often get leading zeros that should not be
5984 interpreted as octal.
5986 64. Added qualify_domain to the redirect router, to override the global
5989 65. If a pathologically long header line contained very many addresses (the
5990 report of this problem mentioned 10 000) and each of them was rewritten,
5991 Exim could use up a very large amount of memory. (It kept on making new
5992 copies of the header line as it rewrote, and never released the old ones.)
5993 At the expense of a bit more processing, the header rewriting function has
5994 been changed so that it no longer eats memory in this way.
5996 66. The generation of the Received: header has been moved from the time that a
5997 message starts to be received, to the time that it finishes. The timestamp
5998 in the Received: header should now be very close to that of the <= log
5999 line. There are two side-effects of this change:
6001 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6002 logged header lines no longer include the local Received: line, because
6003 it has not yet been created. The same applies to a copy of the message
6004 that is returned to a non-SMTP sender when a message is rejected.
6006 (b) When a filter file is tested using -bf, no additional Received: header
6007 is added to the test message. After some thought, I decided that this
6010 This change does not affect the value of $received_for. It is still set
6011 after address rewriting, but before local_scan() is called.
6013 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6015 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6016 gave an unhelpful panic error message, and a defer error. I have managed to
6017 change this behaviour so that it now rejects any supplied certificate,
6018 which seems right, as the list of acceptable certificates is empty.
6020 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6021 gave an unhelpful defer error. I have not managed to make this reject any
6022 supplied certificates, but the error message it gives is "no certificate
6023 supplied", which is not helpful.
6025 70. exigrep's output now also includes lines that are not associated with any
6026 message, but which match the given pattern. Implemented by a patch from
6027 Martin Sluka, which also tidied up the Perl a bit.
6029 71. Recipient callout verification, like sender verification, was using <> in
6030 the MAIL FROM command. This isn't really the right thing, since the actual
6031 sender may affect whether the remote host accepts the recipient or not. I
6032 have changed it to use the actual sender in the callout; this means that
6033 the cache record is now keyed on a recipient/sender pair, not just the
6034 recipient address. There doesn't seem to be a real danger of callout loops,
6035 since a callout by the remote host to check the sender would use <>.
6036 [SEE ABOVE: changed after hitting problems.]
6038 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6039 temporary errors. However, in the case of such a code being given after
6040 the end of a data transmission (i.e. after ".") Exim was failing to write
6041 a retry record for the message. (Yes, there was some broken host that was
6042 actually sending 8xx at this point.)
6044 73. An unknown lookup type in a host list could cause Exim to panic-die when
6045 the list was checked. (An example that provoked this was putting <; in the
6046 middle of a list instead of at the start.) If this happened during a DATA
6047 ACL check, a -D file could be left lying around. This kind of configuration
6048 error no longer causes Exim to die; instead it causes a defer error. The
6049 incident is still logged to the main and panic logs.
6051 74. Buglet left over from Exim 3 conversion. The message "too many messages
6052 in one connection" was written to the rejectlog but not the mainlog, except
6053 when address rewriting (yes!) was being logged.
6055 75. Added write_rejectlog option.
6057 76. When a system filter was run not as root (that is, when system_filter_user
6058 was set), the values of the $n variables were not being returned to the
6059 main process; thus, they were not subsequently available in the $sn
6062 77. Added +return_path_on_delivery log selector.
6064 78. A connection timeout was being treated differently from recipients deferred
6065 when testing hosts_max_try with a message that was older than the host's
6066 retry timeout. (The host should not be counted, thus allowing all hosts to
6067 be tried at least once before bouncing.) This may have been the cause of an
6068 occasionally reported bug whereby a message would remain on the queue
6069 longer than the retry timeout, but would be bounced if a delivery was
6070 forced. I say "may" because I never totally pinned down the problem;
6071 setting up timeout/retry tests is difficult. See also the next item.
6073 79. The ultimate address timeout was not being applied to errors that involved
6074 a combination of host plus message (for example, a timeout on a MAIL
6075 command). When an address resolved to a number of possible hosts, and they
6076 were not all tried for each delivery (e.g. because of hosts_max_try), a
6077 message could remain on the queue longer than the retry timeout.
6079 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6082 81. Fixed an obscure SMTP outgoing bug which required at least the following
6083 conditions: (a) there was another message waiting for the same server;
6084 (b) the server returned 5xx to all RCPT commands in the first message so
6085 that the message was not completed; (c) the server dropped the connection
6086 or gave a negative response to the RSET that Exim sends to abort the
6087 transaction. The observed case was a dropped connection after DATA that had
6088 been sent in pipelining mode. That is, the server had advertised PIPELINING
6089 but was not implementing it correctly. The effect of the bug was incorrect
6090 behaviour, such as trying another host, and this could lead to a crash.
6096 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6097 and daemon.c were passed as pointers to ints; they should have been
6098 pointers to socklen_t variables (which are typically unsigned ints).
6100 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6103 3. Fixed a really odd bug that affected only the testing scheme; patching a
6104 certain fixed string in the binary changed the value of another string that
6105 happened to be identical to the end of the original first string.
6107 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6108 name", it returns that address as the IP address. On some operating
6109 systems (e.g. Solaris), it also passes back the IP address string as the
6110 "host name". However, on others (e.g. Linux), it passes back an empty
6111 string. Exim wasn't checking for this, and was changing the host name to an
6112 empty string, assuming it had been canonicalized.
6114 5. Although rare, it is permitted to have more than one PTR record for a given
6115 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6116 all the names associated with an address, because they do in Solaris.
6117 However, it seems that they do not in Linux for data that comes from the
6118 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6119 I found this out when I moved to a new Linux workstation and tried to run
6120 the Exim test suite.
6122 To get round this problem I have changed the code so that it now does its
6123 own call to the DNS to look up PTR records when searching for a host name.
6124 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6125 addresses that are only in /etc/hosts are still found.
6127 This behaviour is, however, controlled by an option called host_lookup_
6128 order, which defaults to "bydns:byaddr". If people want to use the other
6129 order, or indeed, just use one or the other means of lookup, they can
6130 specify it in this variable.
6132 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6133 some operating systems, this comes back from gethostbyaddr() as an empty
6134 string, and this is what Exim used to test for. However, it seems that in
6135 other systems, "." is yielded. Exim now tests for this case too.
6137 7. The values of check_spool_space and check_log_space are now held internally
6138 as a number of kilobytes instead of an absolute number of bytes. If a
6139 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6140 kilobyte. This means that much larger values can be stored.
6142 8. Exim monitor: an attempt to get the action menu when not actually pointing
6143 at a message produces an empty menu entitled "No message selected". This
6144 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6145 no entries in it ("Shell widget menu has zero width and/or height"). So I
6146 have added a single, blank menu entry in this case.
6148 9. Added ${quote_local_part.
6150 10. MIME decoding is now applied to the contents of Subject: header lines when
6153 11. Now that a reference to $sender_host_address automatically causes a reverse
6154 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6155 host lookup before query-style lookups in lists that might use this
6156 variable. This has therefore been abolished, and the "net-" prefix is no
6157 longer necessary for query-style lookups.
6159 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6160 have been a typo for LFLAGS, so it has been changed.
6162 13. The install script calls Exim with "-C /dev/null" in order to find the
6163 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6164 to be output. However, since Exim outputs its version number before the
6165 error, it didn't break the script. It just looked ugly. I fixed this by
6166 always allowing "-C /dev/null" if the caller is root.
6168 14. Ignore overlarge ACL variable number when reading spool file - insurance
6169 against a later release with more variables having written the file.
6171 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6172 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6173 this, as well as the form without the "IPv6" on the front.
6175 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6176 OS/Makefile-Darwin file.
6178 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6179 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6181 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6182 variable. However, some OS use ints for inodes. I've added cast to long int
6183 to get rid of the compiler warning.
6185 19. I had forgotten to lock out "/../" in configuration file names when
6186 ALT_CONFIG_PREFIX was set.
6188 20. Routers used for verification do not need to specify transports. However,
6189 if such a router generated a host list, and callout was configured, Exim
6190 crashed, because it could not find a port number from the (non-existent)
6191 transport. It now assumes port 25 in this circumstance.
6193 21. Added the -t option to exigrep.
6195 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6196 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6197 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6198 them is set, LOOKUP_LSEARCH is forced.
6200 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6202 24. Added sender and host information to the "rejected by local_scan()" log
6203 line; previously there was no indication of these.
6205 25. Added .include_if_exists.
6207 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6208 Linux. It turns out that not all file systems support this. Apparently some
6209 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6210 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6211 not supported on the file descriptor, is now ignored when Exim is trying to
6212 sync a directory. This applies only to Linux.
6214 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6216 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6217 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6220 29. Added support for maildirsize files from supplied patch (modified a bit).
6222 30. The use of :fail: followed by an empty string could lead Exim to respond to
6223 sender verification failures with (e.g.):
6225 550 Verification failed for <xxx>
6226 550 Sender verify failed
6228 where the first response line was missing the '-' that indicates it is not
6229 the final line of the response.
6231 31. The loop for finding the name of the user that called Exim had a hardwired
6232 limit of 10; it now uses the value of finduser_retries, which is used for
6233 all other user lookups.
6235 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6238 33. Exim was neglecting to zero errno before one call of strtol() when
6239 expanding a string and expecting an integer value. On some systems this
6240 resulted in spurious "integer overflow" errors. Also, it was casting the
6241 result into an int without checking.
6243 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6244 did not work. The code looks as if it has *never* worked, though it appears
6245 to have been documented since at least release 1.62. I have made it work.
6247 35. The "timeout_DNS" error in retry rules, also documented since at least
6248 1.62, also never worked. As it isn't clear exactly what this means, and
6249 clearly it isn't a major issue, I have abolished the feature by treating it
6250 as "timeout", and writing a warning to the main and panic logs.
6252 36. The display of retry rules for -brt wasn't always showing the error code
6255 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6256 timeout_connect_A, timeout_connect_MX.
6258 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6259 to the empty sender.
6261 39. The daemon was not analysing the content of -oX till after it had closed
6262 stderr and disconnected from the controlling terminal. This meant that any
6263 syntax errors were only noted on the panic log, and the return code from
6264 the command was 0. By re-arranging the code a little, I've made the
6265 decoding happen first, so such errors now appear on stderr, and the return
6266 code is 1. However, the actual setting up of the sockets still happens in
6267 the disconnected process, so errors there are still only recorded on the
6270 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6271 connections (as happens on some IP stacks) was logged at start up time as
6272 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6273 it from "IPv6 and IPv4", which means that two separate sockets are being
6276 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6277 says whether AF_INET or AF_INET6 was passed as an argument.
6279 42. Exiwhat output was messed up when time zones were included in log
6282 43. Exiwhat now gives more information about the daemon's listening ports,
6283 and whether -tls-on-connect was used.
6285 44. The "port" option of the smtp transport is now expanded.
6287 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6288 silently ignored. Now an error message is written to the main and panic
6291 46. There's a new ACL modifier called "logwrite" which writes to a log file
6292 as soon as it is encountered.
6294 47. Added $local_user_uid and $local_user_gid at routing time.
6296 48. Exim crashed when trying to verify a sender address that was being
6299 49. Exim was recognizing only a space character after ".include". It now also
6300 recognizes a tab character.
6302 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6303 extracting the relevant information from the specification. The man page no
6304 longer contains scrambled data for the -d option, and I've added a section
6305 at the front about calling Exim under different names.
6307 51. Added "extra_headers" argument to the "mail" command in filter files.
6309 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6312 53. Installed eximstats 1.29.
6314 54. Added transport_filter_timeout as a generic transport option.
6316 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6317 Cc: header lines. This was required by RFC 822, but it not required by RFC
6320 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6321 incoming messages that did not have them. Now it does so only if the
6322 message originates locally, that is, if there is no associated remote host
6323 address. When Resent- header lines are present, this applies to the Resent-
6324 lines rather than the non-Resent- lines.
6326 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6327 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6329 58. Messages for configuration errors now include the name of the main
6330 configuration file - useful now that there may be more than one file in a
6331 list (.included file names were always shown).
6333 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6334 for those rare installations that do not start the daemon as root or run it
6335 setuid root. I've cut out the call to initgroups() if the daemon is not
6338 60. The Exim user and group can now be bound into the binary as text strings
6339 that are looked up at the start of Exim's processing.
6341 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6343 62. Added $mailstore_basename variable.
6345 63. Installed patch to sieve.c from Michael Haardt.
6347 64. When Exim failed to open the panic log after failing to open the main log,
6348 the original message it was trying to log was written to stderr and debug
6349 output, but if they were not available (the usual case in production), it
6350 was lost. Now it is written to syslog before the two lines that record the
6351 failures to open the logs.
6353 65. Users' Exim filters run in subprocesses under the user's uid. It is
6354 possible for a "deliver" command or an alias in a "personal" command to
6355 provoke an address rewrite. If logging of address rewriting is configured,
6356 this fails because the process is not running as root or exim. There may be
6357 a better way of dealing with this, but for the moment (because 4.30 needs
6358 to be released), I have disabled address rewrite logging when running a
6359 filter in a non-root, non-exim process.
6365 1. The buildconfig auxiliary program wasn't quoting the value set for
6366 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6367 not defined. This bug was masked in 4.22 by the effect that was fixed in
6370 2. Some messages that were rejected after a message id was allocated were
6371 shown as "incomplete" by exigrep. It no longer does this for messages that
6372 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6374 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6375 have allow_domain_literals set, the ID did not get logged in the <= line.
6376 Domain literals are now always recognized in Message-ID: header lines.
6378 4. The first argument for a ${extract expansion item is the key name or field
6379 number. Leading and trailing spaces in this item were not being ignored,
6380 causing some misleading effects.
6382 5. When deliver_drop_privilege was set, single queue runner processes started
6383 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6384 same command in the process it spins off) were not dropping privilege.
6386 6. When the daemon running as "exim" started a queue runner, it always
6387 re-executed Exim in the spun-off process. This is a waste of effort when
6388 deliver_drop_privilege is set. The new process now just calls the
6389 queue-runner function directly.
6395 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6398 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6399 ignored. Though the use of -oP was forcing the writing of a pid file, it
6400 was always written to the default place.
6402 3. If the message "no IP address found for host xxxx" is generated during
6403 incoming verification, it is now followed by identification of the incoming
6404 connection (so you can more easily find what provoked it).
6406 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6408 5. Added some features to "harden" Exim a bit more against certain attacks:
6410 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6411 be put in Local/Makefile. This is like the never_users runtime option,
6412 but it cannot be overridden. The default setting is "root".
6414 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6415 prefix string with which any file named in a -C command line option
6418 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6419 is retained for -C and -D only if the caller of Exim is root. Without
6420 it, the exim user may also use -C and -D and retain privilege.
6422 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6423 command line option is disabled.
6425 6. Macro names set by the -D option must start with an upper case letter, just
6426 like macro names defined in the configuration file.
6428 7. Added "dereference=" facility to LDAP.
6430 8. Two instances of the typo "uknown" in the source files are fixed.
6432 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6433 the Configure-Makefile script screwed up while processing it.
6435 10. Incorporated PCRE 4.4.
6437 11. The SMTP synchronization check was not operating right at the start of an
6438 SMTP session. For example, it could not catch a HELO sent before the client
6439 waited for the greeting. There is now a check for outstanding input at the
6440 point when the greeting is written. Because of the duplex, asynchronous
6441 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6442 way, but not yet received, when the check is performed.
6444 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6445 on TCP/IP sockets, because this apparently causes some broken clients to
6448 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6449 unchanged) from the Cygwin maintainer.
6451 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6452 was supported instead of testing for mbx. Effectively a typo.
6454 15. The spa authenticator server code was not checking that the input it
6455 received was valid base64.
6457 16. The debug output line for the "set" modifier in ACLs was not showing the
6458 name of the variable that was being set.
6460 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6462 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6463 lookup on its own. Something else was needed to trigger the lookup. For
6464 example, a match in host_lookup or the need for a host name in a host list.
6465 Now, if $sender_host_name is referenced and the host name has not yet been
6466 looked up, a lookup is performed. If the lookup fails, the variable remains
6467 empty, and $host_lookup_failed is set to "1".
6469 19. Added "eqi" as a case-independent comparison operator.
6471 20. The saslauthd authentication condition could segfault if neither service
6472 nor realm was specified.
6474 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6475 error message that was logged was misleading, and incoming SMTP
6476 connections were dropped. The message is now more accurate, and temporary
6477 errors are given to SMTP connections.
6479 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6480 (except RSET) with a 421 error, until QUIT is received. However, it was
6481 failing to send a response to QUIT.
6483 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6484 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6485 ACL is now run after the helo_try_verify_hosts code.
6487 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6488 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6489 case-independent, so other case variants are also recognized). Apparently
6490 some systems use these upper case variants.
6492 25. If more than two messages were waiting for the same host, and a transport
6493 filter was specified for the transport, Exim sent two messages over the
6494 same TCP/IP connection, and then failed with "socket operation on non-
6495 socket" when it tried to send the third.
6497 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6499 27. The extern definition of crypt16() in expand.c was not being excluded when
6500 the OS had its own crypt16() function.
6502 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6503 as a preferred synonym for return_size_limit, both as an option and as an
6506 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6508 30. Changed the default configuration ACL to relax the local part checking rule
6509 for addresses that are not in any local domains. For these addresses,
6510 slashes and pipe symbols are allowed within local parts, but the sequence
6511 /../ is explicitly forbidden.
6513 31. SPA server authentication was not clearing the challenge buffer before
6516 32. log_message in a "warn" ACL statement was writing to the reject log as
6517 well as to the main log, which contradicts the documentation and doesn't
6518 seem right (because no rejection is happening). So I have stopped it.
6520 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6521 However, I am unable to do any testing of this.
6523 34. Fixed an infelicity in the appendfile transport. When checking directories
6524 for a mailbox, to see if any needed to be created, it was accidentally
6525 using path names with one or more superfluous leading slashes; tracing
6526 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6528 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6529 contents are added to the log line that is written for every discarded
6530 recipient. (Previously a log_message setting was ignored.)
6532 36. The ${quote: operator now quotes the string if it is empty.
6534 37. The install script runs exim in order to find its version number. If for
6535 some reason other than non-existence or emptiness, which it checks, it
6536 could not run './exim', it was installing it with an empty version number,
6537 i.e. as "exim-". This error state is now caught, and the installation is
6540 38. An argument was missing from the function that creates an error message
6541 when Exim fails to connect to the socket for saslauthd authentication.
6542 This could cause Exim to crash, or give a corrupted message.
6544 39. Added isip, isip4, and isip6 to ${if conditions.
6546 40. The ACL variables $acl_xx are now saved with the message, and can be
6547 accessed later in routers, transports, and filters.
6549 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6550 strings in the file are not string-expanded.
6552 42. If a MAIL command specified a SIZE value that was too large to fit into an
6553 int variable, the check against message_size_limit failed. Such values are
6554 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6555 day this will have to be increased, but I don't think I want to be around
6556 when emails are that large.
6563 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6564 iconv() is not standard in FreeBSD.
6566 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6567 IPv6 enabled. The observed symptom was a segmentation fault on return from
6568 the function os_common_find_running_interfaces() in src/os.c.
6570 3. In the check_special_case() function in daemon.c I had used "errno" as an
6571 argument name, which causes warnings on some systems. This was basically a
6572 typo, since it was named "eno" in the comments!
6574 4. The code that waits for the clock to tick (at a resolution of some fraction
6575 of a second) so as to ensure message-id uniqueness was always waiting for
6576 at least one whole tick, when it could have waited for less. [This is
6577 almost certainly not relevant at current processor speeds, where it is
6578 unlikely to ever wait at all. But we try to future-proof.]
6580 5. The function that sleeps for a time interval that includes fractions of a
6581 second contained a race. It did not block SIGALRM between setting the
6582 timer, and suspending (a couple of lines later). If the interval was short
6583 and the sigsuspend() was delayed until after it had expired, the suspension
6584 never ended. On busy systems this could lead to processes getting stuck for
6587 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6588 process, before it forks any delivery processes. The open lookup caching
6589 mechanism meant that the open file or database connection was passed into
6590 the delivery process. The problem was that delivery processes always tidy
6591 up cached lookup data. This could cause a problem for the next delivery
6592 process started by the queue runner, because the external queue runner
6593 process does not know about the closure. So the next delivery process
6594 still has data in the lookup cache. In the case of a file lookup, there was
6595 no problem because closing a file descriptor in a subprocess doesn't affect
6596 the parent. However, if the lookup was caching a connection to a database,
6597 the connection was closed, and the second delivery process was likely to
6598 see errors such as "PGSQL: query failed: server closed the connection
6599 unexpectedly". The problem has been fixed by closing all cached lookups
6600 in a queue runner before running a delivery process.
6602 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6603 seem to have the "const" qualifier which it has on other OS. I've
6606 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6607 *of the same type* (client or server) with the same public name that an
6608 error should be diagnosed.
6610 9. When Exim looked up a host name for an IP address, but failed to find the
6611 original IP address when looking up the host name (a safety check), it
6612 output the message "<ip address> does not match any IP for NULL", which was
6613 confusing, to say the least. The bug was that the host name should have
6614 appeared instead of "NULL".
6616 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6617 user that is built into the binary, and the -C or -D options is used, root
6618 privilege is dropped before the configuration file is read. In addition,
6619 logging is switched to stderr instead of the normal log files. If the
6620 configuration then re-defines the Exim user, the unprivileged environment
6621 is probably not what is expected, so Exim logs a panic warning message (but
6624 However, if deliver_drop_privilege is set, the unprivileged state may well
6625 be exactly what is intended, so the warning has been cut out in that case,
6626 and Exim is allowed to try to write to its normal log files.
6632 1. smtp_return_error_details was not giving details for temporary sender
6633 or receiver verification errors.
6635 2. Diagnose a configuration error if two authenticators have the same public
6638 3. Exim used not to create the message log file for a message until the first
6639 delivery attempt. This could be confusing when incoming messages were held
6640 for policy or load reasons. The message log file is now created at the time
6641 the message is received, and an initial "Received" line is written to it.
6643 4. The automatically generated man page for command line options had a minor
6644 bug that caused no ill effects; however, a more serious problem was that
6645 the procedure for building the man page automatically didn't always
6646 operate. Consequently, release 4.20 contains an out-of-date version. This
6647 shouldn't happen again.
6649 5. When building Exim with embedded Perl support, the script that builds the
6650 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6651 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6653 6. The freeze_tell option was not being used for messages that were frozen on
6654 arrival, either by an ACL or by local_scan().
6656 7. Added the smtp_incomplete_transaction log selector.
6658 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6659 was accepting AUTH without a new EHLO.
6661 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6662 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6665 10. Logging of TCP/IP connections (when configured) now happens in the main
6666 daemon process instead of the child process, so that the TCP/IP connection
6667 count is more accurate (but it can never be perfect).
6669 11. The use of "drop" in a nested ACL was not being handled correctly in the
6670 outer ACL. Now, if condition failure induced by the nested "drop" causes
6671 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6672 or "require"), the connection is dropped.
6674 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6675 that yield "discard" can now be used with an "accept" or a "discard" verb,
6676 but an error is generated for any others (because I can't see a useful way
6677 to define what should happen).
6679 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6680 are now processed in the same way as lines in the Exim configuration file.
6681 In particular, continuation lines are supported.
6683 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6685 15. Added -ti meaning -t -i.
6687 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6688 domains, and warn by logging if others are found.
6690 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6691 returned from GIFCONF when Exim is trying to find the list of interfaces on
6692 a host. The code in os.c has been modified to copy each ifreq to an aligned
6693 structure in all cases.
6695 Also, in some cases, the returned ifreq's were being copied to a 'struct
6696 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6697 means the last couple of bytes of an IPv6 address could be chopped if the
6698 ifreq contained only a normal sockaddr (14 bytes storage).
6700 18. Named domain lists were not supported in the hosts_treat_as_local option.
6701 An entry such as +xxxx was not recognized, and was treated as a literal
6704 19. Ensure that header lines added by a DATA ACL are included in the reject log
6705 if the ACL subsequently rejects the message.
6707 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6708 MD5 (which is deprecated).
6710 21. When testing a filter file using -bf, Exim was writing a message when it
6711 took the sender from a "From " line in the message, but it was not doing so
6712 when it took $return_path from a Return-Path: header line. It now does.
6714 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6715 with a valid header line field name (a series of printing characters
6716 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6718 23. Changed "disc" in the source to "disk" to conform to the documentation and
6719 the book and for uniformity.
6721 24. Ignore Sendmail's -Ooption=value command line item.
6723 25. When execve() failed while trying to run a command in a pipe transport,
6724 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6725 could be confused with a return value of 69 from the command itself. This
6726 has been changed to 127, the value the shell returns if it is asked to run
6727 a non-existent command. The wording for the related log line suggests a
6728 non-existent command as the problem.
6730 26. If received_header_text expands to an empty string, do not add a Received:
6731 header line to the message. (Well, it adds a token one on the spool, but
6732 marks it "old" so that it doesn't get used or transmitted.)
6734 27. Installed eximstats 1.28 (addition of -nt option).
6736 28. There was no check for failure on the call to getsockname() in the daemon
6737 code. This can fail if there is a shortage of resources on the system, with
6738 ENOMEM, for example. A temporary error is now given on failure.
6740 29. Contrary to the C standard, it seems that in some environments, the
6741 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6742 program. Exim now does this explicitly; it affects the formatting of
6743 timestamps using strftime().
6745 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6746 complaints. I've now initialized all the variables, to avoid this.
6748 32. Header lines added by a system filter were not being "seen" during
6749 transport-time rewrites.
6751 33. The info_callback() function passed to OpenSSL is set up with type void
6752 (*)(SSL *, int, int), as described somewhere. However, when calling the
6753 function (actually a macro) that sets it up, the type void(*)() is
6754 expected. I've put in a cast to prevent warnings from picky compilers.
6756 34. If a DNS black list lookup found a CNAME record, but there were no A
6757 records associated with the domain it pointed at, Exim crashed.
6759 35. If a DNS black list lookup returned more than one A record, Exim ignored
6760 all but the first. It now scans all returned addresses if a particular IP
6761 value is being sought. In this situation, the contents of the
6762 $dnslist_value variable are a list of all the addresses, separated by a
6765 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6766 to accept a host name and all its aliases if the forward lookup for any of
6767 them yielded the IP address of the incoming connection. Now it accepts only
6768 those names whose forward lookup yields the correct IP address. Any other
6769 names are discarded. This closes a loophole whereby a rogue DNS
6770 administrator could create reverse DNS records to break through a
6771 wildcarded host restriction in an ACL.
6773 37. If a user filter or a system filter that ran in a subprocess used any of
6774 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6775 the wrong values were passed to the pipe command ($thisaddress had the
6776 value of $0, $0 had the value of $1, etc). This bug was introduced by
6777 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6779 38. Improved the line breaking for long SMTP error messages from ACLs.
6780 Previously, if there was no break point between 40 and 75 characters, Exim
6781 left the rest of the message alone. Two changes have been made: (a) I've
6782 reduced the minimum length to 35 characters; (b) if it can't find a break
6783 point between 35 and 75 characters, it looks ahead and uses the first one
6784 that it finds. This may give the occasional overlong line, but at least the
6785 remaining text gets split now.
6787 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6788 file descriptors might be low, and that setting 1000 would always raise it.
6789 It turns out that in some environments, the limit is already over 1000 and
6790 that lowering it causes trouble. So now Exim takes care not to decrease it.
6792 40. When delivering a message, the value of $return_path is set to $sender_
6793 address at the start of routing (routers may change the value). By an
6794 oversight, this default was not being set up when an address was tested by
6795 -bt or -bv, which affected the outcome if any router or filter referred to
6798 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6799 log only when "message" or "log_message" are set. However, if one of the
6800 conditions was an address verification, or a call to a nested ACL, the
6801 messages generated by the underlying test were being passed through. This
6802 no longer happens. The underlying message is available in $acl_verify_
6803 message for both "message" and "log_message" expansions, so it can be
6804 passed through if needed.
6806 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6807 new expansion $bh_ to give the encoded byte string without charset
6808 translation. Translation happens only if iconv() is available; HAVE_ICONV
6809 indicates this at build time. HEADERS_CHARSET gives the charset to
6810 translate to; headers_charset can change it in the configuration, and
6811 "headers charset" can change it in an individual filter file.
6813 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6814 that creates RFC 2047 encoded "words" labels them as that charset instead
6815 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6816 expansion operator; (ii) when Exim creates a From: line for a local
6817 message; (iii) when a header line is rewritten to include a "phrase" part.
6819 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6820 buggy, causing it to skip the first lines of messages whose message ID
6821 ended in 'D'. This would not have bitten before Exim release 4.14, because
6822 message IDs were unlikely to end in 'D' before then. The effect was to have
6823 incorrect size information for certain domains.
6825 45. #include "config.h" was missing at the start of the crypt16.c module. This
6826 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6829 46. If there was a timeout during a "random" callout check, Exim treated it as
6830 a failure of the random address, and carried on sending RSET and the real
6831 address. If the delay was just some slowness somewhere, the response to the
6832 original RCPT would be taken as a response to RSET and so on, causing
6833 mayhem of various kinds.
6835 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6836 when I implemented it. It didn't allow for the fact that some option values
6837 may legitimately be negative (e.g. size_addition), and it didn't even do
6838 the right test for positive values.
6840 48. Domain names in DNS records are case-independent. Exim always looks them up
6841 in lower case. Some resolvers return domain names in exactly the case they
6842 appear in the zone file, that is, they may contain uppercase letters. Not
6843 all resolvers do this - some return always lower case. Exim was treating a
6844 change of case by a resolver as a change of domain, similar to a widening
6845 of a domain abbreviation. This triggered its re-routing code and so it was
6846 trying to route what was effectively the same domain again. This normally
6847 caused routing to fail (because the router wouldn't handle the domain
6848 twice). Now Exim checks for this case specially, and just changes the
6849 casing of the domain that it ultimately uses when it transmits the message
6852 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6855 50. If a filter generated a file delivery with a non-absolute name (possible if
6856 no home directory exists for the router), the forbid_file option was not
6859 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6860 the existing equality matching.
6862 52. Exim was using ints instead of ino_t variables in some places where it was
6863 dealing with inode numbers.
6865 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6866 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6867 called TMPDIR, and if it finds it is different, it changes its value.
6869 54. The smtp_printf() function is now made available to local_scan() so
6870 additional output lines can be written before returning. There is also an
6871 smtp_fflush() function to enable the detection of a dropped connection.
6872 The variables smtp_input and smtp_batched_input are exported to
6875 55. Changed the default runtime configuration: the message "Unknown user"
6876 has been removed from the ACL, and instead placed on the localuser router,
6877 using the cannot_route_message feature. This means that any verification
6878 failures that generate their own messages won't get overridden. Similarly,
6879 the "Unrouteable address" message that was in the ACL for unverifiable
6880 relay addresses has also been removed.
6882 56. Added hosts_avoid_esmtp to the smtp transport.
6884 57. The exicyclog script was not checking for the esoteric option
6885 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6886 will work only if exicyclog is run under the appropriate euid.
6888 58. Following a discussion on the list, the rules by which Exim recognises line
6889 endings on incoming messages have been changed. The -dropcr and drop_cr
6890 options are now no-ops, retained only for backwards compatibility. The
6891 following line terminators are recognized: LF CRLF CR. However, special
6892 processing applies to CR:
6894 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6895 nor a local message in the state where . is a terminator.
6897 (ii) If a bare CR is encountered in a header line, an extra space is added
6898 after the line terminator so as not to end the header. The reasoning
6899 behind this is that bare CRs in header lines are most likely either
6900 to be mistakes, or people trying to play silly games.
6902 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6903 was being incorrectly given as 18 bytes larger than it should have been.
6904 This is a VOB (very old bug).
6906 60. This may never have affected anything current, but just in case it has:
6907 When the local host is found other than at the start of a list of hosts,
6908 the local host, those with the same MX, and any that follow, are discarded.
6909 When the list in question was part of a longer list of hosts, the following
6910 hosts (not currently being processed) were also being discarded. This no
6911 longer happens. I'm not sure if this situation could ever has previously
6914 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6917 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6918 header. This is something that is recommended in a new Internet Draft, and
6919 is something that is documented as being done by Sendmail. There are two
6920 possible values. For messages generated by the autoreply transport, Exim
6923 Auto-Submitted: auto-replied
6925 whereas for all other generated messages (e.g. bounces) it adds
6927 Auto-Submitted: auto-generated
6929 63. The "personal" condition in filters now includes a test for the
6930 Auto-Submitted: header. If it contains the string "auto-" the message it
6931 not considered personal.
6933 64. Added rcpt_include_affixes as a generic transport option.
6935 65. Added queue_only_override (default true).
6937 66. Added the syslog_duplication option.
6939 67. If what should have been the first header line of a message consisted of
6940 a space followed by a colon, Exim was mis-interpreting it as a header line.
6941 It isn't of course - it is syntactically invalid and should therefore be
6942 treated as the start of the message body. The misbehaviour could have
6943 caused a number of strange effects, including loss of data in subsequent
6944 header lines, and spool format errors.
6946 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6947 client host had authenticated. This control can now be exercised by an ACL
6948 for more flexibility.
6950 69. By default, callouts do not happen when testing with -bh. There is now a
6951 variant, -bhc, which does actually run the callout code, including
6952 consulting and updating the callout cache.
6954 70. Added support for saslauthd authentication, courtesy of Alexander
6957 71. If statvfs() failed on the spool or log directories while checking their
6958 size for availability, Exim confusingly gave the error "space shortage".
6959 Furthermore, in debugging mode it crashed with a floating point exception.
6960 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6961 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6962 serious problem, Exim now writes to the main and panic logs when this
6963 happens, with details of the failure. It then refuses to accept the
6964 incoming message, giving the message "spool directory problem" or "log
6965 directory problem" with a 421 code for SMTP messages.
6967 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6968 0, 1, and 2 exist, because some OS complain for execs without them (see
6969 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6970 descriptors. However, the code omitted to check that the open succeeded,
6971 causing mysterious errors if for some reason the permissions on /dev/null
6972 got screwed. Now Exim writes a message to the main and panic logs, and
6973 bombs out if it can't open /dev/null.
6975 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6976 interact so that it is all more flexible. It is supposed to remain
6977 backwards compatible. Also added extra_local_interfaces.
6979 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6980 to bomb out with an assertion failure - to the client this appears as a
6981 connection drop. This problem occurs in the part of the code that was taken
6982 from the Samba project. Fortunately, the assertion is in a very simple
6983 function, so I have fixed this by reproducing the function inline in the
6984 one place where it is called, and arranging for authentication to fail
6985 instead of killing the process with assert().
6987 75. The SPA client code was not working when the server requested OEM rather
6988 than Unicode encoding.
6990 76. Added code to make require_files with a specific uid setting more usable in
6991 the case where statting the file as root fails - usually a non-root-mounted
6992 NFS file system. When this happens and the failure is EACCES, Exim now
6993 forks a subprocess and does the per-uid checking as the relevant uid.
6995 77. Added process_log_path.
6997 78. If log_file_path was not explicitly set, a setting of check_log_space or
6998 check_log_inodes was ignored.
7000 79. If a space check for the spool or log partitions fails, the incident is now
7001 logged. Of course, in the latter case the data may get lost...
7003 80. Added the %p formatting code to string_format() so that it can be used to
7004 print addresses in debug_print(). Adjusted all the address printing in the
7005 debugging in store.c to use %p rather than %d.
7007 81. There was a concern that a line of code in smtp_in.c could overflow a
7008 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7009 initially expressed, the concern was not well-founded, because trailing
7010 spaces are removed early. However, if the trailing spaces were followed by
7011 a NULL, they did not get removed, so the overflow was possible. Two fixes
7014 (a) I re-wrote the offending code in a cleaner fashion.
7015 (b) If an incoming SMTP command contains a NULL character, it is rejected
7018 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7019 runs initgroups(), so that if the Exim user is in any additional groups,
7020 they will be used during message reception.
7026 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob