1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
78 JH/01 Rework error string handling in TLS interface so that the caller in
79 more cases is responsible for logging. This permits library-sourced
80 string to be attached to addresses during delivery, and collapses
81 pairs of long lines into single ones.
83 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
84 during configuration. Wildcards are allowed and expanded.
86 JH/02 Rework error string handling in DKIM to pass more info back to callers.
87 This permits better logging.
89 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
90 do not close it down and have the child transport start it up again on
91 the passed-on TCP connection. Instead, proxy the child (and any
92 subsequent ones) for TLS via a unix-domain socket channel. Logging is
93 affected: the continued delivery log lines do not have any DNSSEC, TLS
94 Certificate or OCSP information. TLS cipher information is still logged.
96 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
97 identical IP addresses on different listening ports. Will also affect
100 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
101 add noisy ifdef guards to special-case this sillyness.
102 Patch from Bernd Kuhls.
104 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
105 than 255 are no longer allowed.
107 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
108 Disable the session-cache too, which might reduce our load. Since we
109 currrectly use a new context for every connection, both as server and
110 client, there is no benefit for these.
111 GnuTLS appears to not support tickets server-side by default (we don't
112 call gnutls_session_ticket_enable_server()) but client side is enabled
113 by default on recent versions (3.1.3 +) unless the PFS priority string
116 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
117 <https://reproducible-builds.org/specs/source-date-epoch/>.
119 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
120 the check for any unsuccessful recipients did not notice the limit, and
121 erroneously found still-pending ones.
123 JH/08 Pipeline CHUNKING command and data together, on kernels that support
124 MSG_MORE. Only in-clear (not on TLS connections).
126 JH/09 Avoid using a temporary file during transport using dkim. Unless a
127 transport-filter is involved we can buffer the headers in memory for
128 creating the signature, and read the spool data file once for the
129 signature and again for transmission.
131 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
132 4.77 as the kernel support then wasn't solid, having issues in 64bit
133 mode. Now, it's been long enough. Add support for FreeBSD also.
135 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
136 case where the routing stage had gathered several addresses to send to
137 a host before calling the transport for the first, we previously failed
138 to close down TLS in the old transport process before passing the TCP
139 connection to the new process. The new one sent a STARTTLS command
140 which naturally failed, giving a failed delivery and bloating the retry
141 database. Investigation and fix prototype from Wolfgang Breyha.
143 JH/12 Fix check on SMTP command input synchronisation. Previously there were
144 false-negatives in the check that the sender had not preempted a response
145 or prompt from Exim (running as a server), due to that code's lack of
146 awareness of the SMTP input buffering.
148 PP/04 Add commandline_checks_require_admin option.
149 Exim drops privileges sanely, various checks such as -be aren't a
150 security problem, as long as you trust local users with access to their
151 own account. When invoked by services which pass untrusted data to
152 Exim, this might be an issue. Set this option in main configuration
153 AND make fixes to the calling application, such as using `--` to stop
156 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
157 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
158 responses to those, into a single TLS record each way (this usually means
159 a single packet). As a side issue, smtp_enforce_sync now works on TLS
162 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
163 affects you only if you're dancing at the edge of the param size limits.
164 If you are, and this message makes sense to you, then: raise the
165 configured limit or use OpenSSL 1.1. Nothing we can do for older
168 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
169 cmdline element to get the documented default one. Previously it was
172 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
173 are unsuable for non-message-specific reasons - eg. connection timeout,
174 banner-time rejection.
176 JH/16 Fix logging of delivery remote port, when specified by router, under
179 PP/06 Repair manualroute's ability to take options in any order, even if one
180 is the name of a transport.
183 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
185 JH/17 Change the list-building routines interface to use the expanding-string
186 triplet model, for better allocation and copying behaviour.
188 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
189 Previously it was constructed the first time a possibly-matching string
190 was met in the configuration file input during startup; now it is done
193 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
194 compatible one, to avoid the (poorly documented) possibility of a config
195 file in the working directory redirecting the DB files, possibly correpting
196 some existing file. CVE-2017-10140 assigned for BDB.
198 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
199 cache-hot. Previously, although the result was properly cached, the
200 initial verify call returned a defer.
202 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
203 the main verify for receipient in uncached-mode.
205 JH/22 Retire historical build files to an "unsupported" subdir. These are
206 defined as "ones for which we have no current evidence of testing".
208 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
209 if present. Previously it was ignored.
211 JH/24 Start using specified-initialisers in C structure init coding. This is
212 a C99 feature (it's 2017, so now considered safe).
214 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
215 if was a fixed-sized field and bitmask ops via macros; it is now more
218 PP/07 GitHub PR 56: Apply MariaDB build fix.
219 Patch provided by Jaroslav Škarvada.
221 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
222 during Coverity cleanups [4.87 JH/47]
223 Diagnosis and fix provided by Michael Fischer v. Mollard.
225 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
226 the right size to place the terminating semicolon on its own folded
227 line, the header hash was calculated to an incorrect value thanks to
228 the (relaxed) space the fold became.
230 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
231 and confused the parent.
233 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
234 which could crash as a result. This could lead to undeliverable messages.
236 JH/28 Logging: "next input sent too soon" now shows where input was truncated
239 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
240 matters on fast-turnover and PID-randomising systems, which were getting
241 out-of-order delivery.
243 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
244 a possibly-overlapping copy. The symptom was that "Remote host closed
245 connection in response to HELO" was logged instead of the actual 4xx
248 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
249 Previously only that bufferd was discarded, resulting in SYMTP command
252 JH/32 DKIM: when a message has multiple signatures matching an identity given
253 in dkim_verify_signers, run the dkim acl once for each. Previously only
254 one run was done. Bug 2189.
256 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
257 "panic the current process" to "deliberately defer". The panic log is
258 still written with the problem list name; the mail and reject logs now
259 get a temp-reject line for the message that was being handled, saying
260 something like "domains check lookup or other defer". The SMTP 451
261 message is still "Temporary local problem".
263 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
264 A crafted sequence of BDAT commands could result in in-use memory beeing
265 freed. CVE-2017-16943.
267 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
268 from SMTP input. Previously it was always done; now only done for DATA
269 and not BDAT commands. CVE-2017-16944.
271 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
272 to the message (such as an overlong header line). Previously this was
273 not done and we did not exit BDAT mode. Followon from the previous item
274 though a different problem.
280 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
281 than -2003 did; needs libidn2 in addition to libidn.
283 JH/02 The path option on a pipe transport is now expanded before use.
285 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
286 Patch provided by "Björn", documentation fix added too.
288 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
289 missing a wire-to-host endian conversion.
291 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
292 close after a BDAT command line could be taken as a following command,
293 giving a synch failure. Fix by only checking for synch immediately
294 before acknowledging the chunk.
296 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
297 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
298 macro. Patches provided by Josh Soref.
300 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
301 Previously we did not; the RFC seems ambiguous and VRFY is not listed
302 by IANA as a service extension. However, John Klensin suggests that we
305 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
306 the dkim code may be unix-mode line endings rather than smtp wire-format
307 CRLF, so prepend a CR to any bare LF.
309 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
310 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
312 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
313 the first were themselves being wrongly included in the feed into dkim
314 processing; with most chunk sizes in use this resulted in an incorrect
315 body hash calculated value.
317 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
318 DKIM signature block, for verification. Although advised against by
319 standards it is specifically not ruled illegal.
321 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
323 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
324 missing a body hash (the bh= tag).
326 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
327 It seems that HAProxy sends the Proxy Protocol information in clear and
328 only then does a TLS startup, so do the same.
330 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
331 TCP connections (such as for Spamd) unless the daemon successfully set
332 Fast Open mode on its listening sockets. This fixes breakage seen on
333 too-old kernels or those not configured for Fast Open, at the cost of
334 requiring both directions being enabled for TFO, and TFO never being used
335 by non-daemon-related Exim processes.
337 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
338 endings, at least on the first header line. Try to canonify any that get
339 past that check, despite the cost.
341 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
342 now limited to an arbitrary five deep, while parsing addresses with the
343 strip_excess_angle_brackets option enabled.
345 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
346 instead leave the unprompted TLS handshake in socket buffer for the
347 TLS library to consume.
349 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
351 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
353 JH/16 Drop variables when they go out of scope. Memory management drops a whole
354 region in one operation, for speed, and this leaves assigned pointers
355 dangling. Add checks run only under the testsuite which checks all
356 variables at a store-reset and panics on a dangling pointer; add code
357 explicitly nulling out all the variables discovered. Fixes one known
358 bug: a transport crash, where a dangling pointer for $sending_ip_address
359 originally assigned in a verify callout, is re-used.
361 PP/06 Drop '.' from @INC in various Perl scripts.
363 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
365 PP/08 Reduce a number of compilation warnings under clang; building with
366 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
367 should be warning-free.
369 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
371 HS/01 Fix portability problems introduced by PP/08 for platforms where
372 realloc(NULL) is not equivalent to malloc() [SunOS et al].
374 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
375 chunk. This allows us to accept broken chunked messages. We need a more
376 general solution here.
378 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
379 already-broken messages in the queue.
381 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
383 JH/19 Fix reference counting bug in routing-generated-address tracking.
389 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
390 supports it and a size is available (ie. the sending peer gave us one).
392 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
393 years of being deprecated). The replacements are the ACLs
394 acl_smtp_mime and acl_not_smtp_mime.
396 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
397 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
398 or even an in-clear connection were permitted. Now, if the host lookup
399 was dnssec and dane was requested then the host is only used if the
400 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
401 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
402 if one fails this test.
403 This means that a poorly-configured remote DNS will make it incommunicado;
404 but it protects against a DNS-interception attack on it.
406 JH/04 Bug 1810: make continued-use of an open smtp transport connection
407 non-noisy when a race steals the message being considered.
409 JH/05 If main configuration option tls_certificate is unset, generate a
410 self-signed certificate for inbound TLS connections.
412 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
413 in rewrites and routers.
415 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
416 and logged a warning sing 4.83; now they are a configuration file error.
418 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
419 (lacking @domain). Apply the same qualification processing as RCPT.
421 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
423 JH/10 Support ${sha256:} applied to a string (as well as the previous
426 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
427 a cutthrough deliver is pending, as we always want to make a connection.
428 This also avoids re-routing the message when later placing the cutthrough
429 connection after a verify cache hit.
430 Do not update it with the verify result either.
432 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
433 when routing results in more than one destination address.
435 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
436 signing (which inhibits the cutthrough capability). Previously only
437 the presence of an option was tested; now an expansion evaluating as
438 empty is permissible (obviously it should depend only on data available
439 when the cutthrough connection is made).
441 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
442 the relevant preceding SMTP command did not note the pipelining mode.
444 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
445 Previously they were not counted.
447 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
448 as one having no matching records. Previously we deferred the message
449 that needed the lookup.
451 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
452 distinguished as "(=".
454 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
455 for missing MX records. Previously it only worked for missing A records.
457 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
459 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
460 after the data-go-ahead and data-ack. Patch from Jason Betts.
462 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
463 even for a "none" policy. Patch from Tony Meyer.
465 JH/22 Fix continued use of a connection for further deliveries. If a port was
466 specified by a router, it must also match for the delivery to be
469 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
470 When one of the recipients of a message was unsuitable for the connection
471 (has no matching addresses), we lost track of needing to mark it
472 deferred. As a result mail would be lost.
474 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
476 JH/25 Decoding ACL controls is now done using a binary search; the source code
477 takes up less space and should be simpler to maintain. Merge the ACL
478 condition decode tables also, with similar effect.
480 JH/26 Fix problem with one_time used on a redirect router which returned the
481 parent address unchanged. A retry would see the parent address marked as
482 delivered, so not attempt the (identical) child. As a result mail would
485 JH/27 Fix a possible security hole, wherein a process operating with the Exim
486 UID can gain a root shell. Credit to http://www.halfdog.net/ for
487 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
490 JH/28 Enable {spool,log} filesystem space and inode checks as default.
491 Main config options check_{log,spool}_{inodes,space} are now
492 100 inodes, 10MB unless set otherwise in the configuration.
494 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
495 Previously it only applied to the main-section connection policy
498 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
500 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
501 by me. Added RFC7919 DH primes as an alternative.
503 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
504 are not in the system include path.
506 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
507 GnuTLS, when a session startup failed (eg because the client disconnected)
508 Exim did stdio operations after fclose. This was exposed by a recent
509 change which nulled out the file handle after the fclose.
511 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
512 signed directly by the cert-signing cert, rather than an intermediate
513 OCSP-signing cert. This is the model used by LetsEncrypt.
515 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
517 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
518 an incoming connection.
520 HS/02 Bug 1802: Do not half-close the connection after sending a request
523 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
524 fallback to "prime256v1".
526 JH/34 SECURITY: Use proper copy of DATA command in error message.
527 Could leak key material. Remotely exploitable. CVE-2016-9963.
533 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
534 and 3.4.4 - once the server is enabled to respond to an OCSP request
535 it does even when not requested, resulting in a stapling non-aware
536 client dropping the TLS connection.
538 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
539 support variable-length bit vectors. No functional change.
541 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
542 The I= interface field on outgoing lines is now after the H= remote
543 host field, same as incoming lines. There is a separate
544 outgoing_interface log selector which allows you to disable the
547 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
548 If not running log_selector +smtp_connection the mainlog would be held
549 open indefinitely after a "too many connections" event, including to a
550 deleted file after a log rotate. Leave the per net connection logging
551 leaving it open for efficiency as that will be quickly detected by the
552 check on the next write.
554 HS/01 Bug 1671: Fix post transport crash.
555 Processing the wait-<transport> messages could crash the delivery
556 process if the message IDs didn't exist for some reason. When
557 using 'split_spool_directory=yes' the construction of the spool
558 file name failed already, exposing the same netto behaviour.
560 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
561 mime_regex ACL conditions.
563 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
564 to DSN fail messages (bounces): remote IP, remote greeting, remote response
565 to HELO, local diagnostic string.
567 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
568 log line to debug. Even when configured with a tls authenticator many
569 client connections are expected to not authenticate in this way, so
570 an authenticate fail is not an error.
572 HS/02 Add the Exim version string to the process info. This way exiwhat
573 gives some more detail about the running daemon.
575 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
576 matter for fast-change records such as DNSBLs.
578 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
579 retry record, even if constant. There may be multiple transports with
580 different interface settings and the retry behaviour needs to be kept
583 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
585 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
587 JH/10 Bug 840: fix log_defer_output option of pipe transport
589 JH/11 Bug 830: use same host for all RCPTS of a message, even under
590 hosts_randomize. This matters a lot when combined with mua_wrapper.
592 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
593 ${quote_pgsql:<string>} operator.
595 JH/13 Bug 1708: avoid misaligned access in cached lookup.
597 JH/14 Change header file name for freeradius-client. Relevant if compiling
598 with Radius support; from the Gentoo tree and checked under Fedora.
600 JH/15 Bug 1712: Introduce $prdr_requested flag variable
602 JH/16 Bug 1714: Permit an empty string as expansion result for transport
603 option transport_filter, meaning no filtering.
605 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
607 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
608 defaults to "*" (all hosts). The variable is now available when not built
609 with TLS, default unset, mainly to enable keeping the testsuite sane.
610 If a server certificate is not supplied (via tls_certificate) an error is
611 logged, and clients will find TLS connections fail on startup. Presumably
612 they will retry in-clear.
613 Packagers of Exim are strongly encouraged to create a server certificate
614 at installation time.
616 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
617 with the $config_file variable.
619 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
620 in transport context, after the attempt, and per-recipient. The latter type
621 is per host attempted. The event data is the error message, and the errno
622 information encodes the lookup type (A vs. MX) used for the (first) host,
623 and the trailing two digits of the smtp 4xx response.
625 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
626 to write to mainlog (or rejectlog, paniclog) in the window between file
627 creation and permissions/ownership being changed. Particularly affects
628 installations where exicyclog is run as root, rather than exim user;
629 result is that the running daemon panics and dies.
631 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
633 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
634 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
635 "pri" and "weight". Note that the previous implicit priority given by the
636 list order is no longer honoured.
638 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
641 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
642 by defining SUPPORT_SOCKS.
644 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
645 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
646 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
647 variables are renamed to proxy_{local,external}_{address,port}.
649 JH/27 Move Internationalisation support from Experimental to mainline, enabled
650 for a build by defining SUPPORT_I18N
652 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
653 of the query string, and make ${quote_redis:} do that quoting.
655 JH/29 Move Events support from Experimental to mainline, enabled by default
656 and removable for a build by defining DISABLE_EVENT.
658 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
660 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
661 cached by the daemon.
663 JH/32 Move Redis support from Experimental to mainline, enabled for a build
664 by defining LOOKUP_REDIS. The libhiredis library is required.
666 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
667 keys are given for lookup.
669 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
670 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
671 only supported when built with TLS support. The PolarSSL SHA routines
672 are still used when the TLS library is too old for convenient support.
674 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
675 openssl_options), for security. OpenSSL forces this from version 1.1.0
676 server-side so match that on older versions.
678 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
679 allocation for $value could be released as the expansion processing
680 concluded, but leaving the global pointer active for it.
682 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
683 and to use the domains and local_parts ACL conditions.
685 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
686 incorrectly not doubled on cutthrough transmission, hence seen as a
687 body-termination at the receiving system - resulting in truncated mails.
688 Commonly the sender saw a TCP-level error, and retransmitted the message
689 via the normal store-and-forward channel. This could result in duplicates
690 received - but deduplicating mailstores were liable to retain only the
691 initial truncated version.
693 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
695 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
697 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
698 we're in there, support oversigning also; bug 1309.
700 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
702 HS/04 Add support for keep_environment and add_environment options.
704 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
705 either intentional arithmetic overflow during PRNG, or testing config-
708 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
709 delivery resulted in actual delivery. Cancel cutthrough before DATA
712 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
713 rejects a recipient: pass the reject to the originator.
715 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
716 Many were false-positives and ignorable, but it's worth fixing the
719 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
720 for the new environment-manipulation done at startup. Move the routines
721 from being local to tls.c to being global via the os.c file.
723 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
724 an extract embedded as result-arg for a map, the first arg for extract
725 is unavailable so we cannot tell if this is a numbered or keyed
726 extraction. Accept either.
732 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
735 JH/02 The smtp transport option "multi_domain" is now expanded.
737 JH/03 The smtp transport now requests PRDR by default, if the server offers
740 JH/04 Certificate name checking on server certificates, when exim is a client,
741 is now done by default. The transport option tls_verify_cert_hostnames
742 can be used to disable this per-host. The build option
743 EXPERIMENTAL_CERTNAMES is withdrawn.
745 JH/05 The value of the tls_verify_certificates smtp transport and main options
746 default to the word "system" to access the system default CA bundle.
747 For GnuTLS, only version 3.0.20 or later.
749 JH/06 Verification of the server certificate for a TLS connection is now tried
750 (but not required) by default. The verification status is now logged by
751 default, for both outbound TLS and client-certificate supplying inbound
754 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
757 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
758 Status Notification (bounce) messages are now MIME format per RFC 3464.
759 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
760 under the control of the dsn_advertise_hosts option, and routers may
761 have a dsn_lasthop option.
763 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
764 default, modifiable by a malware= option. The list separator for
765 the options can now be changed in the usual way. Bug 68.
767 JH/10 The smtp_receive_timeout main option is now expanded before use.
769 JH/11 The incoming_interface log option now also enables logging of the
770 local interface on delivery outgoing connections.
772 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
773 if the interface and destination host and port all match.
775 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
778 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
779 Patch from Andrew Lewis.
781 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
782 now supports optional time-restrictions, weighting, and priority
783 modifiers per server. Patch originally by <rommer@active.by>.
785 JH/16 The spamd_address main option now supports a mixed list of local
786 and remote servers. Remote servers can be IPv6 addresses, and
787 specify a port-range.
789 JH/17 Bug 68: The spamd_address main option now supports an optional
790 timeout value per server.
792 JH/18 Bug 1581: Router and transport options headers_add/remove can
793 now have the list separator specified.
795 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
798 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
801 JH/21 Support for the A6 type of dns record is withdrawn.
803 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
804 rather than the verbs used.
806 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
807 from 255 to 1024 chars.
809 JH/24 Verification callouts now attempt to use TLS by default.
811 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
812 are generic router options now. The defaults didn't change.
814 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
815 Original patch from Alexander Shikoff, worked over by JH.
817 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
818 files if ZCAT_COMMAND is not executable.
820 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
822 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
824 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
825 Normally benign, it bites when the pair was led to by a CNAME;
826 modern usage is to not canonicalize the domain to a CNAME target
827 (and we were inconsistent anyway for A-only vs AAAA+A).
829 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
831 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
832 when evaluating $sender_host_dnssec.
834 JH/31 Check the HELO verification lookup for DNSSEC, adding new
835 $sender_helo_dnssec variable.
837 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
839 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
841 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
843 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
844 documented as working, but never had. Support all but $spam_report.
846 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
847 added for tls authenticator.
849 HS/03 Add perl_taintmode main config option
855 TL/01 When running the test suite, the README says that variables such as
856 no_msglog_check are global and can be placed anywhere in a specific
857 test's script, however it was observed that placement needed to be near
858 the beginning for it to behave that way. Changed the runtest perl
859 script to read through the entire script once to detect and set these
860 variables, reset to the beginning of the script, and then run through
861 the script parsing/test process like normal.
863 TL/02 The BSD's have an arc4random API. One of the functions to induce
864 adding randomness was arc4random_stir(), but it has been removed in
865 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
866 function when detected.
868 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
869 cause callback expansion.
871 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
872 syntax errors in an expansion can be treated as a string instead of
873 logging or causing an error, due to the internal use of bool_lax
874 instead of bool when processing it.
876 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
877 server certificates when making smtp deliveries.
879 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
881 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
883 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
885 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
886 Merged patch from Sebastian Wiedenroth.
888 JH/05 Fix results-pipe from transport process. Several recipients, combined
889 with certificate use, exposed issues where response data items split
890 over buffer boundaries were not parsed properly. This eventually
891 resulted in duplicates being sent. This issue only became common enough
892 to notice due to the introduction of connection certificate information,
893 the item size being so much larger. Found and fixed by Wolfgang Breyha.
895 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
896 size buffer was used, resulting in syntax errors when an expansion
899 JH/07 Add support for directories of certificates when compiled with a GnuTLS
900 version 3.3.6 or later.
902 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
903 is EXPERIMENTAL_EVENT, the main-configuration and transport options
904 both become "event_action", the variables become $event_name, $event_data
905 and $event_defer_errno. There is a new variable $verify_mode, usable in
906 routers, transports and related events. The tls:cert event is now also
907 raised for inbound connections, if the main configuration event_action
910 TL/06 In test suite, disable OCSP for old versions of openssl which contained
911 early OCSP support, but no stapling (appears to be less than 1.0.0).
913 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
914 server certificate names available under the smtp transport option
915 "tls_verify_cert_hostname" now do not permit multi-component wildcard
918 JH/10 Time-related extraction expansions from certificates now use the main
919 option "timezone" setting for output formatting, and are consistent
920 between OpenSSL and GnuTLS compilations. Bug 1541.
922 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
923 encoded parameter in the incoming message. Bug 1558.
925 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
926 include certificate info, eximon was claiming there were spoolfile
929 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
931 JH/14 Log delivery-related information more consistently, using the sequence
932 "H=<name> [<ip>]" wherever possible.
934 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
935 are problematic for Debian distribution, omit them from the release
938 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
940 JH/16 Fix string representation of time values on 64bit time_t architectures.
943 JH/17 Fix a null-indirection in certextract expansions when a nondefault
944 output list separator was used.
949 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
950 checkers that were complaining about end of non-void function with no
953 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
954 This was a regression introduced in 4.83 by another bugfix.
956 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
958 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
959 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
965 TF/01 Correctly close the server side of TLS when forking for delivery.
967 When a message was received over SMTP with TLS, Exim failed to clear up
968 the incoming connection properly after forking off the child process to
969 deliver the message. In some situations the subsequent outgoing
970 delivery connection happened to have the same fd number as the incoming
971 connection previously had. Exim would try to use TLS and fail, logging
972 a "Bad file descriptor" error.
974 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
975 utilities have not been installed.
977 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
978 temporary space as the ACL may create new global variables.
980 TL/01 LDAP support uses per connection or global context settings, depending
981 upon the detected version of the libraries at build time.
983 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
984 to extract and use the src ip:port in logging and expansions as if it
985 were a direct connection from the outside internet. PPv2 support was
986 updated based on HAProxy spec change in May 2014.
988 JH/02 Add ${listextract {number}{list}{success}{fail}}.
990 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
991 Properly escape header and check for NULL return.
993 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
996 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
998 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
999 characters in header names, implemented as a verify condition.
1000 Contributed by Michael Fischer v. Mollard.
1002 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1003 results permerror and temperror. Previous values are deprecated but
1004 still accepted. In a future release, err_perm and err_temp will be
1005 completely removed, which will be a backward incompatibility if the
1006 ACL tests for either of these two old results. Patch contributed by
1007 user bes-internal on the mailing list.
1009 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1011 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1012 selectors, in both main and reject logs.
1014 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1017 JH/07 Add malware type "sock" for talking to simple daemon.
1019 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1021 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1022 routers/transports under cutthrough routing.
1024 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1025 numbers. Touch up "bool" conditional to keep the same definition.
1027 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1029 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1031 JH/12 Expand items in router/transport headers_add or headers_remove lists
1032 individually rather than the list as a whole. Bug 1452.
1034 Required for reasonable handling of multiple headers_ options when
1035 they may be empty; requires that headers_remove items with embedded
1036 colons must have them doubled (or the list-separator changed).
1038 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1039 view the policy declared in the DMARC record. Currently, $dmarc_status
1040 is a combined value of both the record presence and the result of the
1043 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1045 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1046 dnslookup router and the smtp transport (applying to the forward
1049 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1050 of ldap servers used for a specific lookup. Patch provided by Heiko
1053 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1054 New variable $lookup_dnssec_authenticated for observability.
1056 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1057 Patch submitted by Lars Timman.
1059 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1061 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1062 Requires trusted mode and valid format message id, aborts otherwise.
1063 Patch contributed by Heiko Schlichting.
1065 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1066 certextract with support for various fields. Bug 1358.
1068 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1069 is requested by default, modifiable by smtp transport option
1072 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1073 operate on certificate variables to give certificate fingerprints
1074 Also new ${sha256:cert_variable}.
1076 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1078 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1079 Christian Aistleitner.
1081 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1083 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1084 file. Patch from Wolfgang Breyha.
1086 JH/25 Expand the coverage of the delivery $host and $host_address to
1087 client authenticators run in verify callout. Bug 1476.
1089 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1090 align with daemon_smtp_ports. Bug 72.
1092 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1093 support and error reporting did not work properly.
1095 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1096 and is readable. Patch from Andrew Colin Kissa.
1098 TL/14 Enhance documentation of ${run expansion and how it parses the
1099 commandline after expansion, particularly in the case when an
1100 unquoted variable expansion results in an empty value.
1102 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1104 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1105 by renaming away from it. Addresses GH issue 15, reported by
1108 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1110 TL/15 SECURITY: prevent double expansion in math comparison functions
1111 (can expand unsanitized data). Not remotely exploitable.
1118 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1120 PP/02 Make -n do something, by making it not do something.
1121 When combined with -bP, the name of an option is not output.
1123 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1126 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1127 $sender_host_name and config options to manage this, and basic check
1130 PP/05 DSCP support for outbound connections and control modifier for inbound.
1132 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1133 (Only plugin which currently uses this is kerberos4, which nobody should
1134 be using, but we should make it available and other future plugins might
1135 conceivably use it, even though it would break NAT; stuff *should* be
1136 using channel bindings instead).
1138 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1139 name; added for Sendmail compatibility; requires admin caller.
1140 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1141 just ignore it); requires trusted caller.
1142 Also parse but ignore: -Ac -Am -X<logfile>
1145 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1147 TL/02 Add +smtp_confirmation as a default logging option.
1149 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1150 Patch by Magnus Holmgren from 2007-02-20.
1152 TL/04 Bugzilla 1281 - Spec typo.
1153 Bugzilla 1283 - Spec typo.
1154 Bugzilla 1290 - Spec grammar fixes.
1156 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1158 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1160 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1161 mailing list by by Dmitry Isaikin.
1163 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1165 JH/02 Support "G" suffix to numbers in ${if comparisons.
1167 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1169 NM/01 Bugzilla 1197 - Spec typo
1170 Bugzilla 1196 - Spec examples corrections
1172 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1174 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1175 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1178 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1179 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1181 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1182 "acl {{name}{arg}...}", and optional args on acl condition
1185 JH/05 Permit multiple router/transport headers_add/remove lines.
1187 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1189 JH/07 Avoid using a waiting database for a single-message-only transport.
1190 Performance patch from Paul Fisher. Bugzilla 1262.
1192 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1195 JH/09 Add $headers_added variable, with content from use of ACL modifier
1196 add_header (but not yet added to the message). Bugzilla 199.
1198 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1199 Pulled from Bugzilla 817 by Wolfgang Breyha.
1201 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1203 (nb: this is the same fix as in Exim 4.80.1)
1205 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1208 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1211 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1213 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1214 advertise SMTP AUTH mechanism to us, instead of a generic
1215 protocol violation error. Also, make Exim more robust to bad
1216 data from the Dovecot auth socket.
1218 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1220 When a queue runner is handling a message, Exim first routes the
1221 recipient addresses, during which it prunes them based on the retry
1222 hints database. After that it attempts to deliver the message to
1223 any remaining recipients. It then updates the hints database using
1226 So if a recipient address works intermittently, it can get repeatedly
1227 deferred at routing time. The retry hints record remains fresh so the
1228 address never reaches the final cutoff time.
1230 This is a fairly common occurrence when a user is bumping up against
1231 their storage quota. Exim had some logic in its local delivery code
1232 to deal with this. However it did not apply to per-recipient defers
1233 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1235 This change adds a proper retry rule check during routing so that the
1236 final cutoff time is checked against the message's age. We only do
1237 this check if there is an address retry record and there is not a
1238 domain retry record; this implies that previous attempts to handle
1239 the address had the retry_use_local_parts option turned on. We use
1240 this as an approximation for the destination being like a local
1241 delivery, as in LMTP.
1243 I suspect this new check makes the old local delivery cutoff check
1244 redundant, but I have not verified this so I left the code in place.
1246 TF/02 Correct gecos expansion when From: is a prefix of the username.
1248 Test 0254 submits a message to Exim with the header
1252 When I ran the test suite under the user fanf2, Exim expanded
1253 the header to contain my full name, whereas it should have added
1254 a Resent-Sender: header. It erroneously treats any prefix of the
1255 username as equal to the username.
1257 This change corrects that bug.
1259 GF/01 DCC debug and logging tidyup
1260 Error conditions log to paniclog rather than rejectlog.
1261 Debug lines prefixed by "DCC: " to remove any ambiguity.
1263 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1265 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1266 Bug spotted by Jeremy Harris; was flawed since initial commit.
1267 Would have resulted in OCSP responses post-SNI triggering an Exim
1268 NULL dereference and crash.
1270 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1272 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1273 Bug detection, analysis and fix by Samuel Thibault.
1274 Bugzilla 1331, Debian bug #698092.
1276 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1278 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1279 Server implementation by Todd Lyons, client by JH.
1280 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1281 config variable "prdr_enable" controls whether the server
1282 advertises the facility. If the client requests PRDR a new
1283 acl_data_smtp_prdr ACL is called once for each recipient, after
1284 the body content is received and before the acl_smtp_data ACL.
1285 The client is controlled by both of: a hosts_try_prdr option
1286 on the smtp transport, and the server advertisement.
1287 Default client logging of deliveries and rejections involving
1288 PRDR are flagged with the string "PRDR".
1290 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1291 fclose(). Diagnosis by Todd Lyons.
1293 PP/17 Update configure.default to handle IPv6 localhost better.
1294 Patch by Alain Williams (plus minor tweaks).
1297 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1298 This is now consistent with GnuTLS, and is now documented: the
1299 previous undocumented portable approach to treating the option as
1300 unset was to force an expansion failure. That still works, and
1301 an empty string is now equivalent.
1303 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1304 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1305 not performing validation itself.
1307 PP/20 Added force_command boolean option to pipe transport.
1308 Patch from Nick Koston, of cPanel Inc.
1310 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1313 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1315 PP/21 Fix eximon continuous updating with timestamped log-files.
1316 Broken in a format-string cleanup in 4.80, missed when I repaired the
1317 other false fix of the same issue.
1318 Report and fix from Heiko Schlichting.
1321 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1322 Report from Prashanth Katuri.
1324 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1325 It's SecureTransport, so affects any MacOS clients which use the
1326 system-integrated TLS libraries, including email clients.
1328 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1329 using a MIME ACL for non-SMTP local injection.
1330 Report and assistance in diagnosis by Warren Baker.
1332 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1334 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1336 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1337 last id that failed so it may be referenced in subsequent ACL's.
1339 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1342 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1343 ldap library initialization, allowing self-signed CA's to be
1344 used. Also properly sets require_cert option later in code by
1345 using NULL (global ldap config) instead of ldap handle (per
1346 session). Bug diagnosis and testing by alxgomz.
1348 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1349 the src/util/ subdirectory.
1351 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1352 renamed to Transport Post Delivery Action by Jeremy Harris, as
1355 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1356 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1357 redis_servers = needs to be configured which will be used by the redis
1358 lookup. Patch from Warren Baker, of The Packet Hub.
1360 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1362 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1363 hostname or reverse DNS when processing a host list. Used suggestions
1364 from multiple comments on this bug.
1366 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1368 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1369 Missed a few lines, added it to make the runtest require no keyboard
1372 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1373 contains upper case chars. Make router use caseful_local_part.
1375 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1376 support when GnuTLS has been built with p11-kit.
1382 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1384 This, or similar/improved, will also be change PP/11 of 4.82.
1390 PP/01 Handle short writes when writing local log-files.
1391 In practice, only affects FreeBSD (8 onwards).
1392 Bugzilla 1053, with thanks to Dmitry Isaikin.
1394 NM/01 Bugzilla 949 - Documentation tweak
1396 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1399 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1401 PP/02 Implemented gsasl authenticator.
1403 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1405 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1406 `pkg-config foo` for cflags/libs.
1408 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1409 with rest of GSASL and with heimdal_gssapi.
1411 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1412 `pkg-config foo` for cflags/libs for the TLS implementation.
1414 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1415 properties get this fed in as external SSF. A number of robustness
1416 and debugging improvements to the cyrus_sasl authenticator.
1418 PP/08 cyrus_sasl server now expands the server_realm option.
1420 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1421 Patch by Jeremy Harris.
1423 PP/10 Added dbmjz lookup type.
1425 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1427 PP/12 MAIL args handles TAB as well as SP, for better interop with
1428 non-compliant senders.
1429 Analysis and variant patch by Todd Lyons.
1431 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1432 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1433 Patch from Dirk Mueller <dmueller@suse.com>
1435 PP/13 tls_peerdn now print-escaped for spool files.
1436 Observed some $tls_peerdn in wild which contained \n, which resulted
1437 in spool file corruption.
1439 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1440 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1441 or write after TLS renegotiation, which otherwise led to messages
1444 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1445 as a tracking header (ie: a signed header comes before the signature).
1446 Patch from Wolfgang Breyha.
1448 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1449 comma-sep list; embedded commas doubled.
1451 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1453 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1455 Report and patch from Dmitry Banschikov.
1457 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1458 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1459 failures. We appear to now support TLS1.1+ with Exim.
1461 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1462 lets Exim select keys and certificates based upon TLS SNI from client.
1463 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1464 before an outbound SMTP session. New log_selector, +tls_sni.
1466 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1467 NULL dereference. Report and patch from Alun Jones.
1469 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1470 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1471 Not seeing resolver debug output on NetBSD, but suspect this is a
1472 resolver implementation change.
1474 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1475 Left warnings. Added "eximon gdb" invocation mode.
1477 PP/21 Defaulting "accept_8bitmime" to true, not false.
1479 PP/22 Added -bw for inetd wait mode support.
1481 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1482 locate the relevant includes and libraries. Made this the default.
1484 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1485 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1487 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1488 This may cause build issues on older platforms.
1490 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1491 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1492 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1493 Added SNI support via GnuTLS too.
1494 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1496 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1498 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1499 Applied second patch from Janne, implementing suggestion to default
1500 multiple-strings-in-record handling to match SPF spec.
1502 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1504 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1505 read-only, out of scope).
1506 Patch from Wolfgang Breyha, report from Stuart Northfield.
1508 PP/29 Fix three issues highlighted by clang analyser static analysis.
1509 Only crash-plausible issue would require the Cambridge-specific
1510 iplookup router and a misconfiguration.
1511 Report from Marcin Mirosław.
1513 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1515 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1516 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1517 As part of this, removing so much warning spew let me fix some minor
1518 real issues in debug logging.
1520 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1521 assignment on my part. Fixed.
1523 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1524 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1525 Janne Snabb (who went above and beyond: thank you).
1527 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1528 string otherwise requires a connection and a bunch more work and it's
1529 relatively easy to get wrong. Should also expose TLS library linkage
1532 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1533 64-bit ${eval} (JH/03).
1535 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1536 GNU libc to support some of the 64-bit stuff, should not lead to
1537 conflicts. Defined before os.h is pulled in, so if a given platform
1538 needs to override this, it can.
1540 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1541 protection layer was required, which is not implemented.
1542 Bugzilla 1254, patch from Wolfgang Breyha.
1544 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1545 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1546 tls_dhparam take prime identifiers. Also unbreak combination of
1547 OpenSSL+DH_params+TLSSNI.
1549 PP/39 Disable SSLv2 by default in OpenSSL support.
1555 PP/01 Solaris build fix for Oracle's LDAP libraries.
1556 Bugzilla 1109, patch from Stephen Usher.
1558 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1560 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1563 TF/02 Fix a couple more cases where we did not log the error message
1564 when unlink() failed. See also change 4.74-TF/03.
1566 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1567 lock up or crash if it happened to be inside a call to libc when it
1568 got a SIGUSR1 from exiwhat.
1570 The SIGUSR1 handler appends the current process status to the process
1571 log which is later printed by exiwhat. It used to use the general
1572 purpose logging code to do this, but several functions it calls are
1573 not safe for signals.
1575 The new output code in the SIGUSR1 handler is specific to the process
1576 log, and simple enough that it's easy to inspect for signal safety.
1577 Removing some special cases also simplifies the general logging code.
1578 Removing the spurious timestamps from the process log simplifies
1581 TF/04 Improved ratelimit ACL condition.
1583 The /noupdate option has been deprecated in favour of /readonly which
1584 has clearer semantics. The /leaky, /strict, and /readonly update modes
1585 are mutually exclusive. The update mode is no longer included in the
1586 database key; it just determines when the database is updated. (This
1587 means that when you upgrade Exim will forget old rate measurements.)
1589 Exim now checks that the per_* options are used with an update mode that
1590 makes sense for the current ACL. For example, when Exim is processing a
1591 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1592 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1593 must specify per_mail/readonly. If you omit the update mode it defaults to
1594 /leaky where that makes sense (as before) or /readonly where required.
1596 The /noupdate option is now undocumented but still supported for
1597 backwards compatibility. It is equivalent to /readonly except that in
1598 ACLs where /readonly is required you may specify /leaky/noupdate or
1599 /strict/noupdate which are treated the same as /readonly.
1601 A useful new feature is the /count= option. This is a generalization
1602 of the per_byte option, so that you can measure the throughput of other
1603 aggregate values. For example, the per_byte option is now equivalent
1604 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1606 The per_rcpt option has been generalized using the /count= mechanism
1607 (though it's more complicated than the per_byte equivalence). When it is
1608 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1609 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1610 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1611 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1612 Note that using per_rcpt with a non-readonly update mode in more than
1613 one ACL will cause the recipients to be double-counted. (The per_mail
1614 and per_byte options don't have this problem.)
1616 The handling of very low rates has changed slightly. If the computed rate
1617 is less than the event's count (usually one) then this event is the first
1618 after a long gap. In this case the rate is set to the same as this event's
1619 count, so that the first message of a spam run is counted properly.
1621 The major new feature is a mechanism for counting the rate of unique
1622 events. The new per_addr option counts the number of different
1623 recipients that someone has sent messages to in the last time period. It
1624 behaves like per_rcpt if all the recipient addresses are different, but
1625 duplicate recipient addresses do not increase the measured rate. Like
1626 the /count= option this is a general mechanism, so the per_addr option
1627 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1628 example, measure the rate that a client uses different sender addresses
1629 with the options per_mail/unique=$sender_address. There are further
1630 details in the main documentation.
1632 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1634 TF/06 Removed a few PCRE remnants.
1636 TF/07 Automatically extract Exim's version number from tags in the git
1637 repository when doing development or release builds.
1639 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1640 Bugzilla 879. Patch from Paul Fisher.
1642 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1643 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1646 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1648 PP/05 Variable $av_failed, true if the AV scanner deferred.
1649 Bugzilla 1078. Patch from John Horne.
1651 PP/06 Stop make process more reliably on build failure.
1652 Bugzilla 1087. Patch from Heiko Schlittermann.
1654 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1655 Bugzilla 1089. Patch from Heiko Schlittermann.
1657 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1658 Bugzilla 1131. Patch from Holger Weiß.
1660 PP/09 Handle IPv6 addresses with SPF.
1661 Bugzilla 860. Patch from Wolfgang Breyha.
1663 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1665 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1668 PP/11 match_* no longer expand right-hand-side by default.
1669 New compile-time build option, EXPAND_LISTMATCH_RHS.
1670 New expansion conditions, "inlist", "inlisti".
1672 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1674 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1676 PP/14 fix log_write() format string regression from TF/03.
1677 Bugzilla 1152. Patch from Dmitry Isaikin.
1683 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1685 PP/02 Harmonised TLS library version reporting; only show if debugging.
1686 Layout now matches that introduced for other libraries in 4.74 PP/03.
1688 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1690 PP/04 New "dns_use_edns0" global option.
1692 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1695 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1696 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1698 TK/01 Updated PolarSSL code to 0.14.2.
1699 Bugzilla 1097. Patch from Andreas Metzler.
1701 PP/07 Catch divide-by-zero in ${eval:...}.
1702 Fixes bugzilla 1102.
1704 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1707 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1708 format-string attack -- SECURITY: remote arbitrary code execution.
1710 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1711 time unintentionally subject to list matching rules, letting the header
1712 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1713 arbitrary string expansion). This allowed for information disclosure.
1715 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1716 INT_MIN/-1 -- value coerced to INT_MAX.
1722 NM/01 Workaround for PCRE version dependency in version reporting
1725 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1726 This fixes portability to compilers other than gcc, notably
1727 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1729 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1730 makefiles for portability to HP-UX and POSIX correctness.
1732 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1733 Also via indented variable definition in the Makefile.
1734 (Debugging by Oliver Heesakkers).
1736 PP/02 Restore caching of spamd results with expanded spamd_address.
1737 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1739 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1740 Improves build reliability. Fix from: Frank Elsner
1742 NM/02 Fix wide character breakage in the rfc2047 coding
1743 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1745 NM/03 Allow underscore in dnslist lookups
1746 Fixes bug 1026. Patch from Graeme Fowler
1748 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1749 Code patches from Adam Ciarcinski of NetBSD.
1751 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1754 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1755 is logged, to avoid truncation. Patch from John Horne.
1757 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1758 Patch from Jakob Hirsch.
1760 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1761 SQL string expansion failure details.
1762 Patch from Andrey Oktyabrski.
1764 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1765 Patch from Simon Arlott.
1767 PP/09 New lookups functionality failed to compile on old gcc which rejects
1768 extern declarations in function scope.
1769 Patch from Oliver Fleischmann
1771 PP/10 Use sig_atomic_t for flags set from signal handlers.
1772 Check getgroups() return and improve debugging.
1773 Fixed developed for diagnosis in bug 927 (which turned out to be
1776 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1777 Patch from Mark Zealey.
1779 PP/12 Bugzilla 1056: Improved spamd server selection.
1780 Patch from Mark Zealey.
1782 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1783 Based on patch from Heiko Schlittermann.
1785 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1786 Patch from Uwe Doering, sign-off by Michael Haardt.
1788 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1789 variable declaration deep within a block. Bug and patch from
1792 PP/15 lookups-Makefile IRIX compatibility coercion.
1794 PP/16 Make DISABLE_DKIM build knob functional.
1796 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1797 Patch by Simon Arlott
1799 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1800 variable argument macros. Our copy now differs from upstream.
1806 TF/01 Failure to get a lock on a hints database can have serious
1807 consequences so log it to the panic log.
1809 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1810 controlled using the smtp_confirmation log selector.
1812 TF/03 Include the error message when we fail to unlink a spool file.
1814 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1815 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1816 for maintaining out-of-tree patches for some time.
1818 PP/01 Bugzilla 139: Documentation and portability issues.
1819 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1820 Handle per-OS dynamic-module compilation flags.
1822 PP/02 Let /dev/null have normal permissions.
1823 The 4.73 fixes were a little too stringent and complained about the
1824 permissions on /dev/null. Exempt it from some checks.
1825 Reported by Andreas M. Kirchwitz.
1827 PP/03 Report version information for many libraries, including
1828 Exim version information for dynamically loaded libraries. Created
1829 version.h, now support a version extension string for distributors
1830 who patch heavily. Dynamic module ABI change.
1832 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1833 privilege escalation vulnerability whereby the Exim run-time user
1834 can cause root to append content of the attacker's choosing to
1837 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1840 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1841 If dropping privileges for untrusted macros, we disabled normal logging
1842 on the basis that it would fail; for the Exim run-time user, this is not
1843 the case, and it resulted in successful deliveries going unlogged.
1844 Fixed. Reported by Andreas Metzler.
1850 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1851 only prepend for the Resent-* case. Fixes regression introduced in
1852 Exim 4.70 by NM/22 for Bugzilla 607.
1854 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1855 increasing numbers of administrators be bitten by this.
1857 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1859 PP/03 Bugzilla 994: added openssl_options main configuration option.
1861 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1863 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1865 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1867 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1868 without a peer certificate, leading to a segfault because of an
1869 assumption that peers always have certificates. Be a little more
1870 paranoid. Problem reported by Martin Tscholak.
1872 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1873 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1874 NB: ClamAV planning to remove STREAM in "middle of 2010".
1875 CL also introduces -bmalware, various -d+acl logging additions and
1876 more caution in buffer sizes.
1878 PP/09 Implemented reverse_ip expansion operator.
1880 PP/10 Bugzilla 937: provide a "debug" ACL control.
1882 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1884 PP/12 Bugzilla 973: Implement --version.
1886 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1888 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1890 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1892 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1893 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1894 ignore trailing whitespace.
1896 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1898 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1901 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1902 Notification from Dr Andrew Aitchison.
1904 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1905 ExtendedDetectionInfo response format.
1906 Notification from John Horne.
1908 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1911 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1912 XSL and documented dependency on system catalogs, with examples of how
1915 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1918 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1919 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1920 configuration file which is writeable by the Exim user or group.
1922 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1923 of configuration files to cover files specified with the -C option if
1924 they are going to be used with root privileges, not just the default
1927 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1928 option (effectively making it always true).
1930 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1931 files to be used while preserving root privileges.
1933 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1934 that rogue child processes cannot use them.
1936 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1937 run-time user, instead of root.
1939 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1940 Exim run-time user without dropping privileges.
1942 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1943 result string, instead of calling string_vformat() twice with the same
1946 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1947 for other users. Others should always drop root privileges if they use
1948 -C on the command line, even for a whitelisted configure file.
1950 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1952 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1958 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1959 $data_path, and $header_path variables; fixed documentation bugs and
1962 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1963 exipick to access non-standard spools, including the "frozen" queue
1966 NM/01 Bugzilla 965: Support mysql stored procedures.
1967 Patch from Alain Williams
1969 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1971 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1972 Patch from Andreas Metzler
1974 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1975 Patch from Kirill Miazine
1977 NM/05 Bugzilla 671: Added umask to procmail example.
1979 JJ/03 installed exipick 20100323.0, fixing doc bug
1981 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1982 directory. Notification and patch from Dan Rosenberg.
1984 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1986 TK/02 Improve log output when DKIM signing operation fails.
1988 MH/01 Treat the transport option dkim_domain as a colon separated
1989 list, not as a single string, and sign the message with each element,
1990 omitting multiple occurences of the same signer.
1992 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1993 Bugzilla 985, 986. Patch by Simon Arlott
1995 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1996 Patch by Simon Arlott
1998 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1999 MBX locking. Notification from Dan Rosenberg.
2005 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2007 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2009 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2011 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2013 NM/04 Bugzilla 915: Flag broken perl installation during build.
2019 TK/01 Added patch by Johannes Berg that expands the main option
2020 "spamd_address" if it starts with a dollar sign.
2022 TK/02 Write list of recipients to X-Envelope-Sender header when building
2023 the mbox-format spool file for content scanning (suggested by Jakob
2026 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2027 (http://www.dcc-servers.net/) support via dccifd. Activated by
2028 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2030 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2031 by Mark Daniel Reidel <mr@df.eu>.
2033 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2034 When building exim an external PCRE library is now needed -
2035 PCRE is a system library on the majority of modern systems.
2036 See entry on PCRE_LIBS in EDITME file.
2038 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2039 conversation. Added nologin parameter to request.
2040 Patch contributed by Kirill Miazine.
2042 TF/01 Do not log submission mode rewrites if they do not change the address.
2044 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2046 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2047 log files in place. Contributed by Roberto Lima.
2049 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2051 TF/03 Bugzilla 615: When checking the local_parts router precondition
2052 after a local_part_suffix or local_part_prefix option, Exim now
2053 does not use the address's named list lookup cache, since this
2054 contains cached lookups for the whole local part.
2056 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2057 Robert Millan. Documentation is in experimental-spec.txt.
2059 TF/04 Bugzilla 668: Fix parallel build (make -j).
2061 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2063 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2064 Patch provided by Jan Srzednicki.
2066 TF/05 Leading white space used to be stripped from $spam_report which
2067 wrecked the formatting. Now it is preserved.
2069 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2070 that they are available at delivery time.
2072 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2074 TF/08 TLS error reporting now respects the incoming_interface and
2075 incoming_port log selectors.
2077 TF/09 Produce a more useful error message if an SMTP transport's hosts
2078 setting expands to an empty string.
2080 NM/06 Bugzilla 744: EXPN did not work under TLS.
2081 Patch provided by Phil Pennock.
2083 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2084 Patch provided by Richard Godbee.
2086 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2087 acl_smtp_notquit, added index entry.
2089 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2090 Patch provided by Eugene Bujak.
2092 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2093 accept(). Patch provided by Maxim Dounin.
2095 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2096 Patch provided by Phil Pennock.
2098 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2100 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2101 Patch provided by Brad "anomie" Jorsch.
2103 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2104 Patch provided by Dean Brooks.
2106 TK/05 Add native DKIM support (does not depend on external libraries).
2108 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2109 Patch provided by Graeme Fowler.
2111 NM/16 Bugzilla 851: Documentation example syntax fix.
2113 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2115 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2118 NM/19 Bugzilla 745: TLS version reporting.
2119 Patch provided by Phil Pennock.
2121 NM/20 Bugzilla 167: bool: condition support.
2122 Patch provided by Phil Pennock.
2124 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2125 clients. Patch provided by Phil Pennock.
2127 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2128 Patch provided by Brad "anomie" Jorsch.
2130 NM/23 Bugzilla 687: Fix misparses in eximstats.
2131 Patch provided by Heiko Schlittermann.
2133 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2134 Patch provided by Heiko Schlittermann.
2136 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2137 plus update to original patch.
2139 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2141 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2142 Patch provided by David Brownlee.
2144 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2146 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2148 NM/30 Bugzilla 888: TLS documentation bugfixes.
2150 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2152 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2153 Unlike the original bugzilla I have changed all shell scripts in src tree.
2155 NM/33 Bugzilla 898: Transport filter timeout fix.
2156 Patch by Todd Rinaldo.
2158 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2159 Patch by Serge Demonchaux.
2161 NM/35 Bugzilla 39: Base64 decode bug fixes.
2162 Patch by Jakob Hirsch.
2164 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2166 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2168 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2170 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2176 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2177 ALT-N's libdkim that I have put here:
2178 http://duncanthrax.net/exim-experimental/
2180 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2181 were called 'true' and it seems that C99 defines that as a reserved
2182 keyword to be used with 'bool' variable types. That means you could
2183 not include C99-style headers which use bools without triggering
2184 build errors in sieve.c.
2186 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2187 as mailq or other aliases. Changed the --help handling significantly
2188 to do whats expected. exim_usage() emits usage/help information.
2190 SC/01 Added the -bylocaldomain option to eximstats.
2192 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2194 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2196 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2202 PH/01 Another patch from the Sieve maintainer.
2204 PH/02 When an IPv6 address is converted to a string for single-key lookup
2205 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2206 dots are used instead of colons so that keys in lsearch files need not
2207 contain colons. This was done some time before quoting was made available
2208 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2209 (notated using the quote facility) so as to distinguish them from IPv4
2210 keys. This meant that lookups for IP addresses in host lists did not work
2211 for iplsearch lookups.
2213 This has been fixed by arranging for IPv6 addresses to be expressed with
2214 colons if the lookup type is iplsearch. This is not incompatible, because
2215 previously such lookups could never work.
2217 The situation is now rather anomalous, since one *can* have colons in
2218 ordinary lsearch keys. However, making the change in all cases is
2219 incompatible and would probably break a number of configurations.
2221 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2224 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2225 conversion specification without a maximum field width, thereby enabling
2226 a rogue spamd server to cause a buffer overflow. While nobody in their
2227 right mind would setup Exim to query an untrusted spamd server, an
2228 attacker that gains access to a server running spamd could potentially
2229 exploit this vulnerability to run arbitrary code as the Exim user.
2231 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2232 $primary_hostname instead of what libspf2 thinks the hosts name is.
2234 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2235 a directory entry by the name of the lookup key. Previously, if a
2236 symlink pointed to a non-existing file or a file in a directory that
2237 Exim lacked permissions to read, a lookup for a key matching that
2238 symlink would fail. Now it is enough that a matching directory entry
2239 exists, symlink or not. (Bugzilla 503.)
2241 PH/03 The body_linecount and body_zerocount variables are now exported in the
2244 PH/04 Added the $dnslist_matched variable.
2246 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2247 This means they are set thereafter only if the connection becomes
2250 PH/06 Added the client_condition to authenticators so that some can be skipped
2251 by clients under certain conditions.
2253 PH/07 The error message for a badly-placed control=no_multiline_responses left
2254 "_responses" off the end of the name.
2256 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2258 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2259 (without spaces) instead of just copying the configuration text.
2261 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2263 PH/11 Added $max_received_linelength.
2265 PH/12 Added +ignore_defer and +include_defer to host lists.
2267 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2268 way in which PCRE > 7.0 is built.
2270 PH/14 Implemented queue_only_load_latch.
2272 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2273 MAIL command. The effect was to mangle the value on 64-bit systems.
2275 PH/16 Another patch from the Sieve maintainer.
2277 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2279 PH/18 If a system quota error occurred while trying to create the file for
2280 a maildir delivery, the message "Mailbox is full" was not appended to the
2281 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2282 only to a quota excession during the actual writing of the file.
2284 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2285 characters?) which causes problems in log lines. The DN values are now
2286 passed through string_printing() before being added to log lines.
2288 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2289 and InterBase are left for another time.)
2291 PH/21 Added message_body_newlines option.
2293 PH/22 Guard against possible overflow in moan_check_errorcopy().
2295 PH/23 POSIX allows open() to be a macro; guard against that.
2297 PH/24 If the recipient of an error message contained an @ in the local part
2298 (suitably quoted, of course), incorrect values were put in $domain and
2299 $local_part during the evaluation of errors_copy.
2305 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2306 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2309 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2310 issue a MAIL command.
2312 PH/02 In an ACL statement such as
2314 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2316 if a client was not listed at all, or was listed with a value other than
2317 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2318 the condition was not true (as it should be), so access was not denied.
2319 The bug was that the ! inversion was incorrectly passed on to the second
2320 item. This has been fixed.
2322 PH/03 Added additional dnslists conditions == and =& which are different from
2323 = and & when the dns lookup returns more than one IP address.
2325 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2326 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2328 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2329 FSYNC, which compiles an option called disable_fsync that allows for
2330 bypassing fsync(). The documentation is heavily laced with warnings.
2332 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2334 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2335 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2336 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2337 including adding "make clean"; (3) Added -fPIC when compiling the test
2338 dynamically loaded module, to get rid of a warning.
2340 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2341 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2342 = 0s. The bug is otherwise harmless.
2344 PH/07 There was a bug in the dovecot authenticator such that the value of
2345 $auth1 could be overwritten, and so not correctly preserved, after a
2346 successful authentication. This usually meant that the value preserved by
2347 the server_setid option was incorrect.
2349 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2351 PH/09 Installed PCRE release 7.0.
2353 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2354 run for batched SMTP input. It is now run at the start of every message
2355 in the batch. While fixing this I discovered that the process information
2356 (output by running exiwhat) was not always getting set for -bs and -bS
2357 input. This is fixed, and it now also says "batched" for BSMTP.
2359 PH/11 Added control=no_pipelining.
2361 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2362 patch, slightly modified), and move the expansion of helo_data till after
2363 the connection is made in the smtp transport (so it can use these
2366 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2368 PH/14 Added log_selector = +pid.
2370 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2372 PH/16 Add ${if forany and ${if forall.
2374 PH/17 Added dsn_from option to vary the From: line in DSNs.
2376 PH/18 Flush SMTP output before performing a callout, unless control =
2377 no_callout_flush is set.
2379 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2380 was true (the default) a successful delivery failed to delete the retry
2381 item, thus causing premature timeout of the address. The bug is now
2384 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2386 PH/21 Long custom messages for fakedefer and fakereject are now split up
2387 into multiline responses in the same way that messages for "deny" and
2388 other ACL rejections are.
2390 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2391 with slight modification.
2393 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2394 draft, changing the syntax and factoring some duplicate code".
2396 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2397 for deliveries of the second and subsequent messages over the same SMTP
2400 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2401 ${reduce, with only minor "tidies".
2403 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2405 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2406 expansion side effects.
2408 PH/27 When a message times out after an over-quota error from an Exim-imposed
2409 quota, the bounce message says "mailbox is full". This message was not
2410 being given when it was a system quota that was exceeded. It now should
2413 MH/03 Made $recipients available in local_scan(). local_scan() already has
2414 better access to the recipient list through recipients_list[], but
2415 $recipients can be useful in postmaster-provided expansion strings.
2417 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2418 in the case of a MAIL command with additional options following the
2419 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2420 were accidentally chopped off.
2422 PH/29 SMTP synchronization checks are implemented when a command is read -
2423 there is a check that no more input is waiting when there shouldn't be
2424 any. However, for some commands, a delay in an ACL can mean that it is
2425 some time before the response is written. In this time, more input might
2426 arrive, invalidly. So now there are extra checks after an ACL has run for
2427 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2428 pipelining has not been advertised.
2430 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2432 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2433 returned in the SMTP response when smtp_return_error_details was false.
2434 This has been fixed.
2436 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2437 instead of the C I/O that was originally supplied, because problems were
2438 reported on Solaris.
2440 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2441 Exim which did not show up earlier: it was assuming that a call to
2442 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2443 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2444 was a macro that became an assignment, so it seemed to work. This has
2445 changed to a proper function call with a void return, hence the compile
2446 error. Exim's code has been fixed.
2448 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2451 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2453 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2455 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2456 because it turns out to be too big or there is a timeout) there is no
2457 "Completed" line in the log. When some messages of this type were
2458 selected by exigrep, they were listed as "not completed". Others were
2459 picked up by some special patterns. I have improved the selection
2460 criteria to be more general.
2462 PH/38 The host_find_failed option in the manualroute router can now be set
2463 to "ignore", to completely ignore a host whose IP address cannot be
2464 found. If all hosts are ignored, the behaviour is controlled by the new
2465 host_all_ignored option.
2467 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2468 homing or because of multiple MX records with /mx) generated more than
2469 one IP address, and the following item turned out to be the local host,
2470 all the secondary addresses of the first item were incorrectly removed
2471 from the list, along with the local host and any following hosts (which
2472 is what is supposed to happen).
2474 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2475 whoever called Exim into the -H file. In the case of the daemon it was
2476 behaving confusingly. When first started, it used values for whoever
2477 started the daemon, but after a SIGHUP it used the Exim user (because it
2478 calls itself on a restart). I have changed the code so that it now always
2481 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2482 message are rejected with the same error (e.g. no authentication or bad
2483 sender address), and a DATA command is nevertheless sent (as can happen
2484 with PIPELINING or a stupid MUA), the error message that was given to the
2485 RCPT commands is included in the rejection of the DATA command. This is
2486 intended to be helpful for MUAs that show only the final error to their
2489 PH/42 Another patch from the Sieve maintainer.
2491 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2492 Eximstats - Fixed some broken HTML links and added missing column headers
2494 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2495 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2497 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2499 PH/43 Yet another patch from the Sieve maintainer.
2501 PH/44 I found a way to check for a TCP/IP connection going away before sending
2502 the response to the final '.' that terminates a message, but only in the
2503 case where the client has not sent further data following the '.'
2504 (unfortunately, this is allowed). However, in many cases there won't be
2505 any further data because there won't be any more messages to send. A call
2506 to select() can be used: if it shows that the input is "ready", there is
2507 either input waiting, or the socket has been closed. An attempt to read
2508 the next input character can distinguish the two cases. Previously, Exim
2509 would have sent an OK response which the client would never have see.
2510 This could lead to message repetition. This fix should cure that, at
2511 least in a lot of common cases.
2513 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2514 advertised in response to EHLO.
2520 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2521 fixed by 4.65/MH/01 (is this a record?) are fixed:
2523 (i) An empty string was always treated as zero by the numeric comparison
2524 operators. This behaviour has been restored.
2526 (ii) It is documented that the numeric comparison operators always treat
2527 their arguments as decimal numbers. This was broken in that numbers
2528 starting with 0 were being interpreted as octal.
2530 While fixing these problems I realized that there was another issue that
2531 hadn't been noticed. Values of message_size_limit (both the global option
2532 and the transport option) were treated as octal if they started with 0.
2533 The documentation was vague. These values are now always treated as
2534 decimal, and I will make that clear in the documentation.
2540 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2541 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2544 MH/01 Don't check that the operands of numeric comparison operators are
2545 integers when their expansion is in "skipping" mode (fixes bug
2546 introduced by 4.64-PH/07).
2548 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2549 child addresses, Exim now panics and dies. Previously, because the count
2550 is held in a short int, deliveries were likely to be lost. As such a
2551 large number of recipients for a single message is ridiculous
2552 (performance will be very, very poor), I have chosen to impose a limit
2553 rather than extend the field.
2559 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2560 leftover -K file (the existence of which was triggered by #402).
2561 While we were at it, introduced process PID as part of the -K
2562 filename. This should rule out race conditions when creating
2565 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2566 processing considerably. Previous code took too long for large mails,
2567 triggering a timeout which in turn triggers #401.
2569 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2570 in the DK code in transports.c. sendfile() is not really portable,
2571 hence the _LINUX specificness.
2573 TF/01 In the add_headers option to the mail command in an Exim filter,
2574 there was a bug that Exim would claim a syntax error in any
2575 header after the first one which had an odd number of characters
2578 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2579 callout verification, Exim cached a "reject" for the entire domain. This
2580 is correct for most verifications, but it is not correct for a recipient
2581 verification with use_sender or use_postmaster set, because in that case
2582 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2583 case of MAIL FROM:<> rejection from other early rejections (e.g.
2584 rejection of HELO). When verifying a recipient using a non-null MAIL
2585 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2586 Whatever the result of the callout, the value of the domain cache is
2587 left unchanged (for any other kind of callout, getting as far as trying
2588 RCPT means that the domain itself is ok).
2590 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2593 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2594 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2595 session, the daemon ignores ECONNECT errors and logs others; it now
2596 ignores EPIPE as well.
2598 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2599 (quoted-printable decoding).
2601 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2602 later the small subsequent patch to fix an introduced bug.
2604 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2606 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2608 PH/08 An error is now given if message_size_limit is specified negative.
2610 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2611 to be given (somewhat) arbitrary names.
2613 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2616 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2617 miscellaneous code fixes
2619 PH/10 Added the log_reject_target ACL modifier to specify where to log
2622 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2623 hostname. This is wrong, because it relates to the incoming message (and
2624 probably the interface on which it is arriving) and not to the outgoing
2625 callout (which could be using a different interface). This has been
2626 changed to use the value of the helo_data option from the smtp transport
2627 instead - this is what is used when a message is actually being sent. If
2628 there is no remote transport (possible with a router that sets up host
2629 addresses), $smtp_active_hostname is used.
2631 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2632 tweaks were necessary in order to get it to work (see also 21 below):
2633 (a) The code assumed that strncpy() returns a negative number on buffer
2634 overflow, which isn't the case. Replaced with Exim's string_format()
2636 (b) There were several signed/unsigned issues. I just did the minimum
2637 hacking in of casts. There is scope for a larger refactoring.
2638 (c) The code used strcasecmp() which is not a standard C function.
2639 Replaced with Exim's strcmpic() function.
2640 (d) The code set only $1; it now sets $auth1 as well.
2641 (e) A simple test gave the error "authentication client didn't specify
2642 service in request". It would seem that Dovecot has changed its
2643 interface. Fortunately there's a specification; I followed it and
2644 changed what the client sends and it appears to be working now.
2646 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2649 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2650 address A is aliased to B and C, where B exists and C does not. Without
2651 -v the output is "A verified" because verification stops after a
2652 successful redirection if more than one address is generated. However,
2653 with -v the child addresses are also verified. Exim was outputting "A
2654 failed to verify" and then showing the successful verification for C,
2655 with its parentage. It now outputs "B failed to verify", showing B's
2656 parentage before showing the successful verification of C.
2658 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2659 look up a TXT record in a specific list after matching in a combined
2662 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2663 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2664 they consult the DNS. I had assumed they would set it the way they
2665 wanted; and indeed my experiments on Linux seem to show that in some
2666 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2667 To be on the safe side, however, I have now made the interface to
2668 host_find_byname() similar to host_find_bydns(), with an argument
2669 containing the DNS resolver options. The host_find_byname() function now
2670 sets these options at its start, just as host_find_bydns() does. The smtp
2671 transport options dns_qualify_single and dns_search_parents are passed to
2672 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2673 of host_find_byname() use the default settings of RES_DEFNAMES
2674 (qualify_single) but not RES_DNSRCH (search_parents).
2676 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2677 spool_read_header() do less string testing, by means of a preliminary
2678 switch on the second character of optional "-foo" lines. (This is
2679 overdue, caused by the large number of possibilities that now exist.
2680 Originally there were few.) While I was there, I also converted the
2681 str(n)cmp tests so they don't re-test the leading "-" and the first
2682 character, in the hope this might squeeze out yet more improvement.
2684 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2685 flag allowing group syntax was set by the header_syntax check but not
2686 turned off, possible causing trouble later; (2) The flag was not being
2687 set at all for the header_verify test, causing "group"-style headers to
2688 be rejected. I have now set it in this case, and also caused header_
2689 verify to ignore an empty address taken from a group. While doing this, I
2690 came across some other cases where the code for allowing group syntax
2691 while scanning a header line wasn't quite right (mostly, not resetting
2692 the flag correctly in the right place). These bugs could have caused
2693 trouble for malformed header lines. I hope it is now all correct.
2695 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2696 with the "reply" argument non-NULL. The code, however (which originally
2697 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2698 but it didn't always do it. This confused somebody who was copying the
2699 code for some other use. I have removed all the tests.
2701 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2702 feature that was used to support insecure browsers during the U.S. crypto
2703 embargo. It requires special client support, and Exim is probably the
2704 only MTA that supported it -- and would never use it because real RSA is
2705 always available. This code has been removed, because it had the bad
2706 effect of slowing Exim down by computing (never used) parameters for the
2707 RSA_EXPORT functionality.
2709 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2710 authenticator to fail if there's a tab character in the incoming data
2711 (there should never be unless someone is messing about, as it's supposed
2712 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2713 if the connection is using TLS or if the remote IP is the same as the
2714 local IP, and the "valid-client-cert option" if a client certificate has
2717 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2718 authenticators. This can be used for authorization after authentication
2719 succeeds. (In the case of plaintext, it servers for both authentication
2722 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2723 if any retry times were supplied.
2725 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2726 connection, where there is no client IP address to check. In this
2727 situation, the verify now always succeeds.
2729 PH/25 Applied John Jetmore's -Mset patch.
2731 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2733 PH/27 In a string expansion for a processed (not raw) header when multiple
2734 headers of the same name were present, leading whitespace was being
2735 removed from all of them, but trailing whitespace was being removed only
2736 from the last one. Now trailing whitespace is removed from each header
2737 before concatenation. Completely empty headers in a concatenation (as
2738 before) are ignored.
2740 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2741 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2743 PH/29 [Removed. This was a change that I later backed out, and forgot to
2744 correct the ChangeLog entry (that I had efficiently created) before
2745 committing the later change.]
2747 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2748 address errors (4xx response to RCPT) over the same connection as other
2749 messages routed to the same hosts. Such deliveries are always "forced",
2750 so retry times are not inspected. This resulted in far too many retries
2751 for the affected addresses. The effect occurred only when there were more
2752 hosts than the hosts_max_try setting in the smtp transport when it had
2753 the 4xx errors. Those hosts that it had tried were not added to the list
2754 of hosts for which the message was waiting, so if all were tried, there
2755 was no problem. Two fixes have been applied:
2757 (i) If there are any address or message errors in an SMTP delivery, none
2758 of the hosts (tried or untried) are now added to the list of hosts
2759 for which the message is waiting, so the message should not be a
2760 candidate for sending over the same connection that was used for a
2761 successful delivery of some other message. This seems entirely
2762 reasonable: after all the message is NOT "waiting for some host".
2763 This is so "obvious" that I'm not sure why it wasn't done
2764 previously. Hope I haven't missed anything, but it can't do any
2765 harm, as the worst effect is to miss an optimization.
2767 (ii) If, despite (i), such a delivery is accidentally attempted, the
2768 routing retry time is respected, so at least it doesn't keep
2769 hammering the server.
2771 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2772 in ${readsocket because some servers need this prod.
2774 PH/32 Added some extra debug output when updating a wait-xxx database.
2776 PH/33 The hint "could be header name not terminated by colon", which has been
2777 given for certain expansion errors for a long time, was not being given
2778 for the ${if def:h_colon_omitted{... case.
2780 PH/34 The spec says: "With one important exception, whenever a domain list is
2781 being scanned, $domain contains the subject domain." There was at least
2782 one case where this was not true.
2784 PH/35 The error "getsockname() failed: connection reset by peer" was being
2785 written to the panic log as well as the main log, but it isn't really
2786 panic-worthy as it just means the connection died rather early on. I have
2787 removed the panic log writing for the ECONNRESET error when getsockname()
2790 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2791 runs only) independently of the message's sender address. This meant
2792 that, if the 4xx error was in fact related to the sender, a different
2793 message to the same recipient with a different sender could confuse
2794 things. In particular, this can happen when sending to a greylisting
2795 server, but other circumstances could also provoke similar problems.
2796 I have changed the default so that the retry time for these errors is now
2797 based a combination of the sender and recipient addresses. This change
2798 can be overridden by setting address_retry_include_sender=false in the
2801 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2802 remote server are returned as part of bounce messages. This was not
2803 happening for LMTP over a pipe (the lmtp transport), but now it is the
2804 same for both kinds of LMTP.
2806 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2807 in header lines that were in fact CNAMEs. This is no longer the case.
2809 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2810 and queue runs started by the daemon processed all messages. This has
2811 been fixed so that -R and -S can now usefully be given with -q<time>.
2813 PH/40 Import PCRE release 6.7 (fixes some bugs).
2815 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2817 PH/42 Give an error if -q is specified more than once.
2819 PH/43 Renamed the variables $interface_address and $interface_port as
2820 $received_ip_address and $received_port, to make it clear that these
2821 values apply to message reception, and not to the outgoing interface when
2822 a message is delivered. (The old names remain recognized, of course.)
2824 PH/44 There was no timeout on the connect() call when using a Unix domain
2825 socket in the ${readsocket expansion. There now is.
2827 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2828 be meaningful with "accept".
2830 SC/01 Eximstats V1.43
2831 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2833 SC/02 Eximstats V1.44
2834 Use a glob alias rather than an array ref in the generated
2835 parser. This improves both readability and performance.
2837 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2838 Collect SpamAssassin and rejection statistics.
2839 Don't display local sender or destination tables unless
2840 there is data to show.
2841 Added average volumes into the top table text output.
2843 SC/04 Eximstats V1.46
2844 Collect data on the number of addresses (recipients)
2845 as well as the number of messages.
2847 SC/05 Eximstats V1.47
2848 Added 'Message too big' to the list of mail rejection
2849 reasons (thanks to Marco Gaiarin).
2851 SC/06 Eximstats V1.48
2852 Mainlog lines which have GMT offsets and are too short to
2853 have a flag are now skipped.
2855 SC/07 Eximstats V1.49 (Alain Williams)
2856 Added the -emptyok flag.
2858 SC/08 Eximstats V1.50
2859 Fixes for obtaining the IP address from reject messages.
2861 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2862 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2863 whitespace changes from 4.64-PH/27
2865 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2868 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2869 are found, allow negative numbers in numeric criteria)
2871 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2873 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2874 to match changes made in 4.64-PH/43
2876 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2878 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2879 "domain" is actually an IP address, and force a failure. This locks out
2880 those revolvers/nameservers that support "A-for-A" lookups, in
2881 contravention of the specifications.
2883 PH/48 When a host name was looked up from an IP address, and the subsequent
2884 forward lookup of the name timed out, the host name was left in
2885 $sender_host_name, contrary to the specification.
2887 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2888 restricted to single-key lookups, Exim was not diagnosing an error if
2889 * or *@ was used with a query-style lookup.
2891 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2893 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2894 long ago, but noone interested enough thought of it. Let's just say that
2895 the "1.1" means that there are some new functions that weren't there at
2896 some point in the past.
2898 PH/51 Error processing for expansion failure of helo_data from an smtp
2899 transport during callout processing was broken.
2901 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2902 tested/used via the -bh/-bhc/-bs options.
2904 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2905 bug, fixed in subsequent PCRE releases).
2907 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2908 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2910 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2916 SC/01 Use a glob alias rather than an array ref in eximstats generated
2917 parser. This improves both readability and performance.
2919 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2920 Don't display local sender or destination tables in eximstats unless
2921 there is data to show.
2922 Added average volumes into the eximstats top table text output.
2924 SC/03 Collect data on the number of addresses (recipients) as well
2925 as the number of messages in eximstats.
2927 TF/01 Correct an error in the documentation for the redirect router. Exim
2928 does (usually) call initgroups() when daemonizing.
2930 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2931 with consistent privilege compared to when running as a daemon.
2933 TF/03 Note in the spec that $authenticated_id is not set for local
2934 submissions from trusted users.
2936 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2937 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2939 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2940 by adding some example configuration directives to the default
2941 configuration file. A little bit of work is required to uncomment the
2942 directives and define how usernames and passwords are checked, but
2943 there is now a framework to start from.
2945 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2946 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2947 without this. I don't know how relevant this is to other LDAP libraries.
2949 PH/02 Add the verb name to the "unknown ACL verb" error.
2951 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2953 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2955 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2956 directory not expanded when it should be if an expanded home directory
2957 was set for the address (which is overridden by the transport).
2959 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2962 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2963 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2964 because it is too late at that time, and has no effect.
2966 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2967 security issue with \' (bugzilla #107). I could not use the
2968 PQescapeStringConn() function, because it needs a PGconn value as one of
2971 PH/08 When testing addresses using -bt, indicate those final addresses that
2972 are duplicates that would not cause an additional delivery. At least one
2973 person was confused, thinking that -bt output corresponded to deliveries.
2974 (Suppressing duplicates isn't a good idea as you lose the information
2975 about possibly different redirections that led to the duplicates.)
2977 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2978 systems where poll() doesn't work, in particular OS X.
2980 PH/10 Added more information to debugging output for retry time not reached.
2982 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2983 operations in malware.c.
2985 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2988 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2989 syslog_duplication set false, log lines that would normally be written
2990 both the the main log and to the reject log were not written to syslog at
2993 PH/14 In the default configuration, change the use of "message" in ACL warn
2994 statements to "add_header".
2996 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2997 not followed by a command (e.g. "seen endif").
2999 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3000 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3003 PH/17 Added extra conditions to the default value of delay_warning_condition
3007 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3008 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3009 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3012 The Auto-Submitted: and various List- headers are standardised, whereas I
3013 don't think Precedence: ever was.
3015 PH/18 Refactored debugging code in route_finduser() to show more information,
3016 in particular, the error code if getpwnam() issues one.
3018 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3019 This is apparently needed in addition to the PH/07 change above to avoid
3020 any possible encoding problems.
3022 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3023 but not after initializing Perl.
3025 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3026 output them only if debugging. By default they are written stderr,
3027 apparently, which is not desirable.
3029 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3032 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3035 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3037 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3038 authenticated or an ident call has been made. Suppress the default
3039 values for $authenticated_id and $authenticated_sender (but permit -oMai
3040 and -oMas) when testing with -bh.
3042 PH/24 Re-jigged the order of the tests in the default configuration so that the
3043 tests for valid domains and recipients precede the DNS black list and CSA
3044 tests, on the grounds that those ones are more expensive.
3046 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3047 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3048 This bug exists in every version of Exim that I still have, right back to
3051 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3052 However, an attempt to turn on case-sensitivity in a regex key by
3053 including (?-i) didn't work because the subject string was already
3054 lowercased, and the effects were non-intuitive. It turns out that a
3055 one-line patch can be used to allow (?-i) to work as expected.
3061 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3062 other effects) broke the use of negated acl sub-conditions.
3064 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3067 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3068 "Deny" causes Exim to reject the incoming connection with a 554 error.
3069 Unfortunately, if there is a major crisis, such as a disk failure,
3070 tcp-wrappers gives "deny", whereas what one would like would be some
3071 kind of temporary error. A kludge has been added to help with this.
3072 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3073 554 error is used if errno is still zero or contains ENOENT (which occurs
3074 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3077 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3079 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3080 errors. Otherwise a message that provokes a temporary error (when other
3081 messages do not) can cause a whole host to time out.
3083 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3084 addresses were routed directly to files or pipes from a redirect router.
3085 File deliveries just didn't batch; pipe deliveries might have suffered
3088 PH/06 A failure to get a lock for a hints database would erroneously always say
3089 "Failed to get write lock", even when it was really a read lock.
3091 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3092 of 0600. This has been changed to use the value of the lockfile_mode
3093 option (which defaults to 0600).
3095 PH/08 Applied small patch from the Sieve maintainer.
3097 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3098 folder from quota calculations, a direct delivery into this folder messed
3099 up the contents of the maildirsize file. This was because the regex was
3100 used only to exclude .Trash (or whatever) when the size of the mailbox
3101 was calculated. There was no check that a delivery was happening into an
3102 excluded directory. This bug has been fixed by ignoring all quota
3103 processing for deliveries into excluded directories.
3105 PH/10 Added the maildirfolder_create_regex option to appendfile.
3111 PH/01 The code for finding all the local interface addresses on a FreeBSD
3112 system running IPv6 was broken. This may well have applied to all BSD
3113 systems, as well as to others that have similar system calls. The broken
3114 code found IPv4 interfaces correctly, but gave incorrect values for the
3115 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3116 that it would not match correctly against @[] and not recognize the IPv6
3119 PH/02 The ipliteral router was not recognizing addresses of the form user@
3120 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3122 PH/03 Added disable_ipv6.
3124 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3125 decoded form, because it is most often used to construct To: headers
3126 lines in autoreplies, and the decoded form may well be syntactically
3127 invalid. However, $reply_address has leading white space removed, and all
3128 newlines turned into spaces so that the autoreply transport does not
3131 PH/05 If group was specified without a user on a router, and no group or user
3132 was specified on a transport, the group from the router was ignored.
3134 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3135 for visible compile-time settings that can be used to change these
3136 numbers, for those that want even more. Backwards compatibility with old
3137 spool files has been maintained. However, going back to a previous Exim
3138 release will lost any variables that are in spool files.
3140 PH/07 Two small changes when running in the test harness: increase delay when
3141 passing a TCP/IP connection to a new process, in case the original
3142 process has to generate a bounce, and remove special handling of
3143 127.0.0.2 (sic), which is no longer necessary.
3145 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3146 be the same on different OS.
3148 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3151 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3152 whether --show-vars was specified or not
3154 JJ/02 exipick: Added support for new ACL variable spool format introduced
3157 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3158 syntactically invalid From: or Reply-to: line, and a filter used this to
3159 generate an autoreply, and therefore failed to obtain an address for the
3160 autoreply, Exim could try to deliver to a non-existent relative file
3161 name, causing unrelated and misleading errors. What now happens is that
3162 it logs this as a hard delivery error, but does not attempt to create a
3165 PH/11 The exinext utility has a -C option for testing purposes, but although
3166 the given file was scanned by exinext itself; it wasn't being passed on
3167 when Exim was called.
3169 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3170 an end-of-file indication when reading a command response.
3172 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3173 compiled. In many other places in Exim, IPv6 addresses are always
3174 recognized, so I have changed this. It also means that IPv4 domain
3175 literals of the form [IPV4:n.n.n.n] are now always recognized.
3177 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3178 used if the router is not running as root, for example, when verifying at
3179 ACL time, or when using -bh. The debugging output from this situation was
3180 non-existent - all you got was a failure to exec. I have made two
3183 (a) Failures to set uid/gid, the current directory, or a process leader
3184 in a subprocess such as that created by queryprogram now generate
3185 suitable debugging output when -d is set.
3187 (b) The queryprogram router detects when it is not running as root,
3188 outputs suitable debugging information if -d is set, and then runs
3189 the subprocess without attempting to change uid/gid.
3191 PH/15 Minor change to Makefile for building test_host (undocumented testing
3194 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3195 additional section of a DNS packet that returns MX or SRV records.
3196 Instead, it always explicitly searches for A/AAAA records. This avoids
3197 major problems that occur when a DNS server includes only records of one
3198 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3199 fixed another bug: if SRV records were looked up and the corresponding
3200 address records were *not* found in the additional section, the port
3201 values from the SRV records were lost.
3203 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3204 using the correct key (the original address) when searching the retry
3205 rules in order to find which one to use for generating the retry hint.
3207 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3208 adding the default one. Similarly, if it contains a Reply-To: header, the
3209 errors_reply_to option, if set, is not used.
3211 PH/19 When calculating a retry time, Exim used to measure the "time since
3212 failure" by looking at the "first failed" field in the retry record. Now
3213 it does not use this if it is later than than the arrival time of the
3214 message. Instead it uses the arrival time. This makes for better
3215 behaviour in cases where some deliveries succeed, thus re-setting the
3216 "first failed" field. An example is a quota failure for a huge message
3217 when small messages continue to be delivered. Without this change, the
3218 "time since failure" will always be short, possible causing more frequent
3219 delivery attempts for the huge message than are intended.
3220 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3222 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3223 $1, $2, $3) because the numerical variables can be reset during some
3224 expansion items (e.g. "match"), thereby losing the authentication data.
3226 PH/21 Make -bV show the size of off_t variables so that the test suite can
3227 decide whether to run tests for quotas > 2G.
3229 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3230 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3231 filecount value is greater than 2G or if a quota value is greater than 2G
3232 on a system where the size of off_t is not greater than 4, a panic error
3235 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3236 never match. The debug and -bh output now contains an explicit error
3237 message indicating a malformed IPv4 address or mask.
3239 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3240 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3241 PH/23 above applies.
3243 PH/25 Do not write to syslog when running in the test harness. The only
3244 occasion when this arises is a failure to open the main or panic logs
3245 (for which there is an explicit test).
3247 PH/26 Added the /no_tell option to "control=freeze".
3249 PH/27 If a host name lookup failed very early in a connection, for example, if
3250 the IP address matched host_lookup and the reverse lookup yielded a name
3251 that did not have a forward lookup, an error message of the form "no IP
3252 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3253 could be logged. Now it outputs the IP address instead of "NULL".
3255 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3256 allows the sender and the authenticated sender to be set when
3257 submitting a message from within Exim. Since child_open_exim() is
3258 documented for local_scan(), the new function should be too.
3260 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3261 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3262 results in an empty string is now treated as unset.
3264 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3266 PH/31 Added sender_verify_fail logging option.
3268 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3269 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3270 tidied the source and removed it altogether.
3272 PH/33 When a queue run was abandoned because the load average was too high, a
3273 log line was always written; now it is written only if the queue_run log
3274 selector is set. In addition, the log line for abandonment now contains
3275 information about the queue run such as the pid. This is always present
3276 in "start" and "stop" lines but was omitted from the "abandon" line.
3278 PH/34 Omit spaces between a header name and the colon in the error message that
3279 is given when verify = headers_syntax fails (if there are lots of them,
3280 the message gets confusing).
3282 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3283 names, as there are now some PTR records that contain slashes. This check
3284 is only to protect against broken name servers that fall over on strange
3285 characters, so the fact that it applies to all lookups doesn't matter.
3287 PH/36 Now that the new test suite is complete, we can remove some of the
3288 special code in Exim that was needed for the old test suite. For example,
3289 sorting DNS records because real resolvers return them in an arbitrary
3290 order. The new test suite's fake resolver always returns records in the
3293 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3294 bounces) except when queue_only is set, to avoid logging races between
3295 the different processes.
3297 PH/38 Panic-die if .include specifies a non-absolute path.
3299 PH/39 A tweak to the "H" retry rule from its user.
3301 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3302 a label. They prevented compilation on older perls.
3304 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3305 a warning to be raised on newish perls.
3307 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3308 on queue. Changes to match documented behaviour of showing count of
3309 messages matching specified criteria.
3311 PH/40 Changed the default ident timeout from 30s to 5s.
3313 PH/41 Added support for the use of login_cap features, on those BSD systems
3314 that have them, for controlling the resources used by pipe deliveries.
3316 PH/42 The content-scanning code uses fopen() to create files in which to put
3317 message data. Previously it was not paying any attention to the mode of
3318 the files. Exim runs with umask(0) because the rest of the code creates
3319 files with open(), and sets the required mode explicitly. Thus, these
3320 files were ending up world-writeable. This was not a big issue, because,
3321 being within the spool directory, they were not world-accessible. I have
3322 created a function called modefopen, which takes an additional mode
3323 argument. It sets umask(777), creates the file, chmods it to the required
3324 mode, then resets the umask. All the relevant calls to fopen() in the
3325 content scanning code have been changed to use this function.
3327 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3328 to 24 hours. This avoids potential overflow problems when processing G
3329 and H retry rules. I suspect nobody ever tinkers with this value.
3331 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3333 PH/45 When the plaintext authenticator is running as a client, the server's
3334 challenges are checked to ensure they are valid base64 strings. By
3335 default, the authentication attempt is cancelled if an invalid string is
3336 received. Setting client_ignore_invalid_base64 true ignores these errors.
3337 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3338 they are received. Thus, the responses can be made to depend on the
3339 challenges. If an invalid string is ignored, an empty string is placed in
3342 PH/46 Messages that are created by the autoreply transport now contains a
3343 References: header, in accordance with RFCs 2822 and 3834.
3345 PH/47 Added authenticated_sender_force to the smtp transport.
3347 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3349 PH/49 Installed latest patch from the Sieve maintainer.
3351 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3352 was also set, the appendfile transport was unnecessarily scanning a
3353 directory of message files (e.g. for maildir delivery) to find the count
3354 of files (along with the size), even though it did not need this
3355 information. It now does the scan only if it needs to find either the
3356 size of the count of files.
3358 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3360 PH/52 Two bugs concerned with error handling when the smtp transport is
3363 (i) Exim was not creating retry information for temporary errors given
3364 for individual recipients after the DATA command when the smtp transport
3365 was used in LMTP mode. This meant that they could be retried too
3366 frequently, and not timed out correctly.
3368 (ii) Exim was setting the flag that allows error details to be returned
3369 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3370 recipients that were returned after the DATA command.
3372 PH/53 This is related to PH/52, but is more general: for any failing address,
3373 when detailed error information was permitted to be returned to the
3374 sender, but the error was temporary, then after the final timeout, only
3375 "retry timeout exceeded" was returned. Now it returns the full error as
3376 well as "retry timeout exceeded".
3378 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3379 do this, and (what is worse) MTAs that accept it.
3381 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3382 will now be deprecated.
3384 PH/56 New os.c-cygwin from the Cygwin maintainer.
3386 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3387 formats (previously only available in exim formats via -bpr, -bpru,
3388 and -bpra. Now also available in native and exiqgrep formats)
3390 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3391 with very large, slow to parse queues
3393 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3395 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3397 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3398 responses to authentication challenges, though it was showing the
3399 challenges; (ii) I've removed the CR characters from the debug output for
3402 PH/58 Allow for the insertion of a newline as well as a space when a string
3403 is turned into more than one encoded-word during RFC 2047 encoding. The
3404 Sieve code now uses this.
3406 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3407 data_4xx, lost_connection, tls_required.
3409 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3410 message was being sent as an SMTP response.
3412 PH/61 Add -l and -k options to exicyclog.
3414 PH/62 When verifying, if an address was redirected to one new address, so that
3415 verification continued, and the new address failed or deferred after
3416 having set something in $address_data, the value of $address_data was not
3417 passed back to the ACL. This was different to the case when no
3418 redirection occurred. The value is now passed back in both cases.
3420 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3421 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3422 use login_cap.h, so on its own it isn't the distinguishing feature. The
3423 new name refers directly to the setclassresources() function.
3425 PH/65 Added configuration files for NetBSD3.
3427 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3429 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3430 is preferred over IPv4.
3432 PH/68 The bounce_return_message and bounce_return_body options were not being
3433 honoured for bounces generated during the reception of non-SMTP messages.
3434 In particular, this applied to messages rejected by the ACL. This bug has
3435 been fixed. However, if bounce_return_message is true and bounce_return_
3436 body is false, the headers that are returned for a non-SMTP message
3437 include only those that have been read before the error was detected.
3438 (In the case of an ACL rejection, they have all been read.)
3440 PH/69 The HTML version of the specification is now built in a directory called
3441 spec_html instead of spec.html, because the latter looks like a path with
3442 a MIME-type, and this confuses some software.
3444 PH/70 Catch two compiler warnings in sieve.c.
3446 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3447 function verify_get_ident() calls ip_connect() to connect a socket, but
3448 if the "connect()" function timed out, ip_connect() used to close the
3449 socket. However, verify_get_ident() also closes the socket later, and in
3450 between Exim writes to the log, which may get opened at this point. When
3451 the socket was closed in ip_connect(), the log could get the same file
3452 descriptor number as the socket. This naturally causes chaos. The fix is
3453 not to close the socket in ip_connect(); the socket should be closed by
3454 the function that creates it. There was only one place in the code where
3455 this was missing, in the iplookup router, which I don't think anybody now
3456 uses, but I've fixed it anyway.
3458 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3459 well as to direct DNS lookups. Otherwise the handling of names in host
3460 lists is inconsistent and therefore confusing.
3466 PH/01 Two changes to the default runtime configuration:
3468 (1) Move the checks for relay_from_hosts and authenticated clients from
3469 after to before the (commented out) DNS black list checks.
3471 (2) Add control=submission to the relay_from_hosts and authenticated
3472 clients checks, on the grounds that messages accepted by these
3473 statements are most likely to be submissions.
3475 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3477 (1) Generate an error if the third argument for the ${prvs expansion is
3480 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3483 (3) Reset the variables that are obtained from the first argument of
3484 ${prvscheck and used in the second argument before leaving the code,
3485 because their memory is reclaimed, so using them afterwards may do
3488 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3489 one (it's much easier than Tom thought :-).
3491 (5) Because of (4), we can now allow for the use of $prvscheck_result
3492 inside the third argument.
3494 PH/03 For some reason, the default setting of PATH when running a command from
3495 a pipe transport was just "/usr/bin". I have changed it to
3498 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3499 anything to be listed in the output from -bV.
3501 PH/05 When a filter generated an autoreply, the entire To: header line was
3502 quoted in the delivery log line, like this:
3504 => >A.N.Other <ano@some.domain> <original@ddress> ...
3506 This has been changed so that it extracts the operative address. There
3507 may be more than one such address. If so, they are comma-separated, like
3510 => >ano@some.domain,ona@other.domain <original@ddress> ...
3512 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3513 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3514 not being looked up in the rDNS to get a host name, Exim was showing the
3515 IP address twice in Received: lines, even though the IP addresses were
3516 identical. For example:
3518 Received: from [1.2.3.4] (helo=[1.2.3.4])
3520 However, if the real host name was known, it was omitting the HELO data
3521 if it matched the actual IP address. This has been tidied up so that it
3522 doesn't show the same IP address twice.
3524 PH/07 When both +timestamp and +memory debugging was on, the value given by
3525 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3526 called by the debug printing, thereby overwriting the timestamp buffer.
3527 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3529 PH/08 When the original message was included in an autoreply transport, it
3530 always said "this is a copy of the message, including all the headers",
3531 even if body_only or headers_only was set. It now gives an appropriate
3534 PH/09 Applied a patch from the Sieve maintainer which:
3536 o fixes some comments
3537 o adds the (disabled) notify extension core
3538 o adds some debug output for the result of if/elsif tests
3539 o points to the current vacation draft in the documentation
3540 and documents the missing references header update
3544 o fixes a bug in processing the envelope test (when testing
3545 multiple envelope elements, the last element determined the
3548 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3549 Electronic Mail") by including:
3551 Auto-submitted: auto-generated
3553 in the messages that it generates (bounce messages and others, such as
3554 warnings). In the case of bounce messages for non-SMTP messages, there was
3555 also a typo: it was using "Auto_submitted" (underscore instead of
3556 hyphen). Since every message generated by Exim is necessarily in response
3557 to another message, thes have all been changed to:
3559 Auto-Submitted: auto-replied
3561 in accordance with these statements in the RFC:
3563 The auto-replied keyword:
3565 - SHOULD be used on messages sent in direct response to another
3566 message by an automatic process,
3568 - MUST NOT be used on manually-generated messages,
3570 - MAY be used on Delivery Status Notifications (DSNs) and Message
3571 Disposition Notifications (MDNs),
3573 - MUST NOT be used on messages generated by automatic or periodic
3574 processes, except for messages which are automatic responses to
3577 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3578 to the default Received: header definition.
3580 PH/12 Added log selector acl_warn_skipped (default on).
3582 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3583 variables because (a) they are in the wrong storage pool and (b) even if
3584 they were copied, it wouldn't work properly because of the caching.
3586 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3587 checking when decoding. Apparently there are clients that generate
3588 overlong encoded strings. Why am I not surprised?
3590 PH/15 If the first argument of "${if match_address" was not empty, but did not
3591 contain an "@" character, Exim crashed. Now it writes a panic log message
3592 and treats the condition as false.
3594 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3596 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3597 extension "envelope-auth". The code is finished and in agreement with
3598 other implementations, but there is no documentation so far and in fact,
3599 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3600 not changing the active code.
3602 Print executed "if" and "elsif" statements when debugging is used. This
3603 helps a great deal to understand what a filter does.
3605 Document more things not specified clearly in RFC3028. I had all this
3606 sorted out, when out of a sudden new issues came to my mind. Oops."
3608 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3611 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3612 canonical form (as documented). However, after a host name lookup from
3613 the IP address, check_host() was doing a simple string comparison with
3614 addresses acquired from the DNS when checking that the found name did
3615 have the original IP as one of its addresses. Since any found IPv6
3616 addresses are likely to be in abbreviated form, the comparison could
3617 fail. Luckily, there already exists a function for doing the comparison
3618 by converting both addresses to binary, so now that is used instead of
3619 the text comparison.
3621 PH/20 There was another similar case to PH/19, when a complete host name was
3622 given in a host list; looking up its IP address could give an abbreviated
3623 form, whereas the current host's name might or might not be abbreviated.
3624 The same fix has been applied.
3630 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3631 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3634 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3635 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3637 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3639 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3640 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3641 Also, for "Auto-Submitted", treat anything other than "no" as
3642 non-personal, in accordance with RFC 3834. (Previously it treated
3643 anything starting "auto-" as non-personal.)
3645 TF/01 The control=submission/name=... option had a problem with syntax
3646 errors if the name included a slash character. The /name= option
3647 now slurps the rest of the string, so it can include any characters
3648 but it must come last in the list of options (after /sender_retain
3651 PH/05 Some modifications to the interface to the fake nameserver for the new
3659 TK/01 Added the "success_on_redirect" address verification option. See
3660 NewStuff for rationale and an example.
3662 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3664 PH/02 Patch to exigrep to allow it to work on syslog lines.
3666 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3667 fread() to skip over the body file's header line, because in Cygwin the
3668 header line is locked and is inaccessible.
3670 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3671 co-exist for some time) to make it clear that it is the Exim ID that is
3672 referenced, not the Message-ID: header line.
3674 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3675 string_format() function, because snprintf() does not exist on all
3678 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3679 ${stat: expansion item.
3681 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3682 protocol synchronization error", to keep the pedants happy.
3684 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3685 well as for IRIX systems, when gcc is being used. See the host.c source
3688 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3690 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3693 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3694 transport and to the smtp transport in LMTP mode.
3696 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3698 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3699 run a filter in a subprocess. This could lead to confusion in subsequent
3700 lookups in the parent process. There should also be a search_tidyup() at
3701 the end of the subprocess.
3703 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3704 only if the host matched helo_try_verify_hosts, which caused the
3705 verification to occur when the EHLO/HELO command was issued. The ACL just
3706 tested the remembered result. Now, if a previous verification attempt has
3707 not happened, "verify = helo" does it there and then.
3709 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3711 TK/03 Fix log output including CR from clamd.
3713 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3714 exist provoked a memory error which could cause a segfault.
3716 PH/15 Installed PCRE 6.2
3718 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3720 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3721 of the problem. Specifically, suggested +O2 rather than +O1 for the
3724 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3726 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3727 record, the port was not correctly logged when the outgoing_port log
3728 selector was set (it logged the transort's default port).
3730 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3731 fallback_hosts, and "hosts" in the smtp transport.
3733 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3734 host errors such as "Connection refused".
3736 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3737 authentication with radiusclient 0.4.9:
3739 - Error returned from rc_read_config was caught wrongly
3740 - Username/password not passed on to radius server due to wrong length.
3742 The presumption is that some radiusclient API changes for 4.51/PH/17
3743 were not taken care of correctly. The code is still untested by me (my
3744 Linux distribution still has 0.3.2 of radiusclient), but it was
3745 contributed by a Radius user.
3747 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3748 expanding the "port" option of the smtp transport.
3750 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3751 while reading a MIME header. Thanks to Tom Hughes for a patch.
3753 PH/24 Include config.h inside local_scan.h so that configuration settings are
3756 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3757 that in an ACL for RCPT (for example), you can examine exactly what was
3760 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3761 commands, but it was not correctly comparing the address with the actual
3762 client host address. Thus, it would show the EHLO address in Received:
3763 header lines when this was not necessary.
3765 PH/27 Added the % operator to ${eval:}.
3767 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3768 it should be ignoring failures (because with -C, for example, it has lost
3769 privilege). It wasn't ignoring creation failures other than "already
3772 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3775 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3776 because an input line was too long, either on its own, or by virtue of
3777 too many continuations, the temporary file was not being removed, and the
3778 return code was incorrect.
3780 PH/31 Missing "BOOL" in function definition in filtertest.c.
3782 PH/32 Applied Sieve patches from the maintainer.
3784 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3786 PH/33 Added "verify = not_blind".
3788 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3789 Local/Makefile (with some defaults set). These are used in built scripts
3790 such as exicyclog, but they have never been used in the exim_install
3791 script (though there are many overriding facilities there). I have
3792 arranged that the exim_install script now takes note of these two
3795 PH/35 Installed configuration files for Dragonfly.
3797 PH/36 When a locally submitted message by a trusted user did not contain a
3798 From: header, and the sender address was obtained from -f or from an SMTP
3799 MAIL command, and the trusted user did not use -F to supply a sender
3800 name, $originator_name was incorrectly used when constructing a From:
3801 header. Furthermore, $originator_name was used for submission mode
3802 messages from external hosts without From: headers in a similar way,
3803 which is clearly wrong.
3805 PH/37 Added control=suppress_local_fixups.
3807 PH/38 When log_selector = +received_sender was set, and the addition of the
3808 sender made the log line's construction buffer exactly full, or one byte
3809 less than full, an overflow happened when the terminating "\n" was
3812 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3813 when the result of a list match is failure because a DNS lookup failed.
3815 PH/40 RM_COMMAND is now used in the building process.
3817 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3818 the "build-* directories that it finds.
3820 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3821 address in a domain literal was a prefix of an interface address.
3823 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3824 when verifying a sender address, unless rewrite_headers is false.
3826 PH/44 Wrote a long comment about why errors_to addresses are verified as
3827 recipients, not senders.
3829 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3830 the ratelimit ACL was added.
3832 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3834 PH/46 Added extra information about PostgreSQL errors to the error string.
3836 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3837 avoiding the need to install special zones in a real server. This is
3838 backwards compatible; if it can't find the fake resolver, it drops back.
3839 Thus, both old and new test suites can be run.
3841 TF/02 Added util/ratelimit.pl
3843 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3844 clock is set back in time.
3846 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3847 Candler <B.Candler@pobox.com>.
3849 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3850 OK for addresses that are the result of redirections.
3852 PH/48 A number of further additions for the benefit of the new test suite,
3853 including a fake gethostbyname() that interfaces to the fake DNS resolver
3856 TF/06 The fix for widen_domains has also been applied to qualify_single and
3857 search_parents which are the other dnslookup options that can cause
3860 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3863 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3865 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3866 service" if it encountered an MX record with an empty target hostname.
3867 The message is now "an MX or SRV record indicated no SMTP service".
3869 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3870 if the DNS of the sending site is misconfigured. This is quite a
3871 common situation. This change restores the behaviour of treating a
3872 helo verification defer as a failure.
3874 PH/51 If self=fail was set on a router, the bounce message did not include the
3875 actual error message.
3881 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3883 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3884 command itself ended in error, the underlying message about the transport
3885 filter timeout was being overwritten with the pipe command error. Now the
3886 underlying error message should be appended to the second error message.
3888 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3890 PH/02 Reduce the amount of output that "make" produces by default. Full output
3891 can still be requested.
3893 PH/03 The warning log line about a condition test deferring for a "warn" verb
3894 was being output only once per connection, rather than after each
3895 occurrence (because it was using the same function as for successful
3896 "warn" verbs). This seems wrong, so I have changed it.
3898 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3899 it should not have, which might have caused a crash in the right
3900 circumstances, but probably never did.
3902 PH/04 Installed a modified version of Tony Finch's patch to make submission
3903 mode fix the return path as well as the Sender: header line, and to
3904 add a /name= option so that you can make the user's friendly name appear
3907 TF/03 Added the control = fakedefer ACL modifier.
3909 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3910 Mark Lowes for thorough testing.
3912 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3914 TK/03 Merged latest SRS patch from Miles Wilton.
3916 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3917 with the definition in sysexits.h (which is #included earlier).
3918 Fortunately, Exim does not actually use EX_OK. The code used to try to
3919 preserve the sysexits.h value, by assuming that macro definitions were
3920 scanned for macro replacements. I have been disabused of this notion,
3921 so now the code just undefines EX_OK before #including unistd.h.
3923 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3924 in the smtp transport. When a block could not be written in a single
3925 write() function, the timeout was being re-applied to each part-write.
3926 This seems wrong - if the receiver was accepting one byte at a time it
3927 would take for ever. The timeout is now adjusted when this happens. It
3928 doesn't have to be particularly precise.
3930 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3931 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3933 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3934 without a domain if the check to <postmaster@domain> fails.
3936 SC/01 Eximstats: added -xls and the ability to specify output files
3937 (patch written by Frank Heydlauf).
3939 SC/02 Eximstats: use FileHandles for outputting results.
3941 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3943 SC/04 Eximstats: fixed display of large numbers with -nvr option
3945 SC/05 Eximstats: fixed merging of reports with empty tables.
3947 SC/06 Eximstats: added the -include_original_destination flag
3949 SC/07 Eximstats: removed tabs and trailing whitespace.
3951 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3953 TK/06 MBOX spool code: Add real "From " MBOX separator line
3954 so the .eml file is really in mbox format (even though
3955 most programs do not really care). Patch from Alex Miller.
3957 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3958 The latter is generated from $received_to and is only set if the
3959 message has one envelope recipient. SA can use these headers,
3960 obviously out-of-the-box. Patch from Alex Miller.
3962 PH/08 The ${def test on a variable was returning false if the variable's
3963 value was "0", contrary to what the specification has always said!
3964 The result should be true unless the variable is empty.
3966 PH/09 The syntax error of a character other than { following "${if
3967 def:variable_name" (after optional whitespace) was not being diagnosed.
3968 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3969 accidental colon was present, for example, could give incorrect results.
3971 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3972 result is used (not including appendfile, where other changes are about
3975 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3976 This involved changing a lot of size variables from int to off_t. It
3977 should work with maildirs and everything.
3979 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3980 spamd dying while we are connected to it.
3982 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3985 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3986 function for BDB 4.3.
3988 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3990 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3991 the sources for PCRE have been re-arranged and more files are now
3994 PH/15 The code I had for printing potentially long long variables in PH/11
3995 above was not the best (it lost precision). The length of off_t variables
3996 is now inspected at build time, and an appropriate printing format (%ld
3997 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3998 to be "long long int" or "long int". This is needed for the internal
3999 formatting function string_vformat().
4001 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4002 the configuration file to be ":syslog", then the script "guesses" where
4003 the logs files are, rather than using the compiled in default. In our
4004 case the guess is not the same as the compiled default, so the script
4005 suddenly stopped working when I started to use syslog. The patch checks
4006 to see if log_file_path is "". If so, it attempts to read it from exim
4007 with no configuration file to get the compiled in version, before it
4008 falls back to the previous guessing code."
4010 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4011 implementing BATV in an Exim configuration. See NewStuff for the gory
4014 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4015 Makefile that are specific to HP-UX.
4017 PH/18 If the "use_postmaster" option was set for a recipient callout together
4018 with the "random" option, the postmaster address was used as the MAIL
4019 FROM address for the random test, but not for the subsequent recipient
4020 test. It is now used for both.
4022 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4023 patch removes a few documentation additions to RFC 3028, because the
4024 latest draft now contains them. It adds the new en;ascii-case comparator
4025 and a new error check for 8bit text in MIME parts. Comparator and
4026 require names are now matched exactly. I enabled the subaddress
4027 extension, but it is not well tested yet (read: it works for me)."
4029 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4030 rework some of the code of TK/09 above to avoid the hardwired use of
4031 "%lld" and "long long". Replaced the call to snprintf() with a call to
4034 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4035 records point to non-existent hosts", "retry timeout exceeded", and
4036 "retry time not reached for any host after a long failure period".
4038 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4039 experimental DomainKeys support:
4041 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4042 (2) On an error such as an illegally used "control", the wrong name for
4043 the control was given.
4045 These problems did NOT occur unless DomainKeys support was compiled.
4047 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4049 PH/24 Added ${if match_ip condition.
4051 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4052 databases so that it will be absolutely obvious if a crash occurs in the
4053 DB library. This is a regular occurrence (often caused by mis-matched
4056 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4057 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4058 picked up on a user's system that detects such things. There doesn't seem
4059 to be a gcc warning option for this - only an attribute that has to be
4060 put on the function's prototype. It seems that in Fedora Core 4 they have
4061 set this on a number of new functions. No doubt there will be more in due
4064 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4065 specify a transport. However, if an address that was verified by such a
4066 router was the subject of a callout, Exim crashed because it tried to
4067 read the rcpt_include_affixes from the non-existent transport. Now it
4068 just assumes that the setting of that option is false. This bug was
4069 introduced by 4.51/PH/31.
4071 PH/28 Changed -d+all to exclude +memory, because that information is very
4072 rarely of interest, but it makes the output a lot bigger. People tend to
4073 do -d+all out of habit.
4075 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4076 code in os-type was giving problems when libc.so lives in lib64, like on
4079 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4080 aren't the modern standard, and it seems that some systems' include files
4081 don't always have them. Exim was already checking for some of the newer
4082 ones like T_AAAA, and defining it itself. I've added checks for all the
4083 record types that Exim uses.
4085 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4086 not automatically generating a new one, as it is supposed to. This
4087 prevented TLS from working. If the file did exist, but contained invalid
4088 data, a new version was generated, as expected. It was only the case of a
4089 non-existent file that was broken.
4091 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4092 with a change in libdomainkeys > 0.64.
4094 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4095 from DNS. If the selector record carries the flag, it now has
4096 precedence over the domain-wide flag.
4098 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4100 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4101 the use of a port name that isn't defined in /etc/services) occurred, the
4102 message was deferred as in a normal delivery, and thus remained on the
4103 spool, instead of being failed because of the mua_wrapper setting. This
4104 is now fixed, and I tidied up some of the mua_wrapper messages at the
4107 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4108 the messages in a hash of arrays rather than using individual hashes.
4109 This is a bit cleaner and results in dramatic memory savings, albeit
4110 at a slight CPU cost.
4112 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4113 as requested by Marc Sherman.
4115 SC/10 Eximstats: added histograms for user specified patterns as requested
4118 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4120 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4121 fopen() in the content-scanning modules that did not already have it.
4127 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4128 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4130 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4132 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4134 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4135 to test Sieve filters that use "vacation".
4137 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4138 that changes the way the GnuTLS parameters are stored in the cache file.
4139 The new format can be generated externally. For backward compatibility,
4140 if the data in the cache doesn't make sense, Exim assumes it has read an
4141 old-format file, and it generates new data and writes a new file. This
4142 means that you can't go back to an older release without removing the
4145 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4146 work if there are any delivery delays because "one_time" forces the
4147 parent to be marked "delivered", so its unseen clone is never tried
4148 again. For this reason, Exim now forbids the simultaneous setting of
4151 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4152 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4153 entry if you want to know the details. The fix, however, neglected to
4154 consider the case when local delivery batching is involved. The test for
4155 "previously delivered" was not happening when checking to see if an
4156 address could be batched with a previous (undelivered) one; under
4157 certain circumstances this could lead to multiple deliveries to the same
4160 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4161 in its include files, and this causes problems building Exim.
4163 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4164 header_syntax) but Exim was just ignoring anything given after a slash.
4165 In particular, this caused confusion with an attempt to use "verify =
4166 reverse_host_lookup/defer_ok". An error is now given when options are
4167 supplied for verify items that do not have them. (Maybe reverse_host_
4168 lookup should have a defer_ok option, but that's a different point.)
4170 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4171 defined by RFC 821) to 2048, because there were problems with some AUTH
4172 commands, and RFC 1869 says the size should be increased for extended
4173 SMTP commands that take arguments.
4175 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4178 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4179 "unknown" error; now it says that the functionality isn't in the binary.
4181 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4182 an address' error message when a string expansion fails (syntax or
4183 whatever). Otherwise the password may appear in the log. Following change
4184 PH/42 below, there is no longer a chance of it appearing in a bounce
4187 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4189 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4190 of its addresses was ever tried. (Bugzilla bug #2.)
4192 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4193 the result incorrectly in the debug output. (It correctly added a newline
4194 to what was transported.)
4196 TF/01 Added $received_time.
4198 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4199 commented out examples of how to interface to a virus scanner and to
4200 SpamAssassin. Also added commented examples of av_scanner and
4201 spamd_address settings.
4203 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4204 and controls are allowed in which ACLs. There were a couple of minor
4205 errors. Some of the entries in the conditions table (which is a table of
4206 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4207 negation of where the condition IS allowed.
4209 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4211 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4212 header file does not have a version number, so I've had to invent a new
4213 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4214 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4215 radiusclient), but it was contributed by a Radius user.
4217 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4218 files or directories, for OpenSSL.
4220 PH/19 When an Exim process that is running as root has to create an Exim log
4221 file, it does so in a subprocess that runs as exim:exim so as to get the
4222 ownership right at creation (otherwise, other Exim processes might see
4223 the file with the wrong ownership). There was no test for failure of this
4224 fork() call, which would lead to the process getting stuck as it waited
4225 for a non-existent subprocess. Forks do occasionally fail when resources
4226 run out. I reviewed all the other calls to fork(); they all seem to check
4229 PH/20 When checking for unexpected SMTP input at connect time (before writing
4230 the banner), Exim was not dealing correctly with a non-positive return
4231 from the read() function. If the client had disconnected by this time,
4232 the result was a log entry for a synchronization error with an empty
4233 string after "input=" when read() returned zero. If read() returned -1
4234 (an event I could not check), uninitialized data bytes were printed.
4235 There were reports of junk text (parts of files, etc) appearing after
4238 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4240 PH/22 Added support for macro redefinition, and (re)definition in between
4241 driver and ACL definitions.
4243 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4244 forgetting to use the resulting value; it was using the unexpanded value.
4246 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4247 hadn't been configured. The fix is from Juergen Kreileder, who
4248 understands it better than I do:
4250 "Here's what I see happening with three configured cyrus_sasl
4251 authenticators configured (plain, login, cram-md5):
4253 On startup auth_cyrus_sasl_init() gets called for each of these.
4254 This means three calls to sasl_listmech() without a specified mech_list.
4255 => SASL tests which mechs of all available mechs actually work
4256 => three warnings about OTP not working
4257 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4259 With the patch, sasl_listmech() also gets called three times. But now
4260 SASL's mech_list option is set to the server_mech specified in the the
4261 authenticator. Or in other words, the answer from sasl_listmech()
4262 gets limited to just the mech you're testing for (which is different
4264 => the return list contains just 'plain' or 'login', 'cram-md5' or
4265 nothing depending on the value of ob->server_mech.
4267 I've just tested the patch: Authentication still works fine,
4268 unavailable mechs specified in the exim configuration are still
4269 caught, and the auth.log warnings about OTP are gone."
4271 PH/25 When debugging is enabled, the contents of the command line are added
4272 to the debugging output, even when log_selector=+arguments is not
4275 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4276 answer is "GNU", and only if the return is "GNU/something" is the answer
4279 PH/27 $acl_verify_message is now set immediately after the failure of a
4280 verification in an ACL, and so is available in subsequent modifiers. In
4281 particular, the message can be preserved by coding like this:
4283 warn !verify = sender
4284 set acl_m0 = $acl_verify_message
4286 Previously, $acl_verify_message was set only while expanding "message"
4287 and "log_message" when a very denied access.
4289 PH/28 Modified OS/os.c-Linux with
4291 -#ifndef OS_LOAD_AVERAGE
4292 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4294 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4295 nomenclature these days.)
4297 PH/29 Installed patch from the Sieve maintainer that adds the options
4298 sieve_useraddress and sieve_subaddress to the redirect router.
4300 PH/30 In these circumstances:
4301 . Two addresses routed to the same list of hosts;
4302 . First host does not offer TLS;
4303 . First host accepts first address;
4304 . First host gives temporary error to second address;
4305 . Second host offers TLS and a TLS session is established;
4306 . Second host accepts second address.
4307 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4308 and peerdn, if requested) that were in fact used only for the second
4311 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4312 attention to any local part prefix or suffix that was matched by the
4313 router that accepted the address. It now behaves in the same way as it
4314 does for delivery: the affixes are removed from the local part unless
4315 rcpt_include_affixes is set on the transport.
4317 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4318 timeout during the DATA phase of an incoming message.
4320 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4321 applied a patch sanctioned by the Sieve maintainer.
4323 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4324 the uid or gid is negative. A case of a negative gid caused this to be
4325 noticed. The fix allows for either to be negative.
4327 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4328 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4331 PH/36 The change PH/12 above was broken. Fixed it.
4333 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4334 the grounds that routing the same address twice would always produce the
4335 same answer. This might have been true once, but it is certainly no
4336 longer true now. Routing a child address may depend on the previous
4337 routing that produced that child. Some complicated redirection strategies
4338 went wrong when messages had multiple recipients, and made Exim's
4339 behaviour dependent on the order in which the addresses were given.
4341 I have moved the duplicate checking until after the routing is complete.
4342 Exim scans the addresses that are assigned to local and remote
4343 transports, and removes any duplicates. This means that more work will be
4344 done, as duplicates will always all be routed, but duplicates are
4345 presumably rare, so I don't expect this is of any significance.
4347 For deliveries to pipes, files, and autoreplies, the duplicate checking
4348 still happens during the routing process, since they are not going to be
4351 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4352 It corrects a timeout issue with spamd. This is Ian's comment: "The
4353 background is that sometimes spamd either never reads data from a
4354 connection it has accepted, or it never writes response data. The exiscan
4355 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4356 blindly assumes that writes won't block so it may never time out."
4358 PH/39 Allow G after quota size as well as K and M.
4360 PH/40 The value set for $authenticated_id in an authenticator may not contain
4361 binary zeroes or newlines because the value is written to log lines and
4362 to spool files. There was no check on this. Now the value is run through
4363 the string_printing() function so that such characters are converted to
4364 printable escape sequences.
4366 PH/41 $message_linecount is a new variable that contains the total number of
4367 lines in the message. Compare $body_linecount, which is the count for the
4370 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4371 bounce and delay warning messages, except in certain special cases, which
4374 (a) An SMTP error message from a remote host;
4375 (b) A message specified in a :fail: redirection;
4376 (c) A message specified in a "fail" command in a system filter;
4377 (d) A message specified in a FAIL return from the queryprogram router;
4378 (e) A message specified by the cannot_route_message router option.
4380 In these cases only, Exim does include the error details in bounce and
4381 warning messages. There are also a few cases where bland messages such
4382 as "unrouteable address" or "local delivery error" are given.
4384 PH/43 $value is now also set for the "else" part of a ${run expansion.
4386 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4387 being worked on, but at least Exim now implements the latest version to
4390 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4391 process to complete was treated as a delivery failure, a timeout while
4392 writing the message to the pipe was logged, but erroneously treated as a
4393 successful delivery. Such timeouts include transport filter timeouts. For
4394 consistency with the overall process timeout, these timeouts are now
4395 treated as errors, giving rise to delivery failures by default. However,
4396 there is now a new Boolean option for the pipe transport called
4397 timeout_defer, which, if set TRUE, converts the failures into defers for
4398 both kinds of timeout. A transport filter timeout is now identified in
4401 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4402 systems where "make" and "gmake" are different, calling "gmake" at top
4403 level broke things. I've arranged for the value of $(MAKE) to be passed
4404 from the Makefile to this script so that it can call the same version of
4408 A note about Exim versions 4.44 and 4.50
4409 ----------------------------------------
4411 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4412 changes of various kinds. As a consequence, a big documentation update was
4413 needed. This delayed the release for rather longer than seemed good, especially
4414 in the light of a couple of (minor) security issues. Therefore, the changes
4415 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4416 release. So 4.44 and 4.50 are in effect two different branches that both start
4419 I have left the 4.50 change log unchanged; it contains all the changes since
4420 4.43. The change log for 4.44 is below; many of its items are identical to
4421 those for 4.50. This seems to be the most sensible way to preserve the
4422 historical information.
4428 1. Minor wording change to the doc/README.SIEVE file.
4430 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4431 computation of the current number of files was incorrect.
4433 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4434 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4437 4. Give more explanation in the error message when the command for a transport
4438 filter fails to execute.
4440 5. There are several places where Exim runs a non-Exim command in a
4441 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4442 was being done only for the command run by the queryprogram router. It is
4443 now done for all such subprocesses. The other cases are: ${run, transport
4444 filters, and the commands run by the lmtp and pipe transports.
4446 6. Added CONFIGURE_GROUP build-time option.
4448 7. Some older OS have a limit of 256 on the maximum number of file
4449 descriptors. Exim was using setrlimit() to set 1000 as a large value
4450 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4451 systems. I've change it so that if it can't get 1000, it tries for 256.
4453 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4454 was an oversight, and furthermore, ever since the addition of extra
4455 controls (e.g. 4.43/32), the checks on when to allow different forms of
4456 "control" were broken. There should now be diagnostics for all cases when a
4457 control that does not make sense is encountered.
4459 9. Added the /retain_sender option to "control=submission".
4461 10. $recipients is now available in the predata ACL (oversight).
4463 11. Tidy the search cache before the fork to do a delivery from a message
4464 received from the command line. Otherwise the child will trigger a lookup
4465 failure and thereby defer the delivery if it tries to use (for example) a
4466 cached ldap connection that the parent has called unbind on.
4468 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4469 of $address_data from the recipient verification was clobbered by the
4470 sender verification.
4472 13. The value of address_data from a sender verification is now available in
4473 $sender_address_data in subsequent conditions in the ACL statement.
4475 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4477 15. Added a new option "connect=<time>" to callout options, to set a different
4480 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4481 was its contents. (It was OK if the option was not defined at all.)
4483 17. A "Completed" log line is now written for messages that are removed from
4484 the spool by the -Mrm option.
4486 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4487 information about exactly what failed.
4489 19. Added -dd to debug only the daemon process.
4491 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4492 handles timeouts, both on the server side and network timeouts. Renamed the
4493 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4495 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4496 It is now set to "smtps".
4498 22. $host_address is now set to the target address during the checking of
4499 ignore_target_hosts.
4501 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4502 being passed; this would have caused $sender_host_name to have been used if
4503 matching the list had actually called for a host name (not very likely,
4504 since this list is usually IP addresses). A host name is now passed as
4507 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4508 code that specifies a non-restarting handler (typically sigaction() in
4509 modern systems) in an attempt to fix a rare and obscure crash bug.
4511 25. Narrowed the window for a race in the daemon that could cause it to ignore
4512 SIGCHLD signals. This is not a major problem, because they are used only to
4513 wake it up if nothing else does.
4515 26. A malformed maildirsize file could cause Exim to calculate negative values
4516 for the mailbox size or file count. Odd effects could occur as a result.
4517 The maildirsize information is now recalculated if the size or filecount
4520 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4521 support for a long time. Removed HAVE_SYS_VFS_H.
4523 28. Installed the latest version of exipick from John Jetmore.
4525 29. In an address list, if the pattern was not a regular expression, an empty
4526 subject address (from a bounce message) matched only if the pattern was an
4527 empty string. Non-empty patterns were not even tested. This was the wrong
4528 because it is perfectly reasonable to use an empty address as part of a
4529 database query. An empty address is now tested by patterns that are
4530 lookups. However, all the other forms of pattern expect the subject to
4531 contain a local part and a domain, and therefore, for them, an empty
4532 address still always fails if the pattern is not itself empty.
4534 30. Exim went into a mad DNS loop when attempting to do a callout where the
4535 host was specified on an smtp transport, and looking it up yielded more
4536 than one IP address.
4538 31. Re-factored the code for checking spool and log partition space into a
4539 function that finds that data and another that does the check. The former
4540 is then used to implement four new variables: $spool_space, $log_space,
4541 $spool_inodes, and $log_inodes.
4543 32. The RFC2047 encoding function was originally intended for short strings
4544 such as real names; it was not keeping to the 75-character limit for
4545 encoded words that the RFC imposes. It now respects the limit, and
4546 generates multiple encoded words if necessary. To be on the safe side, I
4547 have increased the buffer size for the ${rfc2047: expansion operator from
4550 33. It is now permitted to omit both strings after an "if" condition; if the
4551 condition is true, the result is "true". As before, when the second string
4552 is omitted, a false condition yields an empty string. This makes it less
4553 cumbersome to write custom ACL and router conditions.
4555 34. Failure to deliver a bounce message always caused it to be frozen, even if
4556 there was an errors_to setting on the router. The errors_to setting is now
4559 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4560 canonical form (fully expanded) before being placed in
4561 $sender_host_address.
4563 36. The table in the code that translates DNS record types into text (T_A to
4564 "A" for instance) was missing entries for NS and CNAME. It is just possible
4565 that this could have caused confusion if both these types were looked up
4566 for the same domain, because the text type is used as part of Exim's
4567 per-process caching. But the chance of anyone hitting this buglet seems
4570 37. The dnsdb lookup has been extended in a number of ways.
4572 (1) There is a new type, "zns", which walks up the domain tree until it
4573 finds some nameserver records. It should be used with care.
4575 (2) There is a new type, "mxh", which is like "mx" except that it returns
4576 just the host names, not the priorities.
4578 (3) It is now possible to give a list of domains (or IP addresses) to be
4579 looked up. The behaviour when one of the lookups defers can be
4580 controlled by a keyword.
4582 (4) It is now possible to specify the separator character for use when
4583 multiple records are returned.
4585 38. The dnslists ACL condition has been extended: it is now possible to supply
4586 a list of IP addresses and/or domains to be looked up in a particular DNS
4589 39. Added log_selector=+queue_time_overall.
4591 40. When running the queue in the test harness, wait just a tad after forking a
4592 delivery process, to get repeatability of debugging output.
4594 41. Include certificate and key file names in error message when GnuTLS fails
4595 to set them up, because the GnuTLS error message doesn't include the name
4596 of the failing file when there is a problem reading it.
4598 42. Allow both -bf and -bF in the same test run.
4600 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4602 44. The "Exiscan patch" is now merged into the mainline Exim source.
4604 45. Sometimes the final signoff response after QUIT could fail to get
4605 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4606 < 0 before doing a fflush(). This bug looks as though it goes back to the
4607 introduction of TLS in release 3.20, but "sometimes" must have been rare
4608 because the tests only now provoked it.
4610 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4611 (this can affect the format of dates).
4613 47. exim_tidydb, when checking for the continued existence of a message for
4614 which it has found a message-specific retry record, was not finding
4615 messages that were in split spool directories. Consequently, it was
4616 deleting retry records that should have stayed in existence.
4618 48. Steve fixed some bugs in eximstats.
4620 49. The SPA authentication driver was not abandoning authentication and moving
4621 on to the next authenticator when an expansion was forced to fail,
4622 contradicting the general specification for all authenticators. Instead it
4623 was generating a temporary error. It now behaves as specified.
4625 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4626 (the order specifies the preference for clients). The order is now AES256,
4627 AES128, 3DES, ARCFOUR128.
4629 51. Small patch to Sieve code - explicitly set From: when generating an
4632 52. Exim crashed if a remote delivery caused a very long error message to be
4633 recorded - for instance if somebody sent an entire SpamAssassin report back
4634 as a large number of 550 error lines. This bug was coincidentally fixed by
4635 increasing the size of one of Exim's internal buffers (big_buffer) that
4636 happened as part of the Exiscan merge. However, to be on the safe side, I
4637 have made the code more robust (and fixed the comments that describe what
4640 53. Now that there can be additional text after "Completed" in log lines (if
4641 the queue_time_overall log selector is set), a one-byte patch to exigrep
4642 was needed to allow it to recognize "Completed" as not the last thing in
4645 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4646 patch that reportedly fixes this has been added. I am not expert enough to
4647 create a test for it. This is what the patch creator wrote:
4649 "I found a little strange behaviour of ldap code when working with
4650 Windows 2003 AD Domain, where users was placed in more than one
4651 Organization Units. When I tried to give exim partial DN, the exit code
4652 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4653 But simultaneously result of request was absolutely normal ldap result,
4654 so I produce this patch..."
4656 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4657 REFERENCE, so I have modified the code to exclude the patch when that macro
4660 55. Some experimental protocols are using DNS PTR records for new purposes. The
4661 keys for these records are domain names, not reversed IP addresses. The
4662 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4663 leaves it alone. Component reversal etc. now happens only for IP addresses.
4666 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4668 57. Double the size of the debug message buffer (to 2048) so that more of very
4669 long debug lines gets shown.
4671 58. The exicyclog utility now does better if the number of log files to keep
4672 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4674 59. Two changes related to the smtp_active_hostname option:
4676 (1) $smtp_active_hostname is now available as a variable.
4677 (2) The default for smtp_banner uses $smtp_active_hostname instead
4678 of $primary_hostname.
4680 60. The host_aton() function is supposed to be passed a string that is known
4681 to be a valid IP address. However, in the case of IPv6 addresses, it was
4682 not checking this. This is a hostage to fortune. Exim now panics and dies
4683 if the condition is not met. A case was found where this could be provoked
4684 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4685 components; fortuitously, this particular loophole had already been fixed
4686 by change 4.50/55 above.
4688 If there are any other similar loopholes, the new check in host_aton()
4689 itself should stop them being exploited. The report I received stated that
4690 data on the command line could provoke the exploit when Exim was running as
4691 exim, but did not say which command line option was involved. All I could
4692 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4693 running as the user.
4696 61. There was a buffer overflow vulnerability in the SPA authentication code
4697 (which came originally from the Samba project). I have added a test to the
4698 spa_base64_to_bits() function which I hope fixes it.
4701 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4702 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4704 63. The daemon start-up calls getloadavg() while still root for those OS that
4705 need the first call to be done as root, but it missed one case: when
4706 deliver_queue_load_max is set with deliver_drop_privilege. This is
4707 necessary for the benefit of the queue runner, because there is no re-exec
4708 when deliver_drop_privilege is set.
4710 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4711 This has been fixed.
4713 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4714 was in use, was not putting the data itself into the right store pool;
4715 consequently, it could be overwritten for a subsequent message in the same
4716 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4719 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4721 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4722 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4723 the function were treating the return as a boolean value, which happened to
4724 work because 0=false and not-0=true, but is not correct code.
4726 68. The host_aton() function was not handling scoped IPv6 addresses (those
4727 with, for example, "%eth0" on the end) correctly.
4729 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4730 negated items (that is, ~something) in unsigned ints. Some compilers
4731 apparently mutter when there is no cast.
4733 70. If an address verification called from an ACL failed, and did not produce a
4734 user-specific message (i.e. there was only a "system" message), nothing was
4735 put in $acl_verify_message. In this situation, it now puts the system
4738 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4739 session; change 4.31/43 added the unwanted input to the log line - except
4740 that it did not do this in the start of session case. It now does.
4742 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4743 This is wrong and can cause the other end to generate a synchronization
4744 error if it is another Exim or anything else that does the synchronization
4745 check. A QUIT command is no longer sent after a timeout.
4747 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4748 during host lookups.
4750 74. The defer_ok option of callout verification was not working if it was used
4751 when verifying addresses in header lines, that is, for this case:
4753 verify = header_sender/callout=defer_ok
4755 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4756 those file descriptors could be used for SMTP connections. If anything
4757 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4758 could be sent to the SMTP client, causing chaos. The daemon now opens
4759 stdin, stdout, and stderr to /dev/null when it puts itself into the
4762 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4763 log by default. The user can override this with suitable Perl magic.
4765 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4766 the log message when discard triggers, was not working for the DATA ACL or
4767 for the non-SMTP ACL.
4769 78. Error message wording change in sieve.c.
4771 79. If smtp_accept_max_per_host was set, the number of connections could be
4772 restricted to fewer than expected, because the daemon was trying to set up
4773 a new connection before checking whether the processes handling previous
4774 connections had finished. The check for completed processes is now done
4775 earlier. On busy systems, this bug wouldn't be noticed because something
4776 else would have woken the daemon, and it would have reaped the completed
4779 80. If a message was submitted locally by a user whose login name contained one
4780 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4781 It caused a spool format error. I have fixed the spool reading code. A
4782 related problem was that the "from" clause in the Received: line became
4783 illegal because of the space(s). It is now covered by ${quote_local_part.
4785 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4788 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4790 83. Patch from Sieve maintainer to fix unterminated string problem in
4791 "vacation" handling.
4793 84. Some minor changes to the Linux configuration files to help with other
4794 OS variants using glibc.
4796 85. One more patch for Sieve to update vacation handling to latest spec.
4799 ----------------------------------------------------
4800 See the note above about the 4.44 and 4.50 releases.
4801 ----------------------------------------------------
4807 1. Change 4.43/35 introduced a bug that caused file counts to be
4808 incorrectly computed when quota_filecount was set in an appendfile
4811 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4812 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4815 3. Give more explanation in the error message when the command for a transport
4816 filter fails to execute.
4818 4. There are several places where Exim runs a non-Exim command in a
4819 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4820 was being done only for the command run by the queryprogram router. It is
4821 now done for all such subprocesses. The other cases are: ${run, transport
4822 filters, and the commands run by the lmtp and pipe transports.
4824 5. Some older OS have a limit of 256 on the maximum number of file
4825 descriptors. Exim was using setrlimit() to set 1000 as a large value
4826 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4827 systems. I've change it so that if it can't get 1000, it tries for 256.
4829 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4830 was an oversight, and furthermore, ever since the addition of extra
4831 controls (e.g. 4.43/32), the checks on when to allow different forms of
4832 "control" were broken. There should now be diagnostics for all cases when a
4833 control that does not make sense is encountered.
4835 7. $recipients is now available in the predata ACL (oversight).
4837 8. Tidy the search cache before the fork to do a delivery from a message
4838 received from the command line. Otherwise the child will trigger a lookup
4839 failure and thereby defer the delivery if it tries to use (for example) a
4840 cached ldap connection that the parent has called unbind on.
4842 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4843 of $address_data from the recipient verification was clobbered by the
4844 sender verification.
4846 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4847 was its contents. (It was OK if the option was not defined at all.)
4849 11. A "Completed" log line is now written for messages that are removed from
4850 the spool by the -Mrm option.
4852 12. $host_address is now set to the target address during the checking of
4853 ignore_target_hosts.
4855 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4856 being passed; this would have caused $sender_host_name to have been used if
4857 matching the list had actually called for a host name (not very likely,
4858 since this list is usually IP addresses). A host name is now passed as
4861 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4862 code that specifies a non-restarting handler (typically sigaction() in
4863 modern systems) in an attempt to fix a rare and obscure crash bug.
4865 15. Narrowed the window for a race in the daemon that could cause it to ignore
4866 SIGCHLD signals. This is not a major problem, because they are used only to
4867 wake it up if nothing else does.
4869 16. A malformed maildirsize file could cause Exim to calculate negative values
4870 for the mailbox size or file count. Odd effects could occur as a result.
4871 The maildirsize information is now recalculated if the size or filecount
4874 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4875 support for a long time. Removed HAVE_SYS_VFS_H.
4877 18. Updated exipick to current release from John Jetmore.
4879 19. Allow an empty sender to be matched against a lookup in an address list.
4880 Previously the only cases considered were a regular expression, or an
4883 20. Exim went into a mad DNS lookup loop when doing a callout where the
4884 host was specified on the transport, if the DNS lookup yielded more than
4887 21. The RFC2047 encoding function was originally intended for short strings
4888 such as real names; it was not keeping to the 75-character limit for
4889 encoded words that the RFC imposes. It now respects the limit, and
4890 generates multiple encoded words if necessary. To be on the safe side, I
4891 have increased the buffer size for the ${rfc2047: expansion operator from
4894 22. Failure to deliver a bounce message always caused it to be frozen, even if
4895 there was an errors_to setting on the router. The errors_to setting is now
4898 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4899 canonical form (fully expanded) before being placed in
4900 $sender_host_address.
4902 24. Updated eximstats to version 1.33
4904 25. Include certificate and key file names in error message when GnuTLS fails
4905 to set them up, because the GnuTLS error message doesn't include the name
4906 of the failing file when there is a problem reading it.
4908 26. Expand error message when OpenSSL has problems setting up cert/key files.
4911 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4912 (this can affect the format of dates).
4914 28. exim_tidydb, when checking for the continued existence of a message for
4915 which it has found a message-specific retry record, was not finding
4916 messages that were in split spool directories. Consequently, it was
4917 deleting retry records that should have stayed in existence.
4919 29. eximstats updated to version 1.35
4920 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4921 1.35 - bugfix such that pie charts by volume are generated correctly
4923 30. The SPA authentication driver was not abandoning authentication and moving
4924 on to the next authenticator when an expansion was forced to fail,
4925 contradicting the general specification for all authenticators. Instead it
4926 was generating a temporary error. It now behaves as specified.
4928 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4929 (the order specifies the preference for clients). The order is now AES256,
4930 AES128, 3DES, ARCFOUR128.
4932 31. Small patch to Sieve code - explicitly set From: when generating an
4935 32. Exim crashed if a remote delivery caused a very long error message to be
4936 recorded - for instance if somebody sent an entire SpamAssassin report back
4937 as a large number of 550 error lines. This bug was coincidentally fixed by
4938 increasing the size of one of Exim's internal buffers (big_buffer) that
4939 happened as part of the Exiscan merge. However, to be on the safe side, I
4940 have made the code more robust (and fixed the comments that describe what
4943 33. Some experimental protocols are using DNS PTR records for new purposes. The
4944 keys for these records are domain names, not reversed IP addresses. The
4945 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4946 leaves it alone. Component reversal etc. now happens only for IP addresses.
4949 34. The host_aton() function is supposed to be passed a string that is known
4950 to be a valid IP address. However, in the case of IPv6 addresses, it was
4951 not checking this. This is a hostage to fortune. Exim now panics and dies
4952 if the condition is not met. A case was found where this could be provoked
4953 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4954 components; fortuitously, this particular loophole had already been fixed
4955 by change 4.50/55 or 4.44/33 above.
4957 If there are any other similar loopholes, the new check in host_aton()
4958 itself should stop them being exploited. The report I received stated that
4959 data on the command line could provoke the exploit when Exim was running as
4960 exim, but did not say which command line option was involved. All I could
4961 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4962 running as the user.
4965 35. There was a buffer overflow vulnerability in the SPA authentication code
4966 (which came originally from the Samba project). I have added a test to the
4967 spa_base64_to_bits() function which I hope fixes it.
4970 36. The daemon start-up calls getloadavg() while still root for those OS that
4971 need the first call to be done as root, but it missed one case: when
4972 deliver_queue_load_max is set with deliver_drop_privilege. This is
4973 necessary for the benefit of the queue runner, because there is no re-exec
4974 when deliver_drop_privilege is set.
4976 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4977 was in use, was not putting the data itself into the right store pool;
4978 consequently, it could be overwritten for a subsequent message in the same
4979 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4982 38. Sometimes the final signoff response after QUIT could fail to get
4983 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4984 < 0 before doing a fflush(). This bug looks as though it goes back to the
4985 introduction of TLS in release 3.20, but "sometimes" must have been rare
4986 because the tests only now provoked it.
4992 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4993 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4994 Now it returns an integer. A number of calls were still expecting a T/F
4995 return. Fortuitously, in all cases, the tests worked in OK situations,
4996 which is the norm. However, things would have gone wrong on any write
4997 failures on the smtp file descriptor. This function is used when sending
4998 messages over SMTP and also when doing verify callouts.
5000 2. When Exim is called to do synchronous delivery of a locally submitted
5001 message (the -odf or -odi options), it no longer closes stderr before doing
5004 3. Implemented the mua_wrapper option.
5006 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5008 5. Implemented the functions header_remove(), header_testname(),
5009 header_add_at_position(), and receive_remove_recipient(), and exported them
5012 6. If an ACL "warn" statement specified the addition of headers, Exim already
5013 inserted X-ACL-Warn: at the start if there was no header name. However, it
5014 was not making this test for the second and subsequent header lines if
5015 there were newlines in the string. This meant that an invalid header could
5016 be inserted if Exim was badly configured.
5018 7. Allow an ACL "warn" statement to add header lines at the start or after all
5019 the Received: headers, as well as at the end.
5021 8. Added the rcpt_4xx retry error code.
5023 9. Added postmaster_mailfrom=xxx to callout verification option.
5025 10. Added mailfrom=xxxx to the callout verification option, for verify=
5028 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5029 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5031 12. Inserted some casts to stop certain compilers warning when using pointer
5032 differences as field lengths or precisions in printf-type calls (mostly
5033 affecting debugging statements).
5035 13. Added optional readline() support for -be (dynamically loaded).
5037 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5038 same clock tick as a message's arrival, so that its received time was the
5039 same as the "first fail" time on the retry record, and that message
5040 remained on the queue past the ultimate address timeout, every queue runner
5041 would try a delivery (because it was past the ultimate address timeout) but
5042 after another failure, the ultimate address timeout, which should have then
5043 bounced the address, did not kick in. This was a "< instead of <=" error;
5044 in most cases the first failure would have been in the next clock tick
5045 after the received time, and all would be well.
5047 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5048 being recognized when the domain list was tested by the match_domain
5049 condition in an expansion string.
5051 16. Added the ${str2b64: operator.
5053 17. Exim was always calling setrlimit() to set a large limit for the number of
5054 processes, without checking whether the existing limit was already
5055 adequate. (It did check for the limit on file descriptors.) Furthermore,
5056 errors from getrlimit() and setrlimit() were being ignored. Now they are
5057 logged to the main and panic logs, but Exim does carry on, to try to do its
5058 job under whatever limits there are.
5060 18. Imported PCRE 5.0.
5062 19. Trivial typo in log message " temporarily refused connection" (the leading
5065 20. If the log selector return_path_on_delivery was set and an address was
5066 redirected to /dev/null, the delivery process crashed because it assumed
5067 that a return path would always be set for a "successful" delivery. In this
5068 case, the whole delivery is bypassed as an optimization, and therefore no
5071 21. Internal re-arrangement: the function for sending a challenge and reading
5072 a response while authentication was assuming a zero-terminated challenge
5073 string. It's now changed to take a pointer and a length, to allow for
5074 binary data in such strings.
5076 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5078 23. Exim was not respecting finduser_retries when seeking the login of the
5079 uid under which it was called; it was always trying 10 times. (The default
5080 setting of finduser_retries is zero.) Also, it was sleeping after the final
5081 failure, which is pointless.
5083 24. Implemented tls_on_connect_ports.
5085 25. Implemented acl_smtp_predata.
5087 26. If the domain in control=submission is set empty, Exim assumes that the
5088 authenticated id is a complete email address when it generates From: or
5089 Sender: header lines.
5091 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5092 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5093 chown and chgrp in /bin and hostname in /usr/bin.
5095 28. Exim was keeping the "process log" file open after each use, just as it
5096 does for the main log. This opens the possibility of it remaining open for
5097 long periods when the USR1 signal hits a daemon. Occasional processlog
5098 errors were reported, that could have been caused by this. Anyway, it seems
5099 much more sensible not to leave this file open at all, so that is what now
5102 29. The long-running daemon process does not normally write to the log once it
5103 has entered its main loop, and it closes the log before doing so. This is
5104 so that log files can straightforwardly be renamed and moved. However,
5105 there are a couple of unusual error situations where the daemon does write
5106 log entries, and I had neglected to close the log afterwards.
5108 30. The text of an SMTP error response that was received during a remote
5109 delivery was being truncated at 512 bytes. This is too short for some of
5110 the long messages that one sometimes sees. I've increased the limit to
5113 31. It is now possible to make retry rules that apply only when a message has a
5114 specific sender, in particular, an empty sender.
5116 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5117 it possible to be selective about when SMTP synchronization is enforced.
5119 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5121 32. Implemented hosts_connection_nolog.
5123 33. Added an ACL for QUIT.
5125 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5128 35. Added mailbox_size and mailbox_filecount to appendfile.
5130 36. Added control = no_multiline_responses to ACLs.
5132 37. There was a bug in the logic of the code that waits for the clock to tick
5133 in the case where the clock went backwards by a substantial amount such
5134 that the microsecond fraction of "now" was more than the microsecond
5135 fraction of "then" (but the whole seconds number was less).
5137 38. Added support for the libradius Radius client library this is found on
5138 FreeBSD (previously only the radiusclient library was supported).
5144 1. When certain lookups returned multiple values in the form name=value, the
5145 quoting of the values was not always being done properly. Specifically:
5146 (a) If the value started with a double quote, but contained no whitespace,
5148 (b) If the value contained whitespace other than a space character (i.e.
5149 tabs or newlines or carriage returns) it was not quoted.
5150 This fix has been applied to the mysql and pgsql lookups by writing a
5151 separate quoting function and calling it from the lookup code. The fix
5152 should probably also be applied to nisplus, ibase and oracle lookups, but
5153 since I cannot test any of those, I have not disturbed their existing code.
5155 2. A hit in the callout cache for a specific address caused a log line with no
5156 reason for rejecting RCPT. Now it says "Previous (cached) callout
5157 verification failure".
5159 3. There was an off-by-one bug in the queryprogram router. An over-long
5160 return line was truncated at 256 instead of 255 characters, thereby
5161 overflowing its buffer with the terminating zero. As well as fixing this, I
5162 have increased the buffer size to 1024 (and made a note to document this).
5164 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5165 when Exim is waiting for an SMTP response from a remote server, Exim
5166 restarts its select() call on the socket, thereby resetting its timeout.
5167 This is not a problem when such interrupts are rare. Somebody set up a cron
5168 job to run exiwhat every 2 minutes, which is less than the normal select()
5169 timeout (5 or 10 minutes). This meant that the select() timeout never
5170 kicked in because it was always reset. I have fixed this by comparing the
5171 time when an interrupt arrives with the time at the start of the first call
5172 to select(). If more time than the timeout has elapsed, the interrupt is
5173 treated as a timeout.
5175 5. Some internal re-factoring in preparation for the addition of Sieve
5176 extensions (by MH). In particular, the "personal" test is moved to a
5177 separate function, and given an option for scanning Cc: and Bcc: (which is
5178 not set for Exim filters).
5180 6. When Exim created an email address using the login of the caller as the
5181 local part (e.g. when creating a From: or Sender: header line), it was not
5182 quoting the local part when it contained special characters such as @.
5184 7. Installed new OpenBSD configuration files.
5186 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5187 try to make them clearer.
5189 9. Callout options, other than the timeout value, were being ignored when
5190 verifying sender addresses in header lines. For example, when using
5192 verify = header_sender/callout=no_cache
5194 the cache was (incorrectly) being used.
5196 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5197 only the Cygwin environment.
5199 11. When return_path_on_delivery was set as a log selector, if different remote
5200 addresses in the same message used different return paths and parallel
5201 remote delivery occurred, the wrong values would sometimes be logged.
5202 (Whenever a remote delivery process finished, the return path value from
5203 the most recently started remote delivery process was logged.)
5205 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5206 lines when AUTH and/or TLS are in use. This is the "received protocol"
5207 field. Exim used to use "asmtp" for authenticated SMTP, without any
5208 indication (in the protocol name) for TLS use. Now it follows the RFC and
5209 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5210 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5211 names appear in log lines as well as in Received: header lines.
5213 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5214 extensions, and comparison tests, and to fix some bugs.
5216 14. Changes to the "personal" filter test:
5218 (1) The test was buggy in that it was just doing the equivalent of
5219 "contains" tests on header lines. For example, if a user's address was
5220 anne@some.where, the "personal" test would incorrectly be true for
5222 To: susanne@some.where
5224 This test is now done by extracting each address from the header in turn,
5225 and checking the entire address. Other tests that are part of "personal"
5226 are now done using regular expressions (for example, to check local parts
5227 of addresses in From: header lines).
5229 (2) The list of non-personal local parts in From: addresses has been
5230 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5231 taken from the Sieve specification recommendations.
5233 (3) If the message contains any header line starting with "List-" it is
5234 treated as non-personal.
5236 (4) The test for "circular" in the Subject: header line has been removed
5237 because it now seems ill-conceived.
5239 15. Minor typos in src/EDITME comments corrected.
5241 16. Installed latest exipick from John Jetmore.
5243 17. If headers_add on a router specified a text string that was too long for
5244 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5245 of string_sprintf() is now avoided.
5247 18. $message_body_size was not set (it was always zero) when running the DATA
5248 ACL and the local_scan() function.
5250 19. For the "mail" command in an Exim filter, no default was being set for
5251 the once_repeat time, causing a random time value to be used if "once" was
5252 specified. (If the value happened to be <= 0, no repeat happened.) The
5253 default is now 0s, meaning "never repeat". The "vacation" command was OK
5254 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5255 (I found it when inspecting the code).
5257 20. There is now an overall timeout for performing a callout verification. It
5258 defaults to 4 times the callout timeout, which applies to individual SMTP
5259 commands during the callout. The overall timeout applies when there is more
5260 than one host that can be tried. The timeout is checked before trying the
5261 next host. This prevents very long delays if there are a large number of
5262 hosts and all are timing out (e.g. when the network connections are timing
5263 out). The value of the overall timeout can be changed by specifying an
5264 additional sub-option for "callout", called "maxwait". For example:
5266 verify = sender/callout=5s,maxwait=20s
5268 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5269 to the end before writing, but this should make it even safer).
5271 22. Exim was forgetting that it had advertised PIPELINING for the second and
5272 subsequent messages on an SMTP connection. It was also not resetting its
5273 memory on STARTTLS and an internal HELO.
5275 23. When Exim logs an SMTP synchronization error within a session, it now
5276 records whether PIPELINING has been advertised or not.
5278 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5279 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5280 rather than long int.
5282 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5284 26. Added the never_mail option to autoreply.
5290 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5291 crash if the getsockname() call failed; this can happen if a connection is
5292 closed very soon after it is established. The problem was simply in the
5293 order in which certain operations were done, causing Exim to try to write
5294 to the SMTP stream before it had set up the file descriptor. The bug has
5295 been fixed by making things happen in the correct order.
5301 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5302 before the connection was closed, thus losing the rejection response.
5304 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5305 some early Solaris releases, but causes trouble in current releases where
5306 socklen_t is defined.
5308 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5311 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5314 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5315 that releasing the top of it at the end releases what was used for sub-
5316 expansions (unless the block got too big). However, discard this block if
5317 the first thing is a variable or header, so that we can use its block when
5318 it is dynamic (useful for very large $message_headers, for example).
5320 6. Lookups now cache *every* query, not just the most recent. A new, separate
5321 store pool is used for this. It can be recovered when all lookup caches are
5322 flushed. Lookups now release memory at the end of their result strings.
5323 This has involved some general refactoring of the lookup sources.
5325 7. Some code has been added to the store_xxx() functions to reduce the amount
5326 of flapping under certain conditions.
5328 8. log_incoming_interface used to affect only the <= reception log lines. Now
5329 it causes the local interface and port to be added to several more SMTP log
5330 lines, for example "SMTP connection from", and rejection lines.
5332 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5334 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5336 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5337 could be overwritten at the end of the current message (or the start of a
5338 new message if it was set in a HELO ACL). The value is now preserved for
5339 the duration of the SMTP connection.
5341 12. If a transport had a headers_rewrite setting, and a matching header line
5342 contained an unqualified address, that address was qualified, even if it
5343 did not match any rewriting rules. The underlying bug was that the values
5344 of the flags that permit the existence of unqualified sender and recipient
5345 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5346 non-local messages, and by -bnq for local messages) were not being
5347 preserved with the message after it was received.
5349 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5350 "next input=" as part of the text comprising the host identity instead of
5351 the correct text. The code was using the same buffer for two different
5352 strings. However, depending on which order the printing function evaluated
5353 its arguments, the bug did not always show up. Under Linux, for example, my
5354 test suite worked just fine.
5356 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5357 doesn't work with some older versions of Perl. It has been changed to "my",
5358 which in any case is probably the better facility to use.
5360 15. A really picky compiler found some instances of statements for creating
5361 error messages that either had too many or two few arguments for the format
5364 16. The size of the buffer for calls to the DNS resolver has been increased
5365 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5366 for addresses that have a lot of PTR records. This alleviates a problem; it
5367 does not fully solve it.
5369 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5370 buffer now truncates the list and logs the incident, which is the same
5371 action as happens when Exim is looking up a host name and its aliases.
5372 Previously in this situation something unpredictable would happen;
5373 sometimes it was "internal error: store_reset failed".
5375 18. If a server dropped the connection unexpectedly when an Exim client was
5376 using GnuTLS and trying to read a response, the client delivery process
5377 crashed while trying to generate an error log message.
5379 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5380 string, for example:
5382 warn message = H1: something\nH2: something
5384 the text was added as a single header line from Exim's point of view
5385 though it ended up OK in the delivered message. However, searching for the
5386 second and subsequent header lines using $h_h2: did not work. This has been
5387 fixed. Similarly, if a system filter added multiple headers in this way,
5388 the routers could not see them.
5390 20. Expanded the error message when iplsearch is called with an invalid key to
5391 suggest using net-iplsearch in a host list.
5393 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5394 ACLs are no longer actually imposed (and a message to that effect is
5397 22. If a "gecos" field in a passwd entry contained escaped characters, in
5398 particular, if it contained a \" sequence, Exim got it wrong when building
5399 a From: or a Sender: header from that name. A second bug also caused
5400 incorrect handling when an unquoted " was present following a character
5401 that needed quoting.
5403 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5404 was not being matched caselessly.
5406 24. Arranged for all hyphens in the exim.8 source to be escaped with
5409 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5410 properly. Recipient callout cache records were still being keyed to include
5411 the sender, even when use_sender was set false. This led to far more
5412 callouts that were necessary. The sender is no longer included in the key
5413 when use_sender is false.
5415 26. Added "control = submission" modifier to ACLs.
5417 27. Added the ${base62d: operator to decode base 62 numbers.
5419 28. dnsdb lookups can now access SRV records.
5421 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5422 the configuration file.
5424 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5425 (-v) mode. This makes the output for a verbose queue run more intelligible.
5427 31. Added a use_postmaster feature to recipient callouts.
5429 32. Added the $body_zerocount variable, containing the number of binary zero
5430 bytes in the message body.
5432 33. The time of last modification of the "new" subdirectory is now used as the
5433 "mailbox time last read" when there is a quota error for a maildir
5436 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5438 35. Added +ignore_unknown as a special item in host lists.
5440 36. Code for decoding IPv6 addresses in host lists is now included, even if
5441 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5442 address was recognized as an IP address, but was then not correctly decoded
5443 into binary, causing unexpected and incorrect effects when compared with
5450 1. Very minor rewording of debugging text in manualroute to say "list of
5451 hosts" instead of "hostlist".
5453 2. If verify=header_syntax was set, and a header line with an unqualified
5454 address (no domain) and a large number of spaces between the end of the
5455 name and the colon was received, the reception process suffered a buffer
5456 overflow, and (when I tested it) crashed. This was caused by some obsolete
5457 code that should have been removed. The fix is to remove it!
5459 3. When running in the test harness, delay a bit after writing a bounce
5460 message to get a bit more predictability in the log output.
5462 4. Added a call to search_tidyup() just before forking a reception process. In
5463 theory, someone could use a lookup in the expansion of smtp_accept_max_
5464 per_host which, without the tidyup, could leave open a database connection.
5466 5. Added the variables $recipient_data and $sender_data which get set from a
5467 lookup success in an ACL "recipients" or "senders" condition, or a router
5468 "senders" option, similar to $domain_data and $local_part_data.
5470 6. Moved the writing of debug_print from before to after the "senders" test
5473 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5474 problems for message scanning, either using a data ACL, or using
5475 local_scan() because the Received: header was not generated till after they
5476 were called (in order to set the time as the time of reception completion).
5477 I have revised the way this works. The header is now generated after the
5478 body is received, but before the ACL or local_scan() are called. After they
5479 are run, the timestamp in the header is updated.
5485 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5486 before starting a queue runner without re-exec. This happened only when
5487 deliver_drop_privilege was set or when the Exim user was set to root. The
5488 effect of the bug was that timeouts during subsequent deliveries caused
5489 crashes instead of being properly handled. The handler is now left at its
5490 default (and expected) setting.
5492 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5493 message, again when deliver_drop_privilege is set or Exim is run as root.
5494 The bug described in (1) was not present in this case, but the tidying up
5495 of the other signals was missing. I have made the two cases consistent.
5497 3. The ignore_target_hosts setting on a manualroute router was being ignored
5498 for hosts that were looked up using the /MX notation.
5500 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5503 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5504 operated on the sender address. After changing the $sender_address to <>
5505 for the sender address verify, Exim was re-instated it as the original
5506 (before rewriting) address, but remembering that it had rewritten it, so it
5507 wasn't rewriting it again. This bug also had the effect of breaking the
5508 sender address verification caching when the sender address was rewritten.
5510 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5511 This has been changed so that if the ip literal address matches
5512 ignore_target_hosts, the router declines.
5514 7. Added expansion conditions match_domain, match_address, and match_local_
5515 part (NOT match_host).
5517 8. The placeholder for the Received: header didn't have a length field set.
5519 9. Added code to Exim itself and to exim_lock to test for a specific race
5520 condition that could lead to file corruption when using MBX delivery. The
5521 issue is with the lockfile that is created in /tmp. If this file is removed
5522 after a process has opened it but before that process has acquired a lock,
5523 there is the potential for a second process to recreate the file and also
5524 acquire a lock. This could lead to two Exim processes writing to the file
5525 at the same time. The added code performs the same test as UW imapd; it
5526 checks after acquiring the lock that its file descriptor still refers to
5527 the same named file.
5529 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5530 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5533 11. Added the smtp_active_hostname option. If used, this will typically be made
5534 to depend on the incoming interface address. Because $interface_address is
5535 not set up until the daemon has forked a reception process, error responses
5536 that can happen earlier (such as "too many connections") no longer contain
5539 12. If an expansion in a condition on a "warn" statement fails because a lookup
5540 defers, the "warn" statement is abandoned, and the next ACL statement is
5541 processed. Previously this caused the whole ACL to be aborted.
5543 13. Added the iplsearch lookup type.
5545 14. Added ident_timeout as a log selector.
5547 15. Added tls_certificate_verified as a log selector.
5549 16. Added a global option tls_require_ciphers (compare the smtp transport
5550 option of the same name). This controls incoming TLS connections.
5552 17. I finally figured out how to make tls_require_ciphers do a similar thing
5553 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5554 before starting the TLS session.
5556 18. Tabs are now shown as \t in -bP output.
5558 19. If the log selector return_path_on_delivery was set, Exim crashed when
5559 bouncing a message because it had too many Received: header lines.
5561 20. If two routers both had headers_remove settings, and the first one included
5562 a superfluous trailing colon, the final name in the first list and the
5563 first name in the second list were incorrectly joined into one item (with a
5564 colon in the middle).
5570 1. Added -C and -D options to the exinext utility, mainly to make it easier
5571 to include in the automated testing, but these could be helpful when
5572 multiple configurations are in use.
5574 2. The exinext utility was not formatting the output nicely when there was
5575 an alternate port involved in the retry record key, nor when there was a
5576 message id as well (for retries that were specific to a specific message
5577 and a specific host). It was also confused by IPv6 addresses, because of
5578 the additional colons they contain. I have fixed the IPv4 problem, and
5579 patched it up to do a reasonable job for IPv6.
5581 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5582 delivery, the log line now contains "pipelined" if PIPELINING was used.
5584 4. An SMTP transport process used to panic and die if the bind() call to set
5585 an explicit outgoing interface failed. This has been changed; it is now
5586 treated in the same way as a connect() failure.
5588 5. A reference to $sender_host_name in the part of a conditional expansion
5589 that was being skipped was still causing a DNS lookup. This no longer
5592 6. The def: expansion condition was not recognizing references to header lines
5593 that used bh_ and bheader_.
5595 7. Added the _cache feature to named lists.
5597 8. The code for checking quota_filecount in the appendfile transport was
5598 allowing one more file than it should have been.
5600 9. For compatibility with Sendmail, the command line option
5608 and sets the incoming protocol and host name (for trusted callers). The
5609 host name and its colon can be omitted when only the protocol is to be set.
5610 Note the Exim already has two private options, -pd and -ps, that refer to
5611 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5612 "s", but I don't think that's a major issue.
5614 10. A number of refactoring changes to the code, none of which should affect
5617 (a) The number of logging options was getting close to filling up the
5618 32-bit word that was used as a bit map. I have split them into two classes:
5619 those that are passed in the argument to log_write(), and those that are
5620 only ever tested independently outside of that function. These are now in
5621 separate 32-bit words, so there is plenty of room for expansion again.
5622 There is no change in the user interface or the logging behaviour.
5624 (b) When building, for example, log lines, the code previously used a
5625 macro that called string_cat() twice, in order to add two strings. This is
5626 not really sufficiently general. Furthermore, there was one instance where
5627 it was actually wrong because one of the argument was used twice, and in
5628 one call a function was used. (As it happened, calling the function twice
5629 did not affect the overall behaviour.) The macro has been replaced by a
5630 function that can join an arbitrary number of extra strings onto a growing
5633 (c) The code for expansion conditions now uses a table and a binary chop
5634 instead of a serial search (which was left over from when there were very
5635 few conditions). Also, it now recognizes conditions like "pam" even when
5636 the relevant support is not compiled in: a suitably worded error message is
5637 given if an attempt is made to use such a condition.
5639 11. Added ${time_interval:xxxxx}.
5641 12. A bug was causing one of the ddress fields not to be passed back correctly
5642 from remote delivery subprocesses. The field in question was not being
5643 subsequently used, so this caused to problems in practice.
5645 13. Added new log selectors queue_time and deliver_time.
5647 14. Might have fixed a bug in maildirsizefile handling that threw up
5648 "unexpected character" debug warnings, and recalculated the data
5649 unnecessarily. In any case, I expanded the warning message to give more
5652 15. Added the message "Restricted characters in address" to the statements in
5653 the default ACL that block characters like @ and % in local parts.
5655 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5656 Three changes have been made:
5658 (a) There was a serious bug; a negative response to MAIL caused the whole
5659 recipient domain to be cached as invalid, thereby blocking all messages
5660 to all local parts at the same domain, from all senders. This bug has
5661 been fixed. The domain is no longer cached after a negative response to
5662 MAIL if the sender used is not empty.
5664 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5667 (c) A new callout option, "use_sender" has been added for people who want
5668 the modified behaviour.
5674 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5677 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5678 indeed breaks things for older releases.
5680 3. Added additional logging to the case where there is a problem reading data
5681 from a filter that is running in a subprocess using a pipe, in order to
5682 try to track down a specific problem.
5684 4. Testing facility fudge: when running in the test harness and attempting
5685 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5686 getting "No route to host". Convert this to a timeout.
5688 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5691 6. Some OS don't have socklen_t but use size_t instead. This affects the
5692 fifth argument of getsockopt() amongst other things. This is now
5693 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5694 can be set for individual OS. I have set it for SunOS5, OSF1, and
5695 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5696 some earlier ones do not.
5698 7. Change 4.30/15 was not doing the test caselessly.
5700 8. The standard form for an IPv6 address literal was being rejected by address
5701 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5702 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5703 this, as well as the form without the "IPv6" on the front (but only when
5704 address literals are enabled, of course).
5706 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5708 10. Exim crashed if a message with an empty sender address specified by -f
5709 encountered a router with an errors_to setting. This could be provoked only
5710 by a command such as
5714 where an empty string was supplied; "<>" did not hit this bug.
5716 11. Installed PCRE release 4.5.
5718 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5719 remained set. It is now erased.
5721 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5722 times from message ids (which are base 36 rather than the normal 62).
5724 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5725 were being counted as actual protocol errors, and logged if the log
5726 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5727 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5728 and DATA following a set of rejected RCPTs do not count as protocol errors.
5729 In other words, Exim assumes they were pipelined, though this may not
5730 actually be the case. Of course, in all cases the client gets an
5731 appropriate error code.
5733 15. If a lookup fails in an ACL condition, a message about the failure may
5734 be available; it is used if testing the ACL cannot continue, because most
5735 such messages specify what the cause of the deferral is. However, some
5736 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5737 that caused an old message to be retained and used if a later statement
5738 caused a defer, replacing the real cause of the deferral.
5740 16. If an IP address had so many PTR records that the DNS lookup buffer
5741 was not large enough to hold them, Exim could crash while trying to process
5742 the truncated data. It now detects and logs this case.
5744 17. Further to 4.21/58, another change has been made: if (and only if) the
5745 first line of a message (the first header line) ends with CRLF, a bare LF
5746 in a subsequent header line has a space inserted after it, so as not to
5747 terminate the header.
5749 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5750 unnecessarily, used atoi() instead of strtol(), and didn't check the
5751 termination when getting file sizes from file names by regex.
5753 19. Completely re-implemented the support for maildirsize files, in the light
5754 of a number of problems with the previous contributed implementation
5755 (4.30/29). In particular:
5757 . If the quota is zero, the maildirsize file is maintained, but no quota is
5760 . If the maildir directory does not exist, it is created before any attempt
5761 to write a maildirsize file.
5763 . The quota value in the file is just a cache; if the quota is changed in
5764 the transport, the new value overrides.
5766 . A regular expression is available for excluding directories from the
5769 20. The autoreply transport checks the characters in options that define the
5770 message's headers; it allows continued headers, but it was checking with
5771 isspace() after an embedded newline instead of explicitly looking for a
5774 21. If all the "regular" hosts to which an address was routed had passed their
5775 expiry times, and had not reached their retry times, the address was
5776 bounced, even if fallback hosts were defined. Now Exim should go on to try
5779 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5780 equivalent code in the SMTP transport. Some hosts send humungous responses
5781 to HELO/EHLO, more than 1024 it seems.
5783 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5784 gives compiler warnings in some environments. I've now done it "properly",
5787 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5788 (because of problems with the built-in one) was declared to return uschar *
5789 instead of char *, causing compiler failure.
5791 25. Fixed a file descriptor leak when processing alias/forward files.
5793 26. Fixed a minor format string issue in dbfn.c.
5795 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5797 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5798 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5801 29. When a sender address is verified, it is cached, to save repeating the test
5802 when there is more than one recipient in a message. However, when the
5803 verification involves a callout, it is possible for different callout
5804 options to be set for different recipients. It is too complicated to keep
5805 track of this in the cache, so now Exim always runs a verification when a
5806 callout is required, relying on the callout cache for the optimization.
5807 The overhead is duplication of the address routing, but this should not be
5810 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5811 to be verified with callout=postmaster, and the main callout worked but the
5812 postmaster check failed, the verification correctly failed. However, if a
5813 subsequent RCPT command asked for sender verification *without* the
5814 postmaster check, incorrect caching caused this verification also to fail,
5817 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5818 it was not caching the DNS options (qualify_single, search_parents) that
5819 were used when the lookup failed. A subsequent lookup with different
5820 options therefore always gave the same answer, though there were cases
5821 where it should not have. (Example: a "domains = !$mx_any" option on a
5822 dnslookup router: the "domains" option is always processed without any
5823 widening, but the router might have qualify_single set.) Now Exim uses the
5824 cached value only when the same options are set.
5826 32. Added John Jetmore's "exipick" utility to the distribution.
5828 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5829 than a timeout (e.g. a certificate is required, and is not provided), an
5830 Exim server now closes the connection immediately. Previously it waited for
5831 the client to close - but if the client is SSL, it seems that they each
5832 wait for each other, leading to a delay before one of them times out.
5834 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5835 maintained 0.8.x compatibility because I don't think many are using it, and
5836 it is clearly obsolete.
5838 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5841 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5842 client certificate was expired. A simple patch fixes this, though I don't
5843 understand the full logic of why the verify callback is called multiple
5846 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5847 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5848 which causes problems with some clients (such as the Certicom SSL Plus
5849 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5850 disables the coutermeasure allowing Eudora to connect."
5852 38. Exim was not checking that a write() to a log file succeeded. This could
5853 lead to Bad Things if a log got too big, in particular if it hit a file
5854 size limit. Exim now panics and dies if it cannot write to a log file, just
5855 as it does if it cannot open a log file.
5857 39. Modified OS/Makefile-Linux so that it now contains
5859 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5861 The two -D definitions ensure that Exim is compiled with large file
5862 support, which makes it possible to handle log files that are bigger than
5865 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5866 instance) a domain was checked against a named list that involved a lookup,
5867 causing $domain_data to be set, then another domain was checked against the
5868 same list, then the first domain was re-checked, the value of $domain_data
5869 after the final check could be wrong. In particular, if the second check
5870 failed, it could be set empty. This bug probably also applied to
5873 41. The strip_trailing_dot option was not being applied to the address given
5874 with the -f command-line option.
5876 42. The code for reading a message's header from the spool was incrementing
5877 $received_count, but never initializing it. This meant that the value was
5878 incorrect (doubled) while delivering a message in the same process in which
5879 it was received. In the most common configuration of Exim, this never
5880 happens - a fresh exec is done - but it can happen when
5881 deliver_drop_privilege is set.
5883 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5884 it now includes up to 150 characters of the unexpected data in the log
5887 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5888 and building data strings. The size of both of these buffers was 10 000
5889 bytes - far larger than anybody would *ever* want, thought I. Needless to
5890 say, somebody hit the limit. I have increased the maximum line length to
5891 20 000 and the maximum data length of concatenated lines to 100 000. I have
5892 also fixed two bugs, because there was no checking on these buffers. Tsk,
5893 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5894 buffer is too small.
5896 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5897 lsearch lookups. Now it does.
5899 46. When parsing a route_list item in a manualroute router, a fixed-length
5900 buffer was used for the list of hosts. I made this 1024 bytes long,
5901 thinking that nobody would ever have a list of hosts that long. Wrong.
5902 Somebody had a whole pile of complicated expansion conditions, and the
5903 string was silently truncated, leading to an expansion error. It turns out
5904 that it is easier to change to an unlimited length (owing to other changes
5905 that have happened since this code was originally written) than to build
5906 structure for giving a limitation error. The length of the item that
5907 expands into the list of hosts is now unlimited.
5909 47. The lsearch lookup could not handle data where the length of text line was
5910 more than 4095 characters. Such lines were truncated, leading to shortened
5911 data being returned. It should now handle lines of any length.
5913 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5914 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5917 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5918 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5919 no longer happens in comment lines. A list of replacements is now placed
5920 at the head of all of the source files, except those whose only change is
5921 to replace PERL_COMMAND in the very first #! line.
5923 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5924 messages on the queue, with a bottom-up merge sort, using code contributed
5925 by Michael Haardt. This should make operations like -bp somewhat faster on
5926 large queues. It won't affect queue runners, except when queue_run_in_order
5929 51. Installed eximstats 1.31 in the distribution.
5931 52. Added support for SRV lookups to the dnslookup router.
5933 53. If an ACL referred to $message_body or $message_body_end, the value was not
5934 reset for any messages that followed in the same SMTP session.
5936 54. The store-handling optimization for building very long strings was not
5937 differentiating between the different store pools. I don't think this
5938 actually made any difference in practice, but I've tidied it.
5940 55. While running the routers to verify a sender address, $sender_address
5941 was still set to the sender address. This is wrong, because when routing to
5942 send a bounce to the sender, it would be empty. Therefore, I have changed
5943 it so that, while verifying a sender address, $sender_address is set to <>.
5944 (There is no change to what happens when verifying a recipient address.)
5946 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5947 target A or AAAA records (if not already returned) without resetting the
5948 qualify_single or search_parents options of the DNS resolver. These are
5949 inappropriate in this case because the targets of MX and SRV records must
5950 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5951 target that, when qualified, matched something in the local domain. These
5952 two options are now turned off when doing these lookups.
5954 57. It seems that at least some releases of Reiserfs (which does not have the
5955 concept of a fixed number of inodes) returns zero and not -1 for the
5956 number of available inodes. This interacted badly with check_spool_inodes,
5957 which assumed that -1 was the "no such thing" setting. What I have done is
5958 to check that the total number of inodes is greater than zero before doing
5959 the test of how many are available.
5961 58. When a "warn" ACL statement has a log_message modifier, the message is
5962 remembered, and not repeated. This is to avoid a lot of repetition when a
5963 message has many recipients that cause the same warning to be written.
5964 However, Exim was preserving the list of already written lines for an
5965 entire SMTP session, which doesn't seem right. The memory is now reset if a
5966 new message is started.
5968 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5969 debugging output unless log_rewrite was also set.
5971 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5972 of (int)(handle) when we know that handle contains (void *)(-1).
5974 61. The Exim daemon panic-logs an error return when it closes the incoming
5975 connection. However "connection reset by peer" seems to be common, and
5976 isn't really an error worthy of noting specially, so that particular error
5979 62. When Exim is trying to find all the local interfaces, it used to panic and
5980 die if the ioctl to get the interface flags failed. However, it seems that
5981 on at least one OS (Solaris 9) it is possible to have an interface that is
5982 included in the list of interfaces, but for which you get a failure error
5983 for this call. This happens when the interface is not "plumbed" into a
5984 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5985 failure of the "get flags" call assumes that the interface is down.
5987 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5988 makes life easier for people who are doing arithmetic on fields extracted
5989 from dates, where you often get leading zeros that should not be
5990 interpreted as octal.
5992 64. Added qualify_domain to the redirect router, to override the global
5995 65. If a pathologically long header line contained very many addresses (the
5996 report of this problem mentioned 10 000) and each of them was rewritten,
5997 Exim could use up a very large amount of memory. (It kept on making new
5998 copies of the header line as it rewrote, and never released the old ones.)
5999 At the expense of a bit more processing, the header rewriting function has
6000 been changed so that it no longer eats memory in this way.
6002 66. The generation of the Received: header has been moved from the time that a
6003 message starts to be received, to the time that it finishes. The timestamp
6004 in the Received: header should now be very close to that of the <= log
6005 line. There are two side-effects of this change:
6007 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6008 logged header lines no longer include the local Received: line, because
6009 it has not yet been created. The same applies to a copy of the message
6010 that is returned to a non-SMTP sender when a message is rejected.
6012 (b) When a filter file is tested using -bf, no additional Received: header
6013 is added to the test message. After some thought, I decided that this
6016 This change does not affect the value of $received_for. It is still set
6017 after address rewriting, but before local_scan() is called.
6019 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6021 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6022 gave an unhelpful panic error message, and a defer error. I have managed to
6023 change this behaviour so that it now rejects any supplied certificate,
6024 which seems right, as the list of acceptable certificates is empty.
6026 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6027 gave an unhelpful defer error. I have not managed to make this reject any
6028 supplied certificates, but the error message it gives is "no certificate
6029 supplied", which is not helpful.
6031 70. exigrep's output now also includes lines that are not associated with any
6032 message, but which match the given pattern. Implemented by a patch from
6033 Martin Sluka, which also tidied up the Perl a bit.
6035 71. Recipient callout verification, like sender verification, was using <> in
6036 the MAIL FROM command. This isn't really the right thing, since the actual
6037 sender may affect whether the remote host accepts the recipient or not. I
6038 have changed it to use the actual sender in the callout; this means that
6039 the cache record is now keyed on a recipient/sender pair, not just the
6040 recipient address. There doesn't seem to be a real danger of callout loops,
6041 since a callout by the remote host to check the sender would use <>.
6042 [SEE ABOVE: changed after hitting problems.]
6044 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6045 temporary errors. However, in the case of such a code being given after
6046 the end of a data transmission (i.e. after ".") Exim was failing to write
6047 a retry record for the message. (Yes, there was some broken host that was
6048 actually sending 8xx at this point.)
6050 73. An unknown lookup type in a host list could cause Exim to panic-die when
6051 the list was checked. (An example that provoked this was putting <; in the
6052 middle of a list instead of at the start.) If this happened during a DATA
6053 ACL check, a -D file could be left lying around. This kind of configuration
6054 error no longer causes Exim to die; instead it causes a defer error. The
6055 incident is still logged to the main and panic logs.
6057 74. Buglet left over from Exim 3 conversion. The message "too many messages
6058 in one connection" was written to the rejectlog but not the mainlog, except
6059 when address rewriting (yes!) was being logged.
6061 75. Added write_rejectlog option.
6063 76. When a system filter was run not as root (that is, when system_filter_user
6064 was set), the values of the $n variables were not being returned to the
6065 main process; thus, they were not subsequently available in the $sn
6068 77. Added +return_path_on_delivery log selector.
6070 78. A connection timeout was being treated differently from recipients deferred
6071 when testing hosts_max_try with a message that was older than the host's
6072 retry timeout. (The host should not be counted, thus allowing all hosts to
6073 be tried at least once before bouncing.) This may have been the cause of an
6074 occasionally reported bug whereby a message would remain on the queue
6075 longer than the retry timeout, but would be bounced if a delivery was
6076 forced. I say "may" because I never totally pinned down the problem;
6077 setting up timeout/retry tests is difficult. See also the next item.
6079 79. The ultimate address timeout was not being applied to errors that involved
6080 a combination of host plus message (for example, a timeout on a MAIL
6081 command). When an address resolved to a number of possible hosts, and they
6082 were not all tried for each delivery (e.g. because of hosts_max_try), a
6083 message could remain on the queue longer than the retry timeout.
6085 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6088 81. Fixed an obscure SMTP outgoing bug which required at least the following
6089 conditions: (a) there was another message waiting for the same server;
6090 (b) the server returned 5xx to all RCPT commands in the first message so
6091 that the message was not completed; (c) the server dropped the connection
6092 or gave a negative response to the RSET that Exim sends to abort the
6093 transaction. The observed case was a dropped connection after DATA that had
6094 been sent in pipelining mode. That is, the server had advertised PIPELINING
6095 but was not implementing it correctly. The effect of the bug was incorrect
6096 behaviour, such as trying another host, and this could lead to a crash.
6102 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6103 and daemon.c were passed as pointers to ints; they should have been
6104 pointers to socklen_t variables (which are typically unsigned ints).
6106 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6109 3. Fixed a really odd bug that affected only the testing scheme; patching a
6110 certain fixed string in the binary changed the value of another string that
6111 happened to be identical to the end of the original first string.
6113 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6114 name", it returns that address as the IP address. On some operating
6115 systems (e.g. Solaris), it also passes back the IP address string as the
6116 "host name". However, on others (e.g. Linux), it passes back an empty
6117 string. Exim wasn't checking for this, and was changing the host name to an
6118 empty string, assuming it had been canonicalized.
6120 5. Although rare, it is permitted to have more than one PTR record for a given
6121 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6122 all the names associated with an address, because they do in Solaris.
6123 However, it seems that they do not in Linux for data that comes from the
6124 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6125 I found this out when I moved to a new Linux workstation and tried to run
6126 the Exim test suite.
6128 To get round this problem I have changed the code so that it now does its
6129 own call to the DNS to look up PTR records when searching for a host name.
6130 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6131 addresses that are only in /etc/hosts are still found.
6133 This behaviour is, however, controlled by an option called host_lookup_
6134 order, which defaults to "bydns:byaddr". If people want to use the other
6135 order, or indeed, just use one or the other means of lookup, they can
6136 specify it in this variable.
6138 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6139 some operating systems, this comes back from gethostbyaddr() as an empty
6140 string, and this is what Exim used to test for. However, it seems that in
6141 other systems, "." is yielded. Exim now tests for this case too.
6143 7. The values of check_spool_space and check_log_space are now held internally
6144 as a number of kilobytes instead of an absolute number of bytes. If a
6145 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6146 kilobyte. This means that much larger values can be stored.
6148 8. Exim monitor: an attempt to get the action menu when not actually pointing
6149 at a message produces an empty menu entitled "No message selected". This
6150 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6151 no entries in it ("Shell widget menu has zero width and/or height"). So I
6152 have added a single, blank menu entry in this case.
6154 9. Added ${quote_local_part.
6156 10. MIME decoding is now applied to the contents of Subject: header lines when
6159 11. Now that a reference to $sender_host_address automatically causes a reverse
6160 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6161 host lookup before query-style lookups in lists that might use this
6162 variable. This has therefore been abolished, and the "net-" prefix is no
6163 longer necessary for query-style lookups.
6165 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6166 have been a typo for LFLAGS, so it has been changed.
6168 13. The install script calls Exim with "-C /dev/null" in order to find the
6169 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6170 to be output. However, since Exim outputs its version number before the
6171 error, it didn't break the script. It just looked ugly. I fixed this by
6172 always allowing "-C /dev/null" if the caller is root.
6174 14. Ignore overlarge ACL variable number when reading spool file - insurance
6175 against a later release with more variables having written the file.
6177 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6178 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6179 this, as well as the form without the "IPv6" on the front.
6181 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6182 OS/Makefile-Darwin file.
6184 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6185 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6187 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6188 variable. However, some OS use ints for inodes. I've added cast to long int
6189 to get rid of the compiler warning.
6191 19. I had forgotten to lock out "/../" in configuration file names when
6192 ALT_CONFIG_PREFIX was set.
6194 20. Routers used for verification do not need to specify transports. However,
6195 if such a router generated a host list, and callout was configured, Exim
6196 crashed, because it could not find a port number from the (non-existent)
6197 transport. It now assumes port 25 in this circumstance.
6199 21. Added the -t option to exigrep.
6201 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6202 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6203 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6204 them is set, LOOKUP_LSEARCH is forced.
6206 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6208 24. Added sender and host information to the "rejected by local_scan()" log
6209 line; previously there was no indication of these.
6211 25. Added .include_if_exists.
6213 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6214 Linux. It turns out that not all file systems support this. Apparently some
6215 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6216 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6217 not supported on the file descriptor, is now ignored when Exim is trying to
6218 sync a directory. This applies only to Linux.
6220 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6222 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6223 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6226 29. Added support for maildirsize files from supplied patch (modified a bit).
6228 30. The use of :fail: followed by an empty string could lead Exim to respond to
6229 sender verification failures with (e.g.):
6231 550 Verification failed for <xxx>
6232 550 Sender verify failed
6234 where the first response line was missing the '-' that indicates it is not
6235 the final line of the response.
6237 31. The loop for finding the name of the user that called Exim had a hardwired
6238 limit of 10; it now uses the value of finduser_retries, which is used for
6239 all other user lookups.
6241 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6244 33. Exim was neglecting to zero errno before one call of strtol() when
6245 expanding a string and expecting an integer value. On some systems this
6246 resulted in spurious "integer overflow" errors. Also, it was casting the
6247 result into an int without checking.
6249 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6250 did not work. The code looks as if it has *never* worked, though it appears
6251 to have been documented since at least release 1.62. I have made it work.
6253 35. The "timeout_DNS" error in retry rules, also documented since at least
6254 1.62, also never worked. As it isn't clear exactly what this means, and
6255 clearly it isn't a major issue, I have abolished the feature by treating it
6256 as "timeout", and writing a warning to the main and panic logs.
6258 36. The display of retry rules for -brt wasn't always showing the error code
6261 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6262 timeout_connect_A, timeout_connect_MX.
6264 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6265 to the empty sender.
6267 39. The daemon was not analysing the content of -oX till after it had closed
6268 stderr and disconnected from the controlling terminal. This meant that any
6269 syntax errors were only noted on the panic log, and the return code from
6270 the command was 0. By re-arranging the code a little, I've made the
6271 decoding happen first, so such errors now appear on stderr, and the return
6272 code is 1. However, the actual setting up of the sockets still happens in
6273 the disconnected process, so errors there are still only recorded on the
6276 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6277 connections (as happens on some IP stacks) was logged at start up time as
6278 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6279 it from "IPv6 and IPv4", which means that two separate sockets are being
6282 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6283 says whether AF_INET or AF_INET6 was passed as an argument.
6285 42. Exiwhat output was messed up when time zones were included in log
6288 43. Exiwhat now gives more information about the daemon's listening ports,
6289 and whether -tls-on-connect was used.
6291 44. The "port" option of the smtp transport is now expanded.
6293 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6294 silently ignored. Now an error message is written to the main and panic
6297 46. There's a new ACL modifier called "logwrite" which writes to a log file
6298 as soon as it is encountered.
6300 47. Added $local_user_uid and $local_user_gid at routing time.
6302 48. Exim crashed when trying to verify a sender address that was being
6305 49. Exim was recognizing only a space character after ".include". It now also
6306 recognizes a tab character.
6308 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6309 extracting the relevant information from the specification. The man page no
6310 longer contains scrambled data for the -d option, and I've added a section
6311 at the front about calling Exim under different names.
6313 51. Added "extra_headers" argument to the "mail" command in filter files.
6315 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6318 53. Installed eximstats 1.29.
6320 54. Added transport_filter_timeout as a generic transport option.
6322 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6323 Cc: header lines. This was required by RFC 822, but it not required by RFC
6326 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6327 incoming messages that did not have them. Now it does so only if the
6328 message originates locally, that is, if there is no associated remote host
6329 address. When Resent- header lines are present, this applies to the Resent-
6330 lines rather than the non-Resent- lines.
6332 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6333 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6335 58. Messages for configuration errors now include the name of the main
6336 configuration file - useful now that there may be more than one file in a
6337 list (.included file names were always shown).
6339 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6340 for those rare installations that do not start the daemon as root or run it
6341 setuid root. I've cut out the call to initgroups() if the daemon is not
6344 60. The Exim user and group can now be bound into the binary as text strings
6345 that are looked up at the start of Exim's processing.
6347 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6349 62. Added $mailstore_basename variable.
6351 63. Installed patch to sieve.c from Michael Haardt.
6353 64. When Exim failed to open the panic log after failing to open the main log,
6354 the original message it was trying to log was written to stderr and debug
6355 output, but if they were not available (the usual case in production), it
6356 was lost. Now it is written to syslog before the two lines that record the
6357 failures to open the logs.
6359 65. Users' Exim filters run in subprocesses under the user's uid. It is
6360 possible for a "deliver" command or an alias in a "personal" command to
6361 provoke an address rewrite. If logging of address rewriting is configured,
6362 this fails because the process is not running as root or exim. There may be
6363 a better way of dealing with this, but for the moment (because 4.30 needs
6364 to be released), I have disabled address rewrite logging when running a
6365 filter in a non-root, non-exim process.
6371 1. The buildconfig auxiliary program wasn't quoting the value set for
6372 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6373 not defined. This bug was masked in 4.22 by the effect that was fixed in
6376 2. Some messages that were rejected after a message id was allocated were
6377 shown as "incomplete" by exigrep. It no longer does this for messages that
6378 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6380 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6381 have allow_domain_literals set, the ID did not get logged in the <= line.
6382 Domain literals are now always recognized in Message-ID: header lines.
6384 4. The first argument for a ${extract expansion item is the key name or field
6385 number. Leading and trailing spaces in this item were not being ignored,
6386 causing some misleading effects.
6388 5. When deliver_drop_privilege was set, single queue runner processes started
6389 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6390 same command in the process it spins off) were not dropping privilege.
6392 6. When the daemon running as "exim" started a queue runner, it always
6393 re-executed Exim in the spun-off process. This is a waste of effort when
6394 deliver_drop_privilege is set. The new process now just calls the
6395 queue-runner function directly.
6401 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6404 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6405 ignored. Though the use of -oP was forcing the writing of a pid file, it
6406 was always written to the default place.
6408 3. If the message "no IP address found for host xxxx" is generated during
6409 incoming verification, it is now followed by identification of the incoming
6410 connection (so you can more easily find what provoked it).
6412 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6414 5. Added some features to "harden" Exim a bit more against certain attacks:
6416 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6417 be put in Local/Makefile. This is like the never_users runtime option,
6418 but it cannot be overridden. The default setting is "root".
6420 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6421 prefix string with which any file named in a -C command line option
6424 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6425 is retained for -C and -D only if the caller of Exim is root. Without
6426 it, the exim user may also use -C and -D and retain privilege.
6428 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6429 command line option is disabled.
6431 6. Macro names set by the -D option must start with an upper case letter, just
6432 like macro names defined in the configuration file.
6434 7. Added "dereference=" facility to LDAP.
6436 8. Two instances of the typo "uknown" in the source files are fixed.
6438 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6439 the Configure-Makefile script screwed up while processing it.
6441 10. Incorporated PCRE 4.4.
6443 11. The SMTP synchronization check was not operating right at the start of an
6444 SMTP session. For example, it could not catch a HELO sent before the client
6445 waited for the greeting. There is now a check for outstanding input at the
6446 point when the greeting is written. Because of the duplex, asynchronous
6447 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6448 way, but not yet received, when the check is performed.
6450 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6451 on TCP/IP sockets, because this apparently causes some broken clients to
6454 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6455 unchanged) from the Cygwin maintainer.
6457 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6458 was supported instead of testing for mbx. Effectively a typo.
6460 15. The spa authenticator server code was not checking that the input it
6461 received was valid base64.
6463 16. The debug output line for the "set" modifier in ACLs was not showing the
6464 name of the variable that was being set.
6466 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6468 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6469 lookup on its own. Something else was needed to trigger the lookup. For
6470 example, a match in host_lookup or the need for a host name in a host list.
6471 Now, if $sender_host_name is referenced and the host name has not yet been
6472 looked up, a lookup is performed. If the lookup fails, the variable remains
6473 empty, and $host_lookup_failed is set to "1".
6475 19. Added "eqi" as a case-independent comparison operator.
6477 20. The saslauthd authentication condition could segfault if neither service
6478 nor realm was specified.
6480 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6481 error message that was logged was misleading, and incoming SMTP
6482 connections were dropped. The message is now more accurate, and temporary
6483 errors are given to SMTP connections.
6485 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6486 (except RSET) with a 421 error, until QUIT is received. However, it was
6487 failing to send a response to QUIT.
6489 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6490 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6491 ACL is now run after the helo_try_verify_hosts code.
6493 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6494 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6495 case-independent, so other case variants are also recognized). Apparently
6496 some systems use these upper case variants.
6498 25. If more than two messages were waiting for the same host, and a transport
6499 filter was specified for the transport, Exim sent two messages over the
6500 same TCP/IP connection, and then failed with "socket operation on non-
6501 socket" when it tried to send the third.
6503 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6505 27. The extern definition of crypt16() in expand.c was not being excluded when
6506 the OS had its own crypt16() function.
6508 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6509 as a preferred synonym for return_size_limit, both as an option and as an
6512 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6514 30. Changed the default configuration ACL to relax the local part checking rule
6515 for addresses that are not in any local domains. For these addresses,
6516 slashes and pipe symbols are allowed within local parts, but the sequence
6517 /../ is explicitly forbidden.
6519 31. SPA server authentication was not clearing the challenge buffer before
6522 32. log_message in a "warn" ACL statement was writing to the reject log as
6523 well as to the main log, which contradicts the documentation and doesn't
6524 seem right (because no rejection is happening). So I have stopped it.
6526 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6527 However, I am unable to do any testing of this.
6529 34. Fixed an infelicity in the appendfile transport. When checking directories
6530 for a mailbox, to see if any needed to be created, it was accidentally
6531 using path names with one or more superfluous leading slashes; tracing
6532 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6534 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6535 contents are added to the log line that is written for every discarded
6536 recipient. (Previously a log_message setting was ignored.)
6538 36. The ${quote: operator now quotes the string if it is empty.
6540 37. The install script runs exim in order to find its version number. If for
6541 some reason other than non-existence or emptiness, which it checks, it
6542 could not run './exim', it was installing it with an empty version number,
6543 i.e. as "exim-". This error state is now caught, and the installation is
6546 38. An argument was missing from the function that creates an error message
6547 when Exim fails to connect to the socket for saslauthd authentication.
6548 This could cause Exim to crash, or give a corrupted message.
6550 39. Added isip, isip4, and isip6 to ${if conditions.
6552 40. The ACL variables $acl_xx are now saved with the message, and can be
6553 accessed later in routers, transports, and filters.
6555 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6556 strings in the file are not string-expanded.
6558 42. If a MAIL command specified a SIZE value that was too large to fit into an
6559 int variable, the check against message_size_limit failed. Such values are
6560 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6561 day this will have to be increased, but I don't think I want to be around
6562 when emails are that large.
6569 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6570 iconv() is not standard in FreeBSD.
6572 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6573 IPv6 enabled. The observed symptom was a segmentation fault on return from
6574 the function os_common_find_running_interfaces() in src/os.c.
6576 3. In the check_special_case() function in daemon.c I had used "errno" as an
6577 argument name, which causes warnings on some systems. This was basically a
6578 typo, since it was named "eno" in the comments!
6580 4. The code that waits for the clock to tick (at a resolution of some fraction
6581 of a second) so as to ensure message-id uniqueness was always waiting for
6582 at least one whole tick, when it could have waited for less. [This is
6583 almost certainly not relevant at current processor speeds, where it is
6584 unlikely to ever wait at all. But we try to future-proof.]
6586 5. The function that sleeps for a time interval that includes fractions of a
6587 second contained a race. It did not block SIGALRM between setting the
6588 timer, and suspending (a couple of lines later). If the interval was short
6589 and the sigsuspend() was delayed until after it had expired, the suspension
6590 never ended. On busy systems this could lead to processes getting stuck for
6593 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6594 process, before it forks any delivery processes. The open lookup caching
6595 mechanism meant that the open file or database connection was passed into
6596 the delivery process. The problem was that delivery processes always tidy
6597 up cached lookup data. This could cause a problem for the next delivery
6598 process started by the queue runner, because the external queue runner
6599 process does not know about the closure. So the next delivery process
6600 still has data in the lookup cache. In the case of a file lookup, there was
6601 no problem because closing a file descriptor in a subprocess doesn't affect
6602 the parent. However, if the lookup was caching a connection to a database,
6603 the connection was closed, and the second delivery process was likely to
6604 see errors such as "PGSQL: query failed: server closed the connection
6605 unexpectedly". The problem has been fixed by closing all cached lookups
6606 in a queue runner before running a delivery process.
6608 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6609 seem to have the "const" qualifier which it has on other OS. I've
6612 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6613 *of the same type* (client or server) with the same public name that an
6614 error should be diagnosed.
6616 9. When Exim looked up a host name for an IP address, but failed to find the
6617 original IP address when looking up the host name (a safety check), it
6618 output the message "<ip address> does not match any IP for NULL", which was
6619 confusing, to say the least. The bug was that the host name should have
6620 appeared instead of "NULL".
6622 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6623 user that is built into the binary, and the -C or -D options is used, root
6624 privilege is dropped before the configuration file is read. In addition,
6625 logging is switched to stderr instead of the normal log files. If the
6626 configuration then re-defines the Exim user, the unprivileged environment
6627 is probably not what is expected, so Exim logs a panic warning message (but
6630 However, if deliver_drop_privilege is set, the unprivileged state may well
6631 be exactly what is intended, so the warning has been cut out in that case,
6632 and Exim is allowed to try to write to its normal log files.
6638 1. smtp_return_error_details was not giving details for temporary sender
6639 or receiver verification errors.
6641 2. Diagnose a configuration error if two authenticators have the same public
6644 3. Exim used not to create the message log file for a message until the first
6645 delivery attempt. This could be confusing when incoming messages were held
6646 for policy or load reasons. The message log file is now created at the time
6647 the message is received, and an initial "Received" line is written to it.
6649 4. The automatically generated man page for command line options had a minor
6650 bug that caused no ill effects; however, a more serious problem was that
6651 the procedure for building the man page automatically didn't always
6652 operate. Consequently, release 4.20 contains an out-of-date version. This
6653 shouldn't happen again.
6655 5. When building Exim with embedded Perl support, the script that builds the
6656 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6657 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6659 6. The freeze_tell option was not being used for messages that were frozen on
6660 arrival, either by an ACL or by local_scan().
6662 7. Added the smtp_incomplete_transaction log selector.
6664 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6665 was accepting AUTH without a new EHLO.
6667 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6668 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6671 10. Logging of TCP/IP connections (when configured) now happens in the main
6672 daemon process instead of the child process, so that the TCP/IP connection
6673 count is more accurate (but it can never be perfect).
6675 11. The use of "drop" in a nested ACL was not being handled correctly in the
6676 outer ACL. Now, if condition failure induced by the nested "drop" causes
6677 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6678 or "require"), the connection is dropped.
6680 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6681 that yield "discard" can now be used with an "accept" or a "discard" verb,
6682 but an error is generated for any others (because I can't see a useful way
6683 to define what should happen).
6685 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6686 are now processed in the same way as lines in the Exim configuration file.
6687 In particular, continuation lines are supported.
6689 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6691 15. Added -ti meaning -t -i.
6693 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6694 domains, and warn by logging if others are found.
6696 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6697 returned from GIFCONF when Exim is trying to find the list of interfaces on
6698 a host. The code in os.c has been modified to copy each ifreq to an aligned
6699 structure in all cases.
6701 Also, in some cases, the returned ifreq's were being copied to a 'struct
6702 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6703 means the last couple of bytes of an IPv6 address could be chopped if the
6704 ifreq contained only a normal sockaddr (14 bytes storage).
6706 18. Named domain lists were not supported in the hosts_treat_as_local option.
6707 An entry such as +xxxx was not recognized, and was treated as a literal
6710 19. Ensure that header lines added by a DATA ACL are included in the reject log
6711 if the ACL subsequently rejects the message.
6713 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6714 MD5 (which is deprecated).
6716 21. When testing a filter file using -bf, Exim was writing a message when it
6717 took the sender from a "From " line in the message, but it was not doing so
6718 when it took $return_path from a Return-Path: header line. It now does.
6720 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6721 with a valid header line field name (a series of printing characters
6722 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6724 23. Changed "disc" in the source to "disk" to conform to the documentation and
6725 the book and for uniformity.
6727 24. Ignore Sendmail's -Ooption=value command line item.
6729 25. When execve() failed while trying to run a command in a pipe transport,
6730 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6731 could be confused with a return value of 69 from the command itself. This
6732 has been changed to 127, the value the shell returns if it is asked to run
6733 a non-existent command. The wording for the related log line suggests a
6734 non-existent command as the problem.
6736 26. If received_header_text expands to an empty string, do not add a Received:
6737 header line to the message. (Well, it adds a token one on the spool, but
6738 marks it "old" so that it doesn't get used or transmitted.)
6740 27. Installed eximstats 1.28 (addition of -nt option).
6742 28. There was no check for failure on the call to getsockname() in the daemon
6743 code. This can fail if there is a shortage of resources on the system, with
6744 ENOMEM, for example. A temporary error is now given on failure.
6746 29. Contrary to the C standard, it seems that in some environments, the
6747 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6748 program. Exim now does this explicitly; it affects the formatting of
6749 timestamps using strftime().
6751 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6752 complaints. I've now initialized all the variables, to avoid this.
6754 32. Header lines added by a system filter were not being "seen" during
6755 transport-time rewrites.
6757 33. The info_callback() function passed to OpenSSL is set up with type void
6758 (*)(SSL *, int, int), as described somewhere. However, when calling the
6759 function (actually a macro) that sets it up, the type void(*)() is
6760 expected. I've put in a cast to prevent warnings from picky compilers.
6762 34. If a DNS black list lookup found a CNAME record, but there were no A
6763 records associated with the domain it pointed at, Exim crashed.
6765 35. If a DNS black list lookup returned more than one A record, Exim ignored
6766 all but the first. It now scans all returned addresses if a particular IP
6767 value is being sought. In this situation, the contents of the
6768 $dnslist_value variable are a list of all the addresses, separated by a
6771 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6772 to accept a host name and all its aliases if the forward lookup for any of
6773 them yielded the IP address of the incoming connection. Now it accepts only
6774 those names whose forward lookup yields the correct IP address. Any other
6775 names are discarded. This closes a loophole whereby a rogue DNS
6776 administrator could create reverse DNS records to break through a
6777 wildcarded host restriction in an ACL.
6779 37. If a user filter or a system filter that ran in a subprocess used any of
6780 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6781 the wrong values were passed to the pipe command ($thisaddress had the
6782 value of $0, $0 had the value of $1, etc). This bug was introduced by
6783 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6785 38. Improved the line breaking for long SMTP error messages from ACLs.
6786 Previously, if there was no break point between 40 and 75 characters, Exim
6787 left the rest of the message alone. Two changes have been made: (a) I've
6788 reduced the minimum length to 35 characters; (b) if it can't find a break
6789 point between 35 and 75 characters, it looks ahead and uses the first one
6790 that it finds. This may give the occasional overlong line, but at least the
6791 remaining text gets split now.
6793 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6794 file descriptors might be low, and that setting 1000 would always raise it.
6795 It turns out that in some environments, the limit is already over 1000 and
6796 that lowering it causes trouble. So now Exim takes care not to decrease it.
6798 40. When delivering a message, the value of $return_path is set to $sender_
6799 address at the start of routing (routers may change the value). By an
6800 oversight, this default was not being set up when an address was tested by
6801 -bt or -bv, which affected the outcome if any router or filter referred to
6804 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6805 log only when "message" or "log_message" are set. However, if one of the
6806 conditions was an address verification, or a call to a nested ACL, the
6807 messages generated by the underlying test were being passed through. This
6808 no longer happens. The underlying message is available in $acl_verify_
6809 message for both "message" and "log_message" expansions, so it can be
6810 passed through if needed.
6812 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6813 new expansion $bh_ to give the encoded byte string without charset
6814 translation. Translation happens only if iconv() is available; HAVE_ICONV
6815 indicates this at build time. HEADERS_CHARSET gives the charset to
6816 translate to; headers_charset can change it in the configuration, and
6817 "headers charset" can change it in an individual filter file.
6819 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6820 that creates RFC 2047 encoded "words" labels them as that charset instead
6821 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6822 expansion operator; (ii) when Exim creates a From: line for a local
6823 message; (iii) when a header line is rewritten to include a "phrase" part.
6825 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6826 buggy, causing it to skip the first lines of messages whose message ID
6827 ended in 'D'. This would not have bitten before Exim release 4.14, because
6828 message IDs were unlikely to end in 'D' before then. The effect was to have
6829 incorrect size information for certain domains.
6831 45. #include "config.h" was missing at the start of the crypt16.c module. This
6832 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6835 46. If there was a timeout during a "random" callout check, Exim treated it as
6836 a failure of the random address, and carried on sending RSET and the real
6837 address. If the delay was just some slowness somewhere, the response to the
6838 original RCPT would be taken as a response to RSET and so on, causing
6839 mayhem of various kinds.
6841 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6842 when I implemented it. It didn't allow for the fact that some option values
6843 may legitimately be negative (e.g. size_addition), and it didn't even do
6844 the right test for positive values.
6846 48. Domain names in DNS records are case-independent. Exim always looks them up
6847 in lower case. Some resolvers return domain names in exactly the case they
6848 appear in the zone file, that is, they may contain uppercase letters. Not
6849 all resolvers do this - some return always lower case. Exim was treating a
6850 change of case by a resolver as a change of domain, similar to a widening
6851 of a domain abbreviation. This triggered its re-routing code and so it was
6852 trying to route what was effectively the same domain again. This normally
6853 caused routing to fail (because the router wouldn't handle the domain
6854 twice). Now Exim checks for this case specially, and just changes the
6855 casing of the domain that it ultimately uses when it transmits the message
6858 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6861 50. If a filter generated a file delivery with a non-absolute name (possible if
6862 no home directory exists for the router), the forbid_file option was not
6865 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6866 the existing equality matching.
6868 52. Exim was using ints instead of ino_t variables in some places where it was
6869 dealing with inode numbers.
6871 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6872 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6873 called TMPDIR, and if it finds it is different, it changes its value.
6875 54. The smtp_printf() function is now made available to local_scan() so
6876 additional output lines can be written before returning. There is also an
6877 smtp_fflush() function to enable the detection of a dropped connection.
6878 The variables smtp_input and smtp_batched_input are exported to
6881 55. Changed the default runtime configuration: the message "Unknown user"
6882 has been removed from the ACL, and instead placed on the localuser router,
6883 using the cannot_route_message feature. This means that any verification
6884 failures that generate their own messages won't get overridden. Similarly,
6885 the "Unrouteable address" message that was in the ACL for unverifiable
6886 relay addresses has also been removed.
6888 56. Added hosts_avoid_esmtp to the smtp transport.
6890 57. The exicyclog script was not checking for the esoteric option
6891 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6892 will work only if exicyclog is run under the appropriate euid.
6894 58. Following a discussion on the list, the rules by which Exim recognises line
6895 endings on incoming messages have been changed. The -dropcr and drop_cr
6896 options are now no-ops, retained only for backwards compatibility. The
6897 following line terminators are recognized: LF CRLF CR. However, special
6898 processing applies to CR:
6900 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6901 nor a local message in the state where . is a terminator.
6903 (ii) If a bare CR is encountered in a header line, an extra space is added
6904 after the line terminator so as not to end the header. The reasoning
6905 behind this is that bare CRs in header lines are most likely either
6906 to be mistakes, or people trying to play silly games.
6908 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6909 was being incorrectly given as 18 bytes larger than it should have been.
6910 This is a VOB (very old bug).
6912 60. This may never have affected anything current, but just in case it has:
6913 When the local host is found other than at the start of a list of hosts,
6914 the local host, those with the same MX, and any that follow, are discarded.
6915 When the list in question was part of a longer list of hosts, the following
6916 hosts (not currently being processed) were also being discarded. This no
6917 longer happens. I'm not sure if this situation could ever has previously
6920 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6923 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6924 header. This is something that is recommended in a new Internet Draft, and
6925 is something that is documented as being done by Sendmail. There are two
6926 possible values. For messages generated by the autoreply transport, Exim
6929 Auto-Submitted: auto-replied
6931 whereas for all other generated messages (e.g. bounces) it adds
6933 Auto-Submitted: auto-generated
6935 63. The "personal" condition in filters now includes a test for the
6936 Auto-Submitted: header. If it contains the string "auto-" the message it
6937 not considered personal.
6939 64. Added rcpt_include_affixes as a generic transport option.
6941 65. Added queue_only_override (default true).
6943 66. Added the syslog_duplication option.
6945 67. If what should have been the first header line of a message consisted of
6946 a space followed by a colon, Exim was mis-interpreting it as a header line.
6947 It isn't of course - it is syntactically invalid and should therefore be
6948 treated as the start of the message body. The misbehaviour could have
6949 caused a number of strange effects, including loss of data in subsequent
6950 header lines, and spool format errors.
6952 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6953 client host had authenticated. This control can now be exercised by an ACL
6954 for more flexibility.
6956 69. By default, callouts do not happen when testing with -bh. There is now a
6957 variant, -bhc, which does actually run the callout code, including
6958 consulting and updating the callout cache.
6960 70. Added support for saslauthd authentication, courtesy of Alexander
6963 71. If statvfs() failed on the spool or log directories while checking their
6964 size for availability, Exim confusingly gave the error "space shortage".
6965 Furthermore, in debugging mode it crashed with a floating point exception.
6966 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6967 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6968 serious problem, Exim now writes to the main and panic logs when this
6969 happens, with details of the failure. It then refuses to accept the
6970 incoming message, giving the message "spool directory problem" or "log
6971 directory problem" with a 421 code for SMTP messages.
6973 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6974 0, 1, and 2 exist, because some OS complain for execs without them (see
6975 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6976 descriptors. However, the code omitted to check that the open succeeded,
6977 causing mysterious errors if for some reason the permissions on /dev/null
6978 got screwed. Now Exim writes a message to the main and panic logs, and
6979 bombs out if it can't open /dev/null.
6981 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6982 interact so that it is all more flexible. It is supposed to remain
6983 backwards compatible. Also added extra_local_interfaces.
6985 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6986 to bomb out with an assertion failure - to the client this appears as a
6987 connection drop. This problem occurs in the part of the code that was taken
6988 from the Samba project. Fortunately, the assertion is in a very simple
6989 function, so I have fixed this by reproducing the function inline in the
6990 one place where it is called, and arranging for authentication to fail
6991 instead of killing the process with assert().
6993 75. The SPA client code was not working when the server requested OEM rather
6994 than Unicode encoding.
6996 76. Added code to make require_files with a specific uid setting more usable in
6997 the case where statting the file as root fails - usually a non-root-mounted
6998 NFS file system. When this happens and the failure is EACCES, Exim now
6999 forks a subprocess and does the per-uid checking as the relevant uid.
7001 77. Added process_log_path.
7003 78. If log_file_path was not explicitly set, a setting of check_log_space or
7004 check_log_inodes was ignored.
7006 79. If a space check for the spool or log partitions fails, the incident is now
7007 logged. Of course, in the latter case the data may get lost...
7009 80. Added the %p formatting code to string_format() so that it can be used to
7010 print addresses in debug_print(). Adjusted all the address printing in the
7011 debugging in store.c to use %p rather than %d.
7013 81. There was a concern that a line of code in smtp_in.c could overflow a
7014 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7015 initially expressed, the concern was not well-founded, because trailing
7016 spaces are removed early. However, if the trailing spaces were followed by
7017 a NULL, they did not get removed, so the overflow was possible. Two fixes
7020 (a) I re-wrote the offending code in a cleaner fashion.
7021 (b) If an incoming SMTP command contains a NULL character, it is rejected
7024 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7025 runs initgroups(), so that if the Exim user is in any additional groups,
7026 they will be used during message reception.
7032 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob