1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Rework error string handling in TLS interface so that the caller in
12 more cases is responsible for logging. This permits library-sourced
13 string to be attached to addresses during delivery, and collapses
14 pairs of long lines into single ones.
16 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
17 during configuration. Wildcards are allowed and expanded.
19 JH/02 Rework error string handling in DKIM to pass more info back to callers.
20 This permits better logging.
22 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
23 do not close it down and have the child transport start it up again on
24 the passed-on TCP connection. Instead, proxy the child (and any
25 subsequent ones) for TLS via a unix-domain socket channel. Logging is
26 affected: the continued delivery log lines do not have any DNSSEC, TLS
27 Certificate or OCSP information. TLS cipher information is still logged.
29 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
30 identical IP addresses on different listening ports. Will also affect
33 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
34 add noisy ifdef guards to special-case this sillyness.
35 Patch from Bernd Kuhls.
37 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
38 than 255 are no longer allowed.
40 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
41 Disable the session-cache too, which might reduce our load. Since we
42 currrectly use a new context for every connection, both as server and
43 client, there is no benefit for these.
44 GnuTLS appears to not support tickets server-side by default (we don't
45 call gnutls_session_ticket_enable_server()) but client side is enabled
46 by default on recent versions (3.1.3 +) unless the PFS priority string
49 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
50 <https://reproducible-builds.org/specs/source-date-epoch/>.
52 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
53 the check for any unsuccessful recipients did not notice the limit, and
54 erroneously found still-pending ones.
56 JH/08 Pipeline CHUNKING command and data together, on kernels that support
57 MSG_MORE. Only in-clear (not on TLS connections).
59 JH/09 Avoid using a temporary file during transport using dkim. Unless a
60 transport-filter is involved we can buffer the headers in memory for
61 creating the signature, and read the spool data file once for the
62 signature and again for transmission.
64 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
65 4.77 as the kernel support then wasn't solid, having issues in 64bit
66 mode. Now, it's been long enough. Add support for FreeBSD also.
68 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
69 case where the routing stage had gathered several addresses to send to
70 a host before calling the transport for the first, we previously failed
71 to close down TLS in the old transport process before passing the TCP
72 connection to the new process. The new one sent a STARTTLS command
73 which naturally failed, giving a failed delivery and bloating the retry
74 database. Investigation and fix prototype from Wolfgang Breyha.
76 JH/12 Fix check on SMTP command input synchronisation. Previously there were
77 false-negatives in the check that the sender had not preempted a response
78 or prompt from Exim (running as a server), due to that code's lack of
79 awareness of the SMTP input buffering.
81 PP/04 Add commandline_checks_require_admin option.
82 Exim drops privileges sanely, various checks such as -be aren't a
83 security problem, as long as you trust local users with access to their
84 own account. When invoked by services which pass untrusted data to
85 Exim, this might be an issue. Set this option in main configuration
86 AND make fixes to the calling application, such as using `--` to stop
89 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
90 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
91 responses to those, into a single TLS record each way (this usually means
92 a single packet). As a side issue, smtp_enforce_sync now works on TLS
95 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
96 affects you only if you're dancing at the edge of the param size limits.
97 If you are, and this message makes sense to you, then: raise the
98 configured limit or use OpenSSL 1.1. Nothing we can do for older
101 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
102 cmdline element to get the documented default one. Previously it was
105 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
106 are unsuable for non-message-specific reasons - eg. connection timeout,
107 banner-time rejection.
109 JH/16 Fix logging of delivery remote port, when specified by router, under
112 PP/06 Repair manualroute's ability to take options in any order, even if one
113 is the name of a transport.
116 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
118 JH/17 Change the list-building routines interface to use the expanding-string
119 triplet model, for better allocation and copying behaviour.
121 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
122 Previously it was constructed the first time a possibly-matching string
123 was met in the configuration file input during startup; now it is done
126 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
127 compatible one, to avoid the (poorly documented) possibility of a config
128 file in the working directory redirecting the DB files, possibly correpting
129 some existing file. CVE-2017-10140 assigned for BDB.
131 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
132 cache-hot. Previously, although the result was properly cached, the
133 initial verify call returned a defer.
135 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
136 the main verify for receipient in uncached-mode.
138 JH/22 Retire historical build files to an "unsupported" subdir. These are
139 defined as "ones for which we have no current evidence of testing".
141 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
142 if present. Previously it was ignored.
144 JH/24 Start using specified-initialisers in C structure init coding. This is
145 a C99 feature (it's 2017, so now considered safe).
151 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
152 than -2003 did; needs libidn2 in addition to libidn.
154 JH/02 The path option on a pipe transport is now expanded before use.
156 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
157 Patch provided by "Björn", documentation fix added too.
159 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
160 missing a wire-to-host endian conversion.
162 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
163 close after a BDAT command line could be taken as a following command,
164 giving a synch failure. Fix by only checking for synch immediately
165 before acknowledging the chunk.
167 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
168 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
169 macro. Patches provided by Josh Soref.
171 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
172 Previously we did not; the RFC seems ambiguous and VRFY is not listed
173 by IANA as a service extension. However, John Klensin suggests that we
176 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
177 the dkim code may be unix-mode line endings rather than smtp wire-format
178 CRLF, so prepend a CR to any bare LF.
180 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
181 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
183 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
184 the first were themselves being wrongly included in the feed into dkim
185 processing; with most chunk sizes in use this resulted in an incorrect
186 body hash calculated value.
188 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
189 DKIM signature block, for verification. Although advised against by
190 standards it is specifically not ruled illegal.
192 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
194 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
195 missing a body hash (the bh= tag).
197 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
198 It seems that HAProxy sends the Proxy Protocol information in clear and
199 only then does a TLS startup, so do the same.
201 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
202 TCP connections (such as for Spamd) unless the daemon successfully set
203 Fast Open mode on its listening sockets. This fixes breakage seen on
204 too-old kernels or those not configured for Fast Open, at the cost of
205 requiring both directions being enabled for TFO, and TFO never being used
206 by non-daemon-related Exim processes.
208 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
209 endings, at least on the first header line. Try to canonify any that get
210 past that check, despite the cost.
212 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
213 now limited to an arbitrary five deep, while parsing addresses with the
214 strip_excess_angle_brackets option enabled.
216 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
217 instead leave the unprompted TLS handshake in socket buffer for the
218 TLS library to consume.
220 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
222 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
224 JH/16 Drop variables when they go out of scope. Memory management drops a whole
225 region in one operation, for speed, and this leaves assigned pointers
226 dangling. Add checks run only under the testsuite which checks all
227 variables at a store-reset and panics on a dangling pointer; add code
228 explicitly nulling out all the variables discovered. Fixes one known
229 bug: a transport crash, where a dangling pointer for $sending_ip_address
230 originally assigned in a verify callout, is re-used.
232 PP/06 Drop '.' from @INC in various Perl scripts.
234 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
236 PP/08 Reduce a number of compilation warnings under clang; building with
237 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
238 should be warning-free.
240 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
242 HS/01 Fix portability problems introduced by PP/08 for platforms where
243 realloc(NULL) is not equivalent to malloc() [SunOS et al].
245 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
246 chunk. This allows us to accept broken chunked messages. We need a more
247 general solution here.
249 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
250 already-broken messages in the queue.
252 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
254 JH/19 Fix reference counting bug in routing-generated-address tracking.
260 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
261 supports it and a size is available (ie. the sending peer gave us one).
263 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
264 years of being deprecated). The replacements are the ACLs
265 acl_smtp_mime and acl_not_smtp_mime.
267 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
268 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
269 or even an in-clear connection were permitted. Now, if the host lookup
270 was dnssec and dane was requested then the host is only used if the
271 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
272 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
273 if one fails this test.
274 This means that a poorly-configured remote DNS will make it incommunicado;
275 but it protects against a DNS-interception attack on it.
277 JH/04 Bug 1810: make continued-use of an open smtp transport connection
278 non-noisy when a race steals the message being considered.
280 JH/05 If main configuration option tls_certificate is unset, generate a
281 self-signed certificate for inbound TLS connections.
283 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
284 in rewrites and routers.
286 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
287 and logged a warning sing 4.83; now they are a configuration file error.
289 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
290 (lacking @domain). Apply the same qualification processing as RCPT.
292 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
294 JH/10 Support ${sha256:} applied to a string (as well as the previous
297 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
298 a cutthrough deliver is pending, as we always want to make a connection.
299 This also avoids re-routing the message when later placing the cutthrough
300 connection after a verify cache hit.
301 Do not update it with the verify result either.
303 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
304 when routing results in more than one destination address.
306 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
307 signing (which inhibits the cutthrough capability). Previously only
308 the presence of an option was tested; now an expansion evaluating as
309 empty is permissible (obviously it should depend only on data available
310 when the cutthrough connection is made).
312 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
313 the relevant preceding SMTP command did not note the pipelining mode.
315 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
316 Previously they were not counted.
318 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
319 as one having no matching records. Previously we deferred the message
320 that needed the lookup.
322 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
323 distinguished as "(=".
325 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
326 for missing MX records. Previously it only worked for missing A records.
328 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
330 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
331 after the data-go-ahead and data-ack. Patch from Jason Betts.
333 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
334 even for a "none" policy. Patch from Tony Meyer.
336 JH/22 Fix continued use of a connection for further deliveries. If a port was
337 specified by a router, it must also match for the delivery to be
340 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
341 When one of the recipients of a message was unsuitable for the connection
342 (has no matching addresses), we lost track of needing to mark it
343 deferred. As a result mail would be lost.
345 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
347 JH/25 Decoding ACL controls is now done using a binary search; the source code
348 takes up less space and should be simpler to maintain. Merge the ACL
349 condition decode tables also, with similar effect.
351 JH/26 Fix problem with one_time used on a redirect router which returned the
352 parent address unchanged. A retry would see the parent address marked as
353 delivered, so not attempt the (identical) child. As a result mail would
356 JH/27 Fix a possible security hole, wherein a process operating with the Exim
357 UID can gain a root shell. Credit to http://www.halfdog.net/ for
358 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
361 JH/28 Enable {spool,log} filesystem space and inode checks as default.
362 Main config options check_{log,spool}_{inodes,space} are now
363 100 inodes, 10MB unless set otherwise in the configuration.
365 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
366 Previously it only applied to the main-section connection policy
369 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
371 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
372 by me. Added RFC7919 DH primes as an alternative.
374 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
375 are not in the system include path.
377 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
378 GnuTLS, when a session startup failed (eg because the client disconnected)
379 Exim did stdio operations after fclose. This was exposed by a recent
380 change which nulled out the file handle after the fclose.
382 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
383 signed directly by the cert-signing cert, rather than an intermediate
384 OCSP-signing cert. This is the model used by LetsEncrypt.
386 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
388 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
389 an incoming connection.
391 HS/02 Bug 1802: Do not half-close the connection after sending a request
394 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
395 fallback to "prime256v1".
397 JH/34 SECURITY: Use proper copy of DATA command in error message.
398 Could leak key material. Remotely exploitable. CVE-2016-9963.
404 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
405 and 3.4.4 - once the server is enabled to respond to an OCSP request
406 it does even when not requested, resulting in a stapling non-aware
407 client dropping the TLS connection.
409 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
410 support variable-length bit vectors. No functional change.
412 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
413 The I= interface field on outgoing lines is now after the H= remote
414 host field, same as incoming lines. There is a separate
415 outgoing_interface log selector which allows you to disable the
418 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
419 If not running log_selector +smtp_connection the mainlog would be held
420 open indefinitely after a "too many connections" event, including to a
421 deleted file after a log rotate. Leave the per net connection logging
422 leaving it open for efficiency as that will be quickly detected by the
423 check on the next write.
425 HS/01 Bug 1671: Fix post transport crash.
426 Processing the wait-<transport> messages could crash the delivery
427 process if the message IDs didn't exist for some reason. When
428 using 'split_spool_directory=yes' the construction of the spool
429 file name failed already, exposing the same netto behaviour.
431 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
432 mime_regex ACL conditions.
434 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
435 to DSN fail messages (bounces): remote IP, remote greeting, remote response
436 to HELO, local diagnostic string.
438 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
439 log line to debug. Even when configured with a tls authenticator many
440 client connections are expected to not authenticate in this way, so
441 an authenticate fail is not an error.
443 HS/02 Add the Exim version string to the process info. This way exiwhat
444 gives some more detail about the running daemon.
446 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
447 matter for fast-change records such as DNSBLs.
449 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
450 retry record, even if constant. There may be multiple transports with
451 different interface settings and the retry behaviour needs to be kept
454 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
456 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
458 JH/10 Bug 840: fix log_defer_output option of pipe transport
460 JH/11 Bug 830: use same host for all RCPTS of a message, even under
461 hosts_randomize. This matters a lot when combined with mua_wrapper.
463 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
464 ${quote_pgsql:<string>} operator.
466 JH/13 Bug 1708: avoid misaligned access in cached lookup.
468 JH/14 Change header file name for freeradius-client. Relevant if compiling
469 with Radius support; from the Gentoo tree and checked under Fedora.
471 JH/15 Bug 1712: Introduce $prdr_requested flag variable
473 JH/16 Bug 1714: Permit an empty string as expansion result for transport
474 option transport_filter, meaning no filtering.
476 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
478 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
479 defaults to "*" (all hosts). The variable is now available when not built
480 with TLS, default unset, mainly to enable keeping the testsuite sane.
481 If a server certificate is not supplied (via tls_certificate) an error is
482 logged, and clients will find TLS connections fail on startup. Presumably
483 they will retry in-clear.
484 Packagers of Exim are strongly encouraged to create a server certificate
485 at installation time.
487 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
488 with the $config_file variable.
490 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
491 in transport context, after the attempt, and per-recipient. The latter type
492 is per host attempted. The event data is the error message, and the errno
493 information encodes the lookup type (A vs. MX) used for the (first) host,
494 and the trailing two digits of the smtp 4xx response.
496 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
497 to write to mainlog (or rejectlog, paniclog) in the window between file
498 creation and permissions/ownership being changed. Particularly affects
499 installations where exicyclog is run as root, rather than exim user;
500 result is that the running daemon panics and dies.
502 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
504 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
505 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
506 "pri" and "weight". Note that the previous implicit priority given by the
507 list order is no longer honoured.
509 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
512 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
513 by defining SUPPORT_SOCKS.
515 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
516 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
517 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
518 variables are renamed to proxy_{local,external}_{address,port}.
520 JH/27 Move Internationalisation support from Experimental to mainline, enabled
521 for a build by defining SUPPORT_I18N
523 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
524 of the query string, and make ${quote_redis:} do that quoting.
526 JH/29 Move Events support from Experimental to mainline, enabled by default
527 and removable for a build by defining DISABLE_EVENT.
529 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
531 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
532 cached by the daemon.
534 JH/32 Move Redis support from Experimental to mainline, enabled for a build
535 by defining LOOKUP_REDIS. The libhiredis library is required.
537 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
538 keys are given for lookup.
540 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
541 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
542 only supported when built with TLS support. The PolarSSL SHA routines
543 are still used when the TLS library is too old for convenient support.
545 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
546 openssl_options), for security. OpenSSL forces this from version 1.1.0
547 server-side so match that on older versions.
549 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
550 allocation for $value could be released as the expansion processing
551 concluded, but leaving the global pointer active for it.
553 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
554 and to use the domains and local_parts ACL conditions.
556 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
557 incorrectly not doubled on cutthrough transmission, hence seen as a
558 body-termination at the receiving system - resulting in truncated mails.
559 Commonly the sender saw a TCP-level error, and retransmitted the message
560 via the normal store-and-forward channel. This could result in duplicates
561 received - but deduplicating mailstores were liable to retain only the
562 initial truncated version.
564 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
566 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
568 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
569 we're in there, support oversigning also; bug 1309.
571 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
573 HS/04 Add support for keep_environment and add_environment options.
575 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
576 either intentional arithmetic overflow during PRNG, or testing config-
579 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
580 delivery resulted in actual delivery. Cancel cutthrough before DATA
583 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
584 rejects a recipient: pass the reject to the originator.
586 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
587 Many were false-positives and ignorable, but it's worth fixing the
590 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
591 for the new environment-manipulation done at startup. Move the routines
592 from being local to tls.c to being global via the os.c file.
594 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
595 an extract embedded as result-arg for a map, the first arg for extract
596 is unavailable so we cannot tell if this is a numbered or keyed
597 extraction. Accept either.
603 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
606 JH/02 The smtp transport option "multi_domain" is now expanded.
608 JH/03 The smtp transport now requests PRDR by default, if the server offers
611 JH/04 Certificate name checking on server certificates, when exim is a client,
612 is now done by default. The transport option tls_verify_cert_hostnames
613 can be used to disable this per-host. The build option
614 EXPERIMENTAL_CERTNAMES is withdrawn.
616 JH/05 The value of the tls_verify_certificates smtp transport and main options
617 default to the word "system" to access the system default CA bundle.
618 For GnuTLS, only version 3.0.20 or later.
620 JH/06 Verification of the server certificate for a TLS connection is now tried
621 (but not required) by default. The verification status is now logged by
622 default, for both outbound TLS and client-certificate supplying inbound
625 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
628 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
629 Status Notification (bounce) messages are now MIME format per RFC 3464.
630 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
631 under the control of the dsn_advertise_hosts option, and routers may
632 have a dsn_lasthop option.
634 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
635 default, modifiable by a malware= option. The list separator for
636 the options can now be changed in the usual way. Bug 68.
638 JH/10 The smtp_receive_timeout main option is now expanded before use.
640 JH/11 The incoming_interface log option now also enables logging of the
641 local interface on delivery outgoing connections.
643 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
644 if the interface and destination host and port all match.
646 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
649 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
650 Patch from Andrew Lewis.
652 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
653 now supports optional time-restrictions, weighting, and priority
654 modifiers per server. Patch originally by <rommer@active.by>.
656 JH/16 The spamd_address main option now supports a mixed list of local
657 and remote servers. Remote servers can be IPv6 addresses, and
658 specify a port-range.
660 JH/17 Bug 68: The spamd_address main option now supports an optional
661 timeout value per server.
663 JH/18 Bug 1581: Router and transport options headers_add/remove can
664 now have the list separator specified.
666 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
669 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
672 JH/21 Support for the A6 type of dns record is withdrawn.
674 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
675 rather than the verbs used.
677 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
678 from 255 to 1024 chars.
680 JH/24 Verification callouts now attempt to use TLS by default.
682 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
683 are generic router options now. The defaults didn't change.
685 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
686 Original patch from Alexander Shikoff, worked over by JH.
688 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
689 files if ZCAT_COMMAND is not executable.
691 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
693 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
695 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
696 Normally benign, it bites when the pair was led to by a CNAME;
697 modern usage is to not canonicalize the domain to a CNAME target
698 (and we were inconsistent anyway for A-only vs AAAA+A).
700 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
702 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
703 when evaluating $sender_host_dnssec.
705 JH/31 Check the HELO verification lookup for DNSSEC, adding new
706 $sender_helo_dnssec variable.
708 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
710 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
712 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
714 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
715 documented as working, but never had. Support all but $spam_report.
717 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
718 added for tls authenticator.
720 HS/03 Add perl_taintmode main config option
726 TL/01 When running the test suite, the README says that variables such as
727 no_msglog_check are global and can be placed anywhere in a specific
728 test's script, however it was observed that placement needed to be near
729 the beginning for it to behave that way. Changed the runtest perl
730 script to read through the entire script once to detect and set these
731 variables, reset to the beginning of the script, and then run through
732 the script parsing/test process like normal.
734 TL/02 The BSD's have an arc4random API. One of the functions to induce
735 adding randomness was arc4random_stir(), but it has been removed in
736 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
737 function when detected.
739 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
740 cause callback expansion.
742 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
743 syntax errors in an expansion can be treated as a string instead of
744 logging or causing an error, due to the internal use of bool_lax
745 instead of bool when processing it.
747 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
748 server certificates when making smtp deliveries.
750 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
752 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
754 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
756 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
757 Merged patch from Sebastian Wiedenroth.
759 JH/05 Fix results-pipe from transport process. Several recipients, combined
760 with certificate use, exposed issues where response data items split
761 over buffer boundaries were not parsed properly. This eventually
762 resulted in duplicates being sent. This issue only became common enough
763 to notice due to the introduction of connection certificate information,
764 the item size being so much larger. Found and fixed by Wolfgang Breyha.
766 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
767 size buffer was used, resulting in syntax errors when an expansion
770 JH/07 Add support for directories of certificates when compiled with a GnuTLS
771 version 3.3.6 or later.
773 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
774 is EXPERIMENTAL_EVENT, the main-configuration and transport options
775 both become "event_action", the variables become $event_name, $event_data
776 and $event_defer_errno. There is a new variable $verify_mode, usable in
777 routers, transports and related events. The tls:cert event is now also
778 raised for inbound connections, if the main configuration event_action
781 TL/06 In test suite, disable OCSP for old versions of openssl which contained
782 early OCSP support, but no stapling (appears to be less than 1.0.0).
784 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
785 server certificate names available under the smtp transport option
786 "tls_verify_cert_hostname" now do not permit multi-component wildcard
789 JH/10 Time-related extraction expansions from certificates now use the main
790 option "timezone" setting for output formatting, and are consistent
791 between OpenSSL and GnuTLS compilations. Bug 1541.
793 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
794 encoded parameter in the incoming message. Bug 1558.
796 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
797 include certificate info, eximon was claiming there were spoolfile
800 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
802 JH/14 Log delivery-related information more consistently, using the sequence
803 "H=<name> [<ip>]" wherever possible.
805 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
806 are problematic for Debian distribution, omit them from the release
809 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
811 JH/16 Fix string representation of time values on 64bit time_t architectures.
814 JH/17 Fix a null-indirection in certextract expansions when a nondefault
815 output list separator was used.
820 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
821 checkers that were complaining about end of non-void function with no
824 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
825 This was a regression introduced in 4.83 by another bugfix.
827 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
829 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
830 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
836 TF/01 Correctly close the server side of TLS when forking for delivery.
838 When a message was received over SMTP with TLS, Exim failed to clear up
839 the incoming connection properly after forking off the child process to
840 deliver the message. In some situations the subsequent outgoing
841 delivery connection happened to have the same fd number as the incoming
842 connection previously had. Exim would try to use TLS and fail, logging
843 a "Bad file descriptor" error.
845 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
846 utilities have not been installed.
848 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
849 temporary space as the ACL may create new global variables.
851 TL/01 LDAP support uses per connection or global context settings, depending
852 upon the detected version of the libraries at build time.
854 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
855 to extract and use the src ip:port in logging and expansions as if it
856 were a direct connection from the outside internet. PPv2 support was
857 updated based on HAProxy spec change in May 2014.
859 JH/02 Add ${listextract {number}{list}{success}{fail}}.
861 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
862 Properly escape header and check for NULL return.
864 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
867 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
869 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
870 characters in header names, implemented as a verify condition.
871 Contributed by Michael Fischer v. Mollard.
873 TL/05 Rename SPF condition results err_perm and err_temp to standardized
874 results permerror and temperror. Previous values are deprecated but
875 still accepted. In a future release, err_perm and err_temp will be
876 completely removed, which will be a backward incompatibility if the
877 ACL tests for either of these two old results. Patch contributed by
878 user bes-internal on the mailing list.
880 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
882 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
883 selectors, in both main and reject logs.
885 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
888 JH/07 Add malware type "sock" for talking to simple daemon.
890 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
892 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
893 routers/transports under cutthrough routing.
895 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
896 numbers. Touch up "bool" conditional to keep the same definition.
898 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
900 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
902 JH/12 Expand items in router/transport headers_add or headers_remove lists
903 individually rather than the list as a whole. Bug 1452.
905 Required for reasonable handling of multiple headers_ options when
906 they may be empty; requires that headers_remove items with embedded
907 colons must have them doubled (or the list-separator changed).
909 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
910 view the policy declared in the DMARC record. Currently, $dmarc_status
911 is a combined value of both the record presence and the result of the
914 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
916 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
917 dnslookup router and the smtp transport (applying to the forward
920 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
921 of ldap servers used for a specific lookup. Patch provided by Heiko
924 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
925 New variable $lookup_dnssec_authenticated for observability.
927 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
928 Patch submitted by Lars Timman.
930 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
932 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
933 Requires trusted mode and valid format message id, aborts otherwise.
934 Patch contributed by Heiko Schlichting.
936 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
937 certextract with support for various fields. Bug 1358.
939 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
940 is requested by default, modifiable by smtp transport option
943 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
944 operate on certificate variables to give certificate fingerprints
945 Also new ${sha256:cert_variable}.
947 JH/23 The PRDR feature is moved from being Experimental into the mainline.
949 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
950 Christian Aistleitner.
952 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
954 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
955 file. Patch from Wolfgang Breyha.
957 JH/25 Expand the coverage of the delivery $host and $host_address to
958 client authenticators run in verify callout. Bug 1476.
960 JH/26 Port service names are now accepted for tls_on_connect_ports, to
961 align with daemon_smtp_ports. Bug 72.
963 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
964 support and error reporting did not work properly.
966 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
967 and is readable. Patch from Andrew Colin Kissa.
969 TL/14 Enhance documentation of ${run expansion and how it parses the
970 commandline after expansion, particularly in the case when an
971 unquoted variable expansion results in an empty value.
973 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
975 PP/02 Fix internal collision of T_APL on systems which support RFC3123
976 by renaming away from it. Addresses GH issue 15, reported by
979 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
981 TL/15 SECURITY: prevent double expansion in math comparison functions
982 (can expand unsanitized data). Not remotely exploitable.
989 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
991 PP/02 Make -n do something, by making it not do something.
992 When combined with -bP, the name of an option is not output.
994 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
997 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
998 $sender_host_name and config options to manage this, and basic check
1001 PP/05 DSCP support for outbound connections and control modifier for inbound.
1003 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1004 (Only plugin which currently uses this is kerberos4, which nobody should
1005 be using, but we should make it available and other future plugins might
1006 conceivably use it, even though it would break NAT; stuff *should* be
1007 using channel bindings instead).
1009 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1010 name; added for Sendmail compatibility; requires admin caller.
1011 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1012 just ignore it); requires trusted caller.
1013 Also parse but ignore: -Ac -Am -X<logfile>
1016 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1018 TL/02 Add +smtp_confirmation as a default logging option.
1020 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1021 Patch by Magnus Holmgren from 2007-02-20.
1023 TL/04 Bugzilla 1281 - Spec typo.
1024 Bugzilla 1283 - Spec typo.
1025 Bugzilla 1290 - Spec grammar fixes.
1027 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1029 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1031 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1032 mailing list by by Dmitry Isaikin.
1034 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1036 JH/02 Support "G" suffix to numbers in ${if comparisons.
1038 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1040 NM/01 Bugzilla 1197 - Spec typo
1041 Bugzilla 1196 - Spec examples corrections
1043 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1045 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1046 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1049 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1050 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1052 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1053 "acl {{name}{arg}...}", and optional args on acl condition
1056 JH/05 Permit multiple router/transport headers_add/remove lines.
1058 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1060 JH/07 Avoid using a waiting database for a single-message-only transport.
1061 Performance patch from Paul Fisher. Bugzilla 1262.
1063 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1066 JH/09 Add $headers_added variable, with content from use of ACL modifier
1067 add_header (but not yet added to the message). Bugzilla 199.
1069 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1070 Pulled from Bugzilla 817 by Wolfgang Breyha.
1072 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1074 (nb: this is the same fix as in Exim 4.80.1)
1076 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1079 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1082 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1084 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1085 advertise SMTP AUTH mechanism to us, instead of a generic
1086 protocol violation error. Also, make Exim more robust to bad
1087 data from the Dovecot auth socket.
1089 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1091 When a queue runner is handling a message, Exim first routes the
1092 recipient addresses, during which it prunes them based on the retry
1093 hints database. After that it attempts to deliver the message to
1094 any remaining recipients. It then updates the hints database using
1097 So if a recipient address works intermittently, it can get repeatedly
1098 deferred at routing time. The retry hints record remains fresh so the
1099 address never reaches the final cutoff time.
1101 This is a fairly common occurrence when a user is bumping up against
1102 their storage quota. Exim had some logic in its local delivery code
1103 to deal with this. However it did not apply to per-recipient defers
1104 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1106 This change adds a proper retry rule check during routing so that the
1107 final cutoff time is checked against the message's age. We only do
1108 this check if there is an address retry record and there is not a
1109 domain retry record; this implies that previous attempts to handle
1110 the address had the retry_use_local_parts option turned on. We use
1111 this as an approximation for the destination being like a local
1112 delivery, as in LMTP.
1114 I suspect this new check makes the old local delivery cutoff check
1115 redundant, but I have not verified this so I left the code in place.
1117 TF/02 Correct gecos expansion when From: is a prefix of the username.
1119 Test 0254 submits a message to Exim with the header
1123 When I ran the test suite under the user fanf2, Exim expanded
1124 the header to contain my full name, whereas it should have added
1125 a Resent-Sender: header. It erroneously treats any prefix of the
1126 username as equal to the username.
1128 This change corrects that bug.
1130 GF/01 DCC debug and logging tidyup
1131 Error conditions log to paniclog rather than rejectlog.
1132 Debug lines prefixed by "DCC: " to remove any ambiguity.
1134 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1136 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1137 Bug spotted by Jeremy Harris; was flawed since initial commit.
1138 Would have resulted in OCSP responses post-SNI triggering an Exim
1139 NULL dereference and crash.
1141 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1143 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1144 Bug detection, analysis and fix by Samuel Thibault.
1145 Bugzilla 1331, Debian bug #698092.
1147 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1149 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1150 Server implementation by Todd Lyons, client by JH.
1151 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1152 config variable "prdr_enable" controls whether the server
1153 advertises the facility. If the client requests PRDR a new
1154 acl_data_smtp_prdr ACL is called once for each recipient, after
1155 the body content is received and before the acl_smtp_data ACL.
1156 The client is controlled by both of: a hosts_try_prdr option
1157 on the smtp transport, and the server advertisement.
1158 Default client logging of deliveries and rejections involving
1159 PRDR are flagged with the string "PRDR".
1161 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1162 fclose(). Diagnosis by Todd Lyons.
1164 PP/17 Update configure.default to handle IPv6 localhost better.
1165 Patch by Alain Williams (plus minor tweaks).
1168 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1169 This is now consistent with GnuTLS, and is now documented: the
1170 previous undocumented portable approach to treating the option as
1171 unset was to force an expansion failure. That still works, and
1172 an empty string is now equivalent.
1174 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1175 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1176 not performing validation itself.
1178 PP/20 Added force_command boolean option to pipe transport.
1179 Patch from Nick Koston, of cPanel Inc.
1181 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1184 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1186 PP/21 Fix eximon continuous updating with timestamped log-files.
1187 Broken in a format-string cleanup in 4.80, missed when I repaired the
1188 other false fix of the same issue.
1189 Report and fix from Heiko Schlichting.
1192 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1193 Report from Prashanth Katuri.
1195 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1196 It's SecureTransport, so affects any MacOS clients which use the
1197 system-integrated TLS libraries, including email clients.
1199 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1200 using a MIME ACL for non-SMTP local injection.
1201 Report and assistance in diagnosis by Warren Baker.
1203 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1205 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1207 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1208 last id that failed so it may be referenced in subsequent ACL's.
1210 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1213 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1214 ldap library initialization, allowing self-signed CA's to be
1215 used. Also properly sets require_cert option later in code by
1216 using NULL (global ldap config) instead of ldap handle (per
1217 session). Bug diagnosis and testing by alxgomz.
1219 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1220 the src/util/ subdirectory.
1222 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1223 renamed to Transport Post Delivery Action by Jeremy Harris, as
1226 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1227 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1228 redis_servers = needs to be configured which will be used by the redis
1229 lookup. Patch from Warren Baker, of The Packet Hub.
1231 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1233 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1234 hostname or reverse DNS when processing a host list. Used suggestions
1235 from multiple comments on this bug.
1237 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1239 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1240 Missed a few lines, added it to make the runtest require no keyboard
1243 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1244 contains upper case chars. Make router use caseful_local_part.
1246 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1247 support when GnuTLS has been built with p11-kit.
1253 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1255 This, or similar/improved, will also be change PP/11 of 4.82.
1261 PP/01 Handle short writes when writing local log-files.
1262 In practice, only affects FreeBSD (8 onwards).
1263 Bugzilla 1053, with thanks to Dmitry Isaikin.
1265 NM/01 Bugzilla 949 - Documentation tweak
1267 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1270 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1272 PP/02 Implemented gsasl authenticator.
1274 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1276 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1277 `pkg-config foo` for cflags/libs.
1279 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1280 with rest of GSASL and with heimdal_gssapi.
1282 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1283 `pkg-config foo` for cflags/libs for the TLS implementation.
1285 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1286 properties get this fed in as external SSF. A number of robustness
1287 and debugging improvements to the cyrus_sasl authenticator.
1289 PP/08 cyrus_sasl server now expands the server_realm option.
1291 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1292 Patch by Jeremy Harris.
1294 PP/10 Added dbmjz lookup type.
1296 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1298 PP/12 MAIL args handles TAB as well as SP, for better interop with
1299 non-compliant senders.
1300 Analysis and variant patch by Todd Lyons.
1302 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1303 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1304 Patch from Dirk Mueller <dmueller@suse.com>
1306 PP/13 tls_peerdn now print-escaped for spool files.
1307 Observed some $tls_peerdn in wild which contained \n, which resulted
1308 in spool file corruption.
1310 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1311 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1312 or write after TLS renegotiation, which otherwise led to messages
1315 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1316 as a tracking header (ie: a signed header comes before the signature).
1317 Patch from Wolfgang Breyha.
1319 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1320 comma-sep list; embedded commas doubled.
1322 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1324 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1326 Report and patch from Dmitry Banschikov.
1328 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1329 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1330 failures. We appear to now support TLS1.1+ with Exim.
1332 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1333 lets Exim select keys and certificates based upon TLS SNI from client.
1334 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1335 before an outbound SMTP session. New log_selector, +tls_sni.
1337 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1338 NULL dereference. Report and patch from Alun Jones.
1340 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1341 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1342 Not seeing resolver debug output on NetBSD, but suspect this is a
1343 resolver implementation change.
1345 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1346 Left warnings. Added "eximon gdb" invocation mode.
1348 PP/21 Defaulting "accept_8bitmime" to true, not false.
1350 PP/22 Added -bw for inetd wait mode support.
1352 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1353 locate the relevant includes and libraries. Made this the default.
1355 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1356 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1358 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1359 This may cause build issues on older platforms.
1361 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1362 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1363 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1364 Added SNI support via GnuTLS too.
1365 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1367 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1369 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1370 Applied second patch from Janne, implementing suggestion to default
1371 multiple-strings-in-record handling to match SPF spec.
1373 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1375 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1376 read-only, out of scope).
1377 Patch from Wolfgang Breyha, report from Stuart Northfield.
1379 PP/29 Fix three issues highlighted by clang analyser static analysis.
1380 Only crash-plausible issue would require the Cambridge-specific
1381 iplookup router and a misconfiguration.
1382 Report from Marcin Mirosław.
1384 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1386 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1387 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1388 As part of this, removing so much warning spew let me fix some minor
1389 real issues in debug logging.
1391 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1392 assignment on my part. Fixed.
1394 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1395 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1396 Janne Snabb (who went above and beyond: thank you).
1398 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1399 string otherwise requires a connection and a bunch more work and it's
1400 relatively easy to get wrong. Should also expose TLS library linkage
1403 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1404 64-bit ${eval} (JH/03).
1406 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1407 GNU libc to support some of the 64-bit stuff, should not lead to
1408 conflicts. Defined before os.h is pulled in, so if a given platform
1409 needs to override this, it can.
1411 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1412 protection layer was required, which is not implemented.
1413 Bugzilla 1254, patch from Wolfgang Breyha.
1415 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1416 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1417 tls_dhparam take prime identifiers. Also unbreak combination of
1418 OpenSSL+DH_params+TLSSNI.
1420 PP/39 Disable SSLv2 by default in OpenSSL support.
1426 PP/01 Solaris build fix for Oracle's LDAP libraries.
1427 Bugzilla 1109, patch from Stephen Usher.
1429 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1431 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1434 TF/02 Fix a couple more cases where we did not log the error message
1435 when unlink() failed. See also change 4.74-TF/03.
1437 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1438 lock up or crash if it happened to be inside a call to libc when it
1439 got a SIGUSR1 from exiwhat.
1441 The SIGUSR1 handler appends the current process status to the process
1442 log which is later printed by exiwhat. It used to use the general
1443 purpose logging code to do this, but several functions it calls are
1444 not safe for signals.
1446 The new output code in the SIGUSR1 handler is specific to the process
1447 log, and simple enough that it's easy to inspect for signal safety.
1448 Removing some special cases also simplifies the general logging code.
1449 Removing the spurious timestamps from the process log simplifies
1452 TF/04 Improved ratelimit ACL condition.
1454 The /noupdate option has been deprecated in favour of /readonly which
1455 has clearer semantics. The /leaky, /strict, and /readonly update modes
1456 are mutually exclusive. The update mode is no longer included in the
1457 database key; it just determines when the database is updated. (This
1458 means that when you upgrade Exim will forget old rate measurements.)
1460 Exim now checks that the per_* options are used with an update mode that
1461 makes sense for the current ACL. For example, when Exim is processing a
1462 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1463 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1464 must specify per_mail/readonly. If you omit the update mode it defaults to
1465 /leaky where that makes sense (as before) or /readonly where required.
1467 The /noupdate option is now undocumented but still supported for
1468 backwards compatibility. It is equivalent to /readonly except that in
1469 ACLs where /readonly is required you may specify /leaky/noupdate or
1470 /strict/noupdate which are treated the same as /readonly.
1472 A useful new feature is the /count= option. This is a generalization
1473 of the per_byte option, so that you can measure the throughput of other
1474 aggregate values. For example, the per_byte option is now equivalent
1475 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1477 The per_rcpt option has been generalized using the /count= mechanism
1478 (though it's more complicated than the per_byte equivalence). When it is
1479 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1480 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1481 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1482 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1483 Note that using per_rcpt with a non-readonly update mode in more than
1484 one ACL will cause the recipients to be double-counted. (The per_mail
1485 and per_byte options don't have this problem.)
1487 The handling of very low rates has changed slightly. If the computed rate
1488 is less than the event's count (usually one) then this event is the first
1489 after a long gap. In this case the rate is set to the same as this event's
1490 count, so that the first message of a spam run is counted properly.
1492 The major new feature is a mechanism for counting the rate of unique
1493 events. The new per_addr option counts the number of different
1494 recipients that someone has sent messages to in the last time period. It
1495 behaves like per_rcpt if all the recipient addresses are different, but
1496 duplicate recipient addresses do not increase the measured rate. Like
1497 the /count= option this is a general mechanism, so the per_addr option
1498 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1499 example, measure the rate that a client uses different sender addresses
1500 with the options per_mail/unique=$sender_address. There are further
1501 details in the main documentation.
1503 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1505 TF/06 Removed a few PCRE remnants.
1507 TF/07 Automatically extract Exim's version number from tags in the git
1508 repository when doing development or release builds.
1510 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1511 Bugzilla 879. Patch from Paul Fisher.
1513 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1514 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1517 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1519 PP/05 Variable $av_failed, true if the AV scanner deferred.
1520 Bugzilla 1078. Patch from John Horne.
1522 PP/06 Stop make process more reliably on build failure.
1523 Bugzilla 1087. Patch from Heiko Schlittermann.
1525 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1526 Bugzilla 1089. Patch from Heiko Schlittermann.
1528 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1529 Bugzilla 1131. Patch from Holger Weiß.
1531 PP/09 Handle IPv6 addresses with SPF.
1532 Bugzilla 860. Patch from Wolfgang Breyha.
1534 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1536 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1539 PP/11 match_* no longer expand right-hand-side by default.
1540 New compile-time build option, EXPAND_LISTMATCH_RHS.
1541 New expansion conditions, "inlist", "inlisti".
1543 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1545 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1547 PP/14 fix log_write() format string regression from TF/03.
1548 Bugzilla 1152. Patch from Dmitry Isaikin.
1554 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1556 PP/02 Harmonised TLS library version reporting; only show if debugging.
1557 Layout now matches that introduced for other libraries in 4.74 PP/03.
1559 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1561 PP/04 New "dns_use_edns0" global option.
1563 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1566 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1567 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1569 TK/01 Updated PolarSSL code to 0.14.2.
1570 Bugzilla 1097. Patch from Andreas Metzler.
1572 PP/07 Catch divide-by-zero in ${eval:...}.
1573 Fixes bugzilla 1102.
1575 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1578 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1579 format-string attack -- SECURITY: remote arbitrary code execution.
1581 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1582 time unintentionally subject to list matching rules, letting the header
1583 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1584 arbitrary string expansion). This allowed for information disclosure.
1586 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1587 INT_MIN/-1 -- value coerced to INT_MAX.
1593 NM/01 Workaround for PCRE version dependency in version reporting
1596 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1597 This fixes portability to compilers other than gcc, notably
1598 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1600 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1601 makefiles for portability to HP-UX and POSIX correctness.
1603 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1604 Also via indented variable definition in the Makefile.
1605 (Debugging by Oliver Heesakkers).
1607 PP/02 Restore caching of spamd results with expanded spamd_address.
1608 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1610 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1611 Improves build reliability. Fix from: Frank Elsner
1613 NM/02 Fix wide character breakage in the rfc2047 coding
1614 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1616 NM/03 Allow underscore in dnslist lookups
1617 Fixes bug 1026. Patch from Graeme Fowler
1619 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1620 Code patches from Adam Ciarcinski of NetBSD.
1622 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1625 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1626 is logged, to avoid truncation. Patch from John Horne.
1628 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1629 Patch from Jakob Hirsch.
1631 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1632 SQL string expansion failure details.
1633 Patch from Andrey Oktyabrski.
1635 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1636 Patch from Simon Arlott.
1638 PP/09 New lookups functionality failed to compile on old gcc which rejects
1639 extern declarations in function scope.
1640 Patch from Oliver Fleischmann
1642 PP/10 Use sig_atomic_t for flags set from signal handlers.
1643 Check getgroups() return and improve debugging.
1644 Fixed developed for diagnosis in bug 927 (which turned out to be
1647 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1648 Patch from Mark Zealey.
1650 PP/12 Bugzilla 1056: Improved spamd server selection.
1651 Patch from Mark Zealey.
1653 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1654 Based on patch from Heiko Schlittermann.
1656 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1657 Patch from Uwe Doering, sign-off by Michael Haardt.
1659 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1660 variable declaration deep within a block. Bug and patch from
1663 PP/15 lookups-Makefile IRIX compatibility coercion.
1665 PP/16 Make DISABLE_DKIM build knob functional.
1667 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1668 Patch by Simon Arlott
1670 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1671 variable argument macros. Our copy now differs from upstream.
1677 TF/01 Failure to get a lock on a hints database can have serious
1678 consequences so log it to the panic log.
1680 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1681 controlled using the smtp_confirmation log selector.
1683 TF/03 Include the error message when we fail to unlink a spool file.
1685 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1686 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1687 for maintaining out-of-tree patches for some time.
1689 PP/01 Bugzilla 139: Documentation and portability issues.
1690 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1691 Handle per-OS dynamic-module compilation flags.
1693 PP/02 Let /dev/null have normal permissions.
1694 The 4.73 fixes were a little too stringent and complained about the
1695 permissions on /dev/null. Exempt it from some checks.
1696 Reported by Andreas M. Kirchwitz.
1698 PP/03 Report version information for many libraries, including
1699 Exim version information for dynamically loaded libraries. Created
1700 version.h, now support a version extension string for distributors
1701 who patch heavily. Dynamic module ABI change.
1703 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1704 privilege escalation vulnerability whereby the Exim run-time user
1705 can cause root to append content of the attacker's choosing to
1708 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1711 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1712 If dropping privileges for untrusted macros, we disabled normal logging
1713 on the basis that it would fail; for the Exim run-time user, this is not
1714 the case, and it resulted in successful deliveries going unlogged.
1715 Fixed. Reported by Andreas Metzler.
1721 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1722 only prepend for the Resent-* case. Fixes regression introduced in
1723 Exim 4.70 by NM/22 for Bugzilla 607.
1725 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1726 increasing numbers of administrators be bitten by this.
1728 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1730 PP/03 Bugzilla 994: added openssl_options main configuration option.
1732 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1734 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1736 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1738 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1739 without a peer certificate, leading to a segfault because of an
1740 assumption that peers always have certificates. Be a little more
1741 paranoid. Problem reported by Martin Tscholak.
1743 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1744 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1745 NB: ClamAV planning to remove STREAM in "middle of 2010".
1746 CL also introduces -bmalware, various -d+acl logging additions and
1747 more caution in buffer sizes.
1749 PP/09 Implemented reverse_ip expansion operator.
1751 PP/10 Bugzilla 937: provide a "debug" ACL control.
1753 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1755 PP/12 Bugzilla 973: Implement --version.
1757 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1759 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1761 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1763 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1764 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1765 ignore trailing whitespace.
1767 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1769 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1772 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1773 Notification from Dr Andrew Aitchison.
1775 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1776 ExtendedDetectionInfo response format.
1777 Notification from John Horne.
1779 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1782 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1783 XSL and documented dependency on system catalogs, with examples of how
1786 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1789 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1790 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1791 configuration file which is writeable by the Exim user or group.
1793 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1794 of configuration files to cover files specified with the -C option if
1795 they are going to be used with root privileges, not just the default
1798 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1799 option (effectively making it always true).
1801 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1802 files to be used while preserving root privileges.
1804 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1805 that rogue child processes cannot use them.
1807 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1808 run-time user, instead of root.
1810 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1811 Exim run-time user without dropping privileges.
1813 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1814 result string, instead of calling string_vformat() twice with the same
1817 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1818 for other users. Others should always drop root privileges if they use
1819 -C on the command line, even for a whitelisted configure file.
1821 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1823 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1829 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1830 $data_path, and $header_path variables; fixed documentation bugs and
1833 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1834 exipick to access non-standard spools, including the "frozen" queue
1837 NM/01 Bugzilla 965: Support mysql stored procedures.
1838 Patch from Alain Williams
1840 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1842 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1843 Patch from Andreas Metzler
1845 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1846 Patch from Kirill Miazine
1848 NM/05 Bugzilla 671: Added umask to procmail example.
1850 JJ/03 installed exipick 20100323.0, fixing doc bug
1852 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1853 directory. Notification and patch from Dan Rosenberg.
1855 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1857 TK/02 Improve log output when DKIM signing operation fails.
1859 MH/01 Treat the transport option dkim_domain as a colon separated
1860 list, not as a single string, and sign the message with each element,
1861 omitting multiple occurences of the same signer.
1863 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1864 Bugzilla 985, 986. Patch by Simon Arlott
1866 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1867 Patch by Simon Arlott
1869 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1870 MBX locking. Notification from Dan Rosenberg.
1876 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1878 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1880 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1882 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1884 NM/04 Bugzilla 915: Flag broken perl installation during build.
1890 TK/01 Added patch by Johannes Berg that expands the main option
1891 "spamd_address" if it starts with a dollar sign.
1893 TK/02 Write list of recipients to X-Envelope-Sender header when building
1894 the mbox-format spool file for content scanning (suggested by Jakob
1897 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1898 (http://www.dcc-servers.net/) support via dccifd. Activated by
1899 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1901 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1902 by Mark Daniel Reidel <mr@df.eu>.
1904 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1905 When building exim an external PCRE library is now needed -
1906 PCRE is a system library on the majority of modern systems.
1907 See entry on PCRE_LIBS in EDITME file.
1909 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1910 conversation. Added nologin parameter to request.
1911 Patch contributed by Kirill Miazine.
1913 TF/01 Do not log submission mode rewrites if they do not change the address.
1915 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1917 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1918 log files in place. Contributed by Roberto Lima.
1920 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1922 TF/03 Bugzilla 615: When checking the local_parts router precondition
1923 after a local_part_suffix or local_part_prefix option, Exim now
1924 does not use the address's named list lookup cache, since this
1925 contains cached lookups for the whole local part.
1927 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1928 Robert Millan. Documentation is in experimental-spec.txt.
1930 TF/04 Bugzilla 668: Fix parallel build (make -j).
1932 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1934 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1935 Patch provided by Jan Srzednicki.
1937 TF/05 Leading white space used to be stripped from $spam_report which
1938 wrecked the formatting. Now it is preserved.
1940 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1941 that they are available at delivery time.
1943 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1945 TF/08 TLS error reporting now respects the incoming_interface and
1946 incoming_port log selectors.
1948 TF/09 Produce a more useful error message if an SMTP transport's hosts
1949 setting expands to an empty string.
1951 NM/06 Bugzilla 744: EXPN did not work under TLS.
1952 Patch provided by Phil Pennock.
1954 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1955 Patch provided by Richard Godbee.
1957 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1958 acl_smtp_notquit, added index entry.
1960 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1961 Patch provided by Eugene Bujak.
1963 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1964 accept(). Patch provided by Maxim Dounin.
1966 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1967 Patch provided by Phil Pennock.
1969 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1971 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1972 Patch provided by Brad "anomie" Jorsch.
1974 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1975 Patch provided by Dean Brooks.
1977 TK/05 Add native DKIM support (does not depend on external libraries).
1979 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1980 Patch provided by Graeme Fowler.
1982 NM/16 Bugzilla 851: Documentation example syntax fix.
1984 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1986 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1989 NM/19 Bugzilla 745: TLS version reporting.
1990 Patch provided by Phil Pennock.
1992 NM/20 Bugzilla 167: bool: condition support.
1993 Patch provided by Phil Pennock.
1995 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1996 clients. Patch provided by Phil Pennock.
1998 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1999 Patch provided by Brad "anomie" Jorsch.
2001 NM/23 Bugzilla 687: Fix misparses in eximstats.
2002 Patch provided by Heiko Schlittermann.
2004 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2005 Patch provided by Heiko Schlittermann.
2007 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2008 plus update to original patch.
2010 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2012 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2013 Patch provided by David Brownlee.
2015 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2017 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2019 NM/30 Bugzilla 888: TLS documentation bugfixes.
2021 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2023 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2024 Unlike the original bugzilla I have changed all shell scripts in src tree.
2026 NM/33 Bugzilla 898: Transport filter timeout fix.
2027 Patch by Todd Rinaldo.
2029 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2030 Patch by Serge Demonchaux.
2032 NM/35 Bugzilla 39: Base64 decode bug fixes.
2033 Patch by Jakob Hirsch.
2035 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2037 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2039 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2041 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2047 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2048 ALT-N's libdkim that I have put here:
2049 http://duncanthrax.net/exim-experimental/
2051 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2052 were called 'true' and it seems that C99 defines that as a reserved
2053 keyword to be used with 'bool' variable types. That means you could
2054 not include C99-style headers which use bools without triggering
2055 build errors in sieve.c.
2057 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2058 as mailq or other aliases. Changed the --help handling significantly
2059 to do whats expected. exim_usage() emits usage/help information.
2061 SC/01 Added the -bylocaldomain option to eximstats.
2063 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2065 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2067 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2073 PH/01 Another patch from the Sieve maintainer.
2075 PH/02 When an IPv6 address is converted to a string for single-key lookup
2076 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2077 dots are used instead of colons so that keys in lsearch files need not
2078 contain colons. This was done some time before quoting was made available
2079 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2080 (notated using the quote facility) so as to distinguish them from IPv4
2081 keys. This meant that lookups for IP addresses in host lists did not work
2082 for iplsearch lookups.
2084 This has been fixed by arranging for IPv6 addresses to be expressed with
2085 colons if the lookup type is iplsearch. This is not incompatible, because
2086 previously such lookups could never work.
2088 The situation is now rather anomalous, since one *can* have colons in
2089 ordinary lsearch keys. However, making the change in all cases is
2090 incompatible and would probably break a number of configurations.
2092 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2095 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2096 conversion specification without a maximum field width, thereby enabling
2097 a rogue spamd server to cause a buffer overflow. While nobody in their
2098 right mind would setup Exim to query an untrusted spamd server, an
2099 attacker that gains access to a server running spamd could potentially
2100 exploit this vulnerability to run arbitrary code as the Exim user.
2102 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2103 $primary_hostname instead of what libspf2 thinks the hosts name is.
2105 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2106 a directory entry by the name of the lookup key. Previously, if a
2107 symlink pointed to a non-existing file or a file in a directory that
2108 Exim lacked permissions to read, a lookup for a key matching that
2109 symlink would fail. Now it is enough that a matching directory entry
2110 exists, symlink or not. (Bugzilla 503.)
2112 PH/03 The body_linecount and body_zerocount variables are now exported in the
2115 PH/04 Added the $dnslist_matched variable.
2117 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2118 This means they are set thereafter only if the connection becomes
2121 PH/06 Added the client_condition to authenticators so that some can be skipped
2122 by clients under certain conditions.
2124 PH/07 The error message for a badly-placed control=no_multiline_responses left
2125 "_responses" off the end of the name.
2127 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2129 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2130 (without spaces) instead of just copying the configuration text.
2132 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2134 PH/11 Added $max_received_linelength.
2136 PH/12 Added +ignore_defer and +include_defer to host lists.
2138 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2139 way in which PCRE > 7.0 is built.
2141 PH/14 Implemented queue_only_load_latch.
2143 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2144 MAIL command. The effect was to mangle the value on 64-bit systems.
2146 PH/16 Another patch from the Sieve maintainer.
2148 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2150 PH/18 If a system quota error occurred while trying to create the file for
2151 a maildir delivery, the message "Mailbox is full" was not appended to the
2152 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2153 only to a quota excession during the actual writing of the file.
2155 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2156 characters?) which causes problems in log lines. The DN values are now
2157 passed through string_printing() before being added to log lines.
2159 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2160 and InterBase are left for another time.)
2162 PH/21 Added message_body_newlines option.
2164 PH/22 Guard against possible overflow in moan_check_errorcopy().
2166 PH/23 POSIX allows open() to be a macro; guard against that.
2168 PH/24 If the recipient of an error message contained an @ in the local part
2169 (suitably quoted, of course), incorrect values were put in $domain and
2170 $local_part during the evaluation of errors_copy.
2176 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2177 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2180 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2181 issue a MAIL command.
2183 PH/02 In an ACL statement such as
2185 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2187 if a client was not listed at all, or was listed with a value other than
2188 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2189 the condition was not true (as it should be), so access was not denied.
2190 The bug was that the ! inversion was incorrectly passed on to the second
2191 item. This has been fixed.
2193 PH/03 Added additional dnslists conditions == and =& which are different from
2194 = and & when the dns lookup returns more than one IP address.
2196 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2197 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2199 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2200 FSYNC, which compiles an option called disable_fsync that allows for
2201 bypassing fsync(). The documentation is heavily laced with warnings.
2203 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2205 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2206 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2207 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2208 including adding "make clean"; (3) Added -fPIC when compiling the test
2209 dynamically loaded module, to get rid of a warning.
2211 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2212 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2213 = 0s. The bug is otherwise harmless.
2215 PH/07 There was a bug in the dovecot authenticator such that the value of
2216 $auth1 could be overwritten, and so not correctly preserved, after a
2217 successful authentication. This usually meant that the value preserved by
2218 the server_setid option was incorrect.
2220 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2222 PH/09 Installed PCRE release 7.0.
2224 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2225 run for batched SMTP input. It is now run at the start of every message
2226 in the batch. While fixing this I discovered that the process information
2227 (output by running exiwhat) was not always getting set for -bs and -bS
2228 input. This is fixed, and it now also says "batched" for BSMTP.
2230 PH/11 Added control=no_pipelining.
2232 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2233 patch, slightly modified), and move the expansion of helo_data till after
2234 the connection is made in the smtp transport (so it can use these
2237 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2239 PH/14 Added log_selector = +pid.
2241 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2243 PH/16 Add ${if forany and ${if forall.
2245 PH/17 Added dsn_from option to vary the From: line in DSNs.
2247 PH/18 Flush SMTP output before performing a callout, unless control =
2248 no_callout_flush is set.
2250 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2251 was true (the default) a successful delivery failed to delete the retry
2252 item, thus causing premature timeout of the address. The bug is now
2255 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2257 PH/21 Long custom messages for fakedefer and fakereject are now split up
2258 into multiline responses in the same way that messages for "deny" and
2259 other ACL rejections are.
2261 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2262 with slight modification.
2264 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2265 draft, changing the syntax and factoring some duplicate code".
2267 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2268 for deliveries of the second and subsequent messages over the same SMTP
2271 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2272 ${reduce, with only minor "tidies".
2274 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2276 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2277 expansion side effects.
2279 PH/27 When a message times out after an over-quota error from an Exim-imposed
2280 quota, the bounce message says "mailbox is full". This message was not
2281 being given when it was a system quota that was exceeded. It now should
2284 MH/03 Made $recipients available in local_scan(). local_scan() already has
2285 better access to the recipient list through recipients_list[], but
2286 $recipients can be useful in postmaster-provided expansion strings.
2288 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2289 in the case of a MAIL command with additional options following the
2290 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2291 were accidentally chopped off.
2293 PH/29 SMTP synchronization checks are implemented when a command is read -
2294 there is a check that no more input is waiting when there shouldn't be
2295 any. However, for some commands, a delay in an ACL can mean that it is
2296 some time before the response is written. In this time, more input might
2297 arrive, invalidly. So now there are extra checks after an ACL has run for
2298 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2299 pipelining has not been advertised.
2301 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2303 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2304 returned in the SMTP response when smtp_return_error_details was false.
2305 This has been fixed.
2307 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2308 instead of the C I/O that was originally supplied, because problems were
2309 reported on Solaris.
2311 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2312 Exim which did not show up earlier: it was assuming that a call to
2313 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2314 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2315 was a macro that became an assignment, so it seemed to work. This has
2316 changed to a proper function call with a void return, hence the compile
2317 error. Exim's code has been fixed.
2319 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2322 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2324 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2326 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2327 because it turns out to be too big or there is a timeout) there is no
2328 "Completed" line in the log. When some messages of this type were
2329 selected by exigrep, they were listed as "not completed". Others were
2330 picked up by some special patterns. I have improved the selection
2331 criteria to be more general.
2333 PH/38 The host_find_failed option in the manualroute router can now be set
2334 to "ignore", to completely ignore a host whose IP address cannot be
2335 found. If all hosts are ignored, the behaviour is controlled by the new
2336 host_all_ignored option.
2338 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2339 homing or because of multiple MX records with /mx) generated more than
2340 one IP address, and the following item turned out to be the local host,
2341 all the secondary addresses of the first item were incorrectly removed
2342 from the list, along with the local host and any following hosts (which
2343 is what is supposed to happen).
2345 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2346 whoever called Exim into the -H file. In the case of the daemon it was
2347 behaving confusingly. When first started, it used values for whoever
2348 started the daemon, but after a SIGHUP it used the Exim user (because it
2349 calls itself on a restart). I have changed the code so that it now always
2352 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2353 message are rejected with the same error (e.g. no authentication or bad
2354 sender address), and a DATA command is nevertheless sent (as can happen
2355 with PIPELINING or a stupid MUA), the error message that was given to the
2356 RCPT commands is included in the rejection of the DATA command. This is
2357 intended to be helpful for MUAs that show only the final error to their
2360 PH/42 Another patch from the Sieve maintainer.
2362 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2363 Eximstats - Fixed some broken HTML links and added missing column headers
2365 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2366 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2368 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2370 PH/43 Yet another patch from the Sieve maintainer.
2372 PH/44 I found a way to check for a TCP/IP connection going away before sending
2373 the response to the final '.' that terminates a message, but only in the
2374 case where the client has not sent further data following the '.'
2375 (unfortunately, this is allowed). However, in many cases there won't be
2376 any further data because there won't be any more messages to send. A call
2377 to select() can be used: if it shows that the input is "ready", there is
2378 either input waiting, or the socket has been closed. An attempt to read
2379 the next input character can distinguish the two cases. Previously, Exim
2380 would have sent an OK response which the client would never have see.
2381 This could lead to message repetition. This fix should cure that, at
2382 least in a lot of common cases.
2384 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2385 advertised in response to EHLO.
2391 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2392 fixed by 4.65/MH/01 (is this a record?) are fixed:
2394 (i) An empty string was always treated as zero by the numeric comparison
2395 operators. This behaviour has been restored.
2397 (ii) It is documented that the numeric comparison operators always treat
2398 their arguments as decimal numbers. This was broken in that numbers
2399 starting with 0 were being interpreted as octal.
2401 While fixing these problems I realized that there was another issue that
2402 hadn't been noticed. Values of message_size_limit (both the global option
2403 and the transport option) were treated as octal if they started with 0.
2404 The documentation was vague. These values are now always treated as
2405 decimal, and I will make that clear in the documentation.
2411 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2412 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2415 MH/01 Don't check that the operands of numeric comparison operators are
2416 integers when their expansion is in "skipping" mode (fixes bug
2417 introduced by 4.64-PH/07).
2419 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2420 child addresses, Exim now panics and dies. Previously, because the count
2421 is held in a short int, deliveries were likely to be lost. As such a
2422 large number of recipients for a single message is ridiculous
2423 (performance will be very, very poor), I have chosen to impose a limit
2424 rather than extend the field.
2430 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2431 leftover -K file (the existence of which was triggered by #402).
2432 While we were at it, introduced process PID as part of the -K
2433 filename. This should rule out race conditions when creating
2436 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2437 processing considerably. Previous code took too long for large mails,
2438 triggering a timeout which in turn triggers #401.
2440 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2441 in the DK code in transports.c. sendfile() is not really portable,
2442 hence the _LINUX specificness.
2444 TF/01 In the add_headers option to the mail command in an Exim filter,
2445 there was a bug that Exim would claim a syntax error in any
2446 header after the first one which had an odd number of characters
2449 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2450 callout verification, Exim cached a "reject" for the entire domain. This
2451 is correct for most verifications, but it is not correct for a recipient
2452 verification with use_sender or use_postmaster set, because in that case
2453 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2454 case of MAIL FROM:<> rejection from other early rejections (e.g.
2455 rejection of HELO). When verifying a recipient using a non-null MAIL
2456 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2457 Whatever the result of the callout, the value of the domain cache is
2458 left unchanged (for any other kind of callout, getting as far as trying
2459 RCPT means that the domain itself is ok).
2461 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2464 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2465 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2466 session, the daemon ignores ECONNECT errors and logs others; it now
2467 ignores EPIPE as well.
2469 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2470 (quoted-printable decoding).
2472 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2473 later the small subsequent patch to fix an introduced bug.
2475 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2477 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2479 PH/08 An error is now given if message_size_limit is specified negative.
2481 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2482 to be given (somewhat) arbitrary names.
2484 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2487 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2488 miscellaneous code fixes
2490 PH/10 Added the log_reject_target ACL modifier to specify where to log
2493 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2494 hostname. This is wrong, because it relates to the incoming message (and
2495 probably the interface on which it is arriving) and not to the outgoing
2496 callout (which could be using a different interface). This has been
2497 changed to use the value of the helo_data option from the smtp transport
2498 instead - this is what is used when a message is actually being sent. If
2499 there is no remote transport (possible with a router that sets up host
2500 addresses), $smtp_active_hostname is used.
2502 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2503 tweaks were necessary in order to get it to work (see also 21 below):
2504 (a) The code assumed that strncpy() returns a negative number on buffer
2505 overflow, which isn't the case. Replaced with Exim's string_format()
2507 (b) There were several signed/unsigned issues. I just did the minimum
2508 hacking in of casts. There is scope for a larger refactoring.
2509 (c) The code used strcasecmp() which is not a standard C function.
2510 Replaced with Exim's strcmpic() function.
2511 (d) The code set only $1; it now sets $auth1 as well.
2512 (e) A simple test gave the error "authentication client didn't specify
2513 service in request". It would seem that Dovecot has changed its
2514 interface. Fortunately there's a specification; I followed it and
2515 changed what the client sends and it appears to be working now.
2517 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2520 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2521 address A is aliased to B and C, where B exists and C does not. Without
2522 -v the output is "A verified" because verification stops after a
2523 successful redirection if more than one address is generated. However,
2524 with -v the child addresses are also verified. Exim was outputting "A
2525 failed to verify" and then showing the successful verification for C,
2526 with its parentage. It now outputs "B failed to verify", showing B's
2527 parentage before showing the successful verification of C.
2529 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2530 look up a TXT record in a specific list after matching in a combined
2533 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2534 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2535 they consult the DNS. I had assumed they would set it the way they
2536 wanted; and indeed my experiments on Linux seem to show that in some
2537 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2538 To be on the safe side, however, I have now made the interface to
2539 host_find_byname() similar to host_find_bydns(), with an argument
2540 containing the DNS resolver options. The host_find_byname() function now
2541 sets these options at its start, just as host_find_bydns() does. The smtp
2542 transport options dns_qualify_single and dns_search_parents are passed to
2543 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2544 of host_find_byname() use the default settings of RES_DEFNAMES
2545 (qualify_single) but not RES_DNSRCH (search_parents).
2547 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2548 spool_read_header() do less string testing, by means of a preliminary
2549 switch on the second character of optional "-foo" lines. (This is
2550 overdue, caused by the large number of possibilities that now exist.
2551 Originally there were few.) While I was there, I also converted the
2552 str(n)cmp tests so they don't re-test the leading "-" and the first
2553 character, in the hope this might squeeze out yet more improvement.
2555 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2556 flag allowing group syntax was set by the header_syntax check but not
2557 turned off, possible causing trouble later; (2) The flag was not being
2558 set at all for the header_verify test, causing "group"-style headers to
2559 be rejected. I have now set it in this case, and also caused header_
2560 verify to ignore an empty address taken from a group. While doing this, I
2561 came across some other cases where the code for allowing group syntax
2562 while scanning a header line wasn't quite right (mostly, not resetting
2563 the flag correctly in the right place). These bugs could have caused
2564 trouble for malformed header lines. I hope it is now all correct.
2566 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2567 with the "reply" argument non-NULL. The code, however (which originally
2568 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2569 but it didn't always do it. This confused somebody who was copying the
2570 code for some other use. I have removed all the tests.
2572 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2573 feature that was used to support insecure browsers during the U.S. crypto
2574 embargo. It requires special client support, and Exim is probably the
2575 only MTA that supported it -- and would never use it because real RSA is
2576 always available. This code has been removed, because it had the bad
2577 effect of slowing Exim down by computing (never used) parameters for the
2578 RSA_EXPORT functionality.
2580 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2581 authenticator to fail if there's a tab character in the incoming data
2582 (there should never be unless someone is messing about, as it's supposed
2583 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2584 if the connection is using TLS or if the remote IP is the same as the
2585 local IP, and the "valid-client-cert option" if a client certificate has
2588 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2589 authenticators. This can be used for authorization after authentication
2590 succeeds. (In the case of plaintext, it servers for both authentication
2593 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2594 if any retry times were supplied.
2596 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2597 connection, where there is no client IP address to check. In this
2598 situation, the verify now always succeeds.
2600 PH/25 Applied John Jetmore's -Mset patch.
2602 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2604 PH/27 In a string expansion for a processed (not raw) header when multiple
2605 headers of the same name were present, leading whitespace was being
2606 removed from all of them, but trailing whitespace was being removed only
2607 from the last one. Now trailing whitespace is removed from each header
2608 before concatenation. Completely empty headers in a concatenation (as
2609 before) are ignored.
2611 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2612 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2614 PH/29 [Removed. This was a change that I later backed out, and forgot to
2615 correct the ChangeLog entry (that I had efficiently created) before
2616 committing the later change.]
2618 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2619 address errors (4xx response to RCPT) over the same connection as other
2620 messages routed to the same hosts. Such deliveries are always "forced",
2621 so retry times are not inspected. This resulted in far too many retries
2622 for the affected addresses. The effect occurred only when there were more
2623 hosts than the hosts_max_try setting in the smtp transport when it had
2624 the 4xx errors. Those hosts that it had tried were not added to the list
2625 of hosts for which the message was waiting, so if all were tried, there
2626 was no problem. Two fixes have been applied:
2628 (i) If there are any address or message errors in an SMTP delivery, none
2629 of the hosts (tried or untried) are now added to the list of hosts
2630 for which the message is waiting, so the message should not be a
2631 candidate for sending over the same connection that was used for a
2632 successful delivery of some other message. This seems entirely
2633 reasonable: after all the message is NOT "waiting for some host".
2634 This is so "obvious" that I'm not sure why it wasn't done
2635 previously. Hope I haven't missed anything, but it can't do any
2636 harm, as the worst effect is to miss an optimization.
2638 (ii) If, despite (i), such a delivery is accidentally attempted, the
2639 routing retry time is respected, so at least it doesn't keep
2640 hammering the server.
2642 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2643 in ${readsocket because some servers need this prod.
2645 PH/32 Added some extra debug output when updating a wait-xxx database.
2647 PH/33 The hint "could be header name not terminated by colon", which has been
2648 given for certain expansion errors for a long time, was not being given
2649 for the ${if def:h_colon_omitted{... case.
2651 PH/34 The spec says: "With one important exception, whenever a domain list is
2652 being scanned, $domain contains the subject domain." There was at least
2653 one case where this was not true.
2655 PH/35 The error "getsockname() failed: connection reset by peer" was being
2656 written to the panic log as well as the main log, but it isn't really
2657 panic-worthy as it just means the connection died rather early on. I have
2658 removed the panic log writing for the ECONNRESET error when getsockname()
2661 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2662 runs only) independently of the message's sender address. This meant
2663 that, if the 4xx error was in fact related to the sender, a different
2664 message to the same recipient with a different sender could confuse
2665 things. In particular, this can happen when sending to a greylisting
2666 server, but other circumstances could also provoke similar problems.
2667 I have changed the default so that the retry time for these errors is now
2668 based a combination of the sender and recipient addresses. This change
2669 can be overridden by setting address_retry_include_sender=false in the
2672 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2673 remote server are returned as part of bounce messages. This was not
2674 happening for LMTP over a pipe (the lmtp transport), but now it is the
2675 same for both kinds of LMTP.
2677 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2678 in header lines that were in fact CNAMEs. This is no longer the case.
2680 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2681 and queue runs started by the daemon processed all messages. This has
2682 been fixed so that -R and -S can now usefully be given with -q<time>.
2684 PH/40 Import PCRE release 6.7 (fixes some bugs).
2686 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2688 PH/42 Give an error if -q is specified more than once.
2690 PH/43 Renamed the variables $interface_address and $interface_port as
2691 $received_ip_address and $received_port, to make it clear that these
2692 values apply to message reception, and not to the outgoing interface when
2693 a message is delivered. (The old names remain recognized, of course.)
2695 PH/44 There was no timeout on the connect() call when using a Unix domain
2696 socket in the ${readsocket expansion. There now is.
2698 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2699 be meaningful with "accept".
2701 SC/01 Eximstats V1.43
2702 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2704 SC/02 Eximstats V1.44
2705 Use a glob alias rather than an array ref in the generated
2706 parser. This improves both readability and performance.
2708 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2709 Collect SpamAssassin and rejection statistics.
2710 Don't display local sender or destination tables unless
2711 there is data to show.
2712 Added average volumes into the top table text output.
2714 SC/04 Eximstats V1.46
2715 Collect data on the number of addresses (recipients)
2716 as well as the number of messages.
2718 SC/05 Eximstats V1.47
2719 Added 'Message too big' to the list of mail rejection
2720 reasons (thanks to Marco Gaiarin).
2722 SC/06 Eximstats V1.48
2723 Mainlog lines which have GMT offsets and are too short to
2724 have a flag are now skipped.
2726 SC/07 Eximstats V1.49 (Alain Williams)
2727 Added the -emptyok flag.
2729 SC/08 Eximstats V1.50
2730 Fixes for obtaining the IP address from reject messages.
2732 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2733 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2734 whitespace changes from 4.64-PH/27
2736 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2739 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2740 are found, allow negative numbers in numeric criteria)
2742 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2744 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2745 to match changes made in 4.64-PH/43
2747 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2749 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2750 "domain" is actually an IP address, and force a failure. This locks out
2751 those revolvers/nameservers that support "A-for-A" lookups, in
2752 contravention of the specifications.
2754 PH/48 When a host name was looked up from an IP address, and the subsequent
2755 forward lookup of the name timed out, the host name was left in
2756 $sender_host_name, contrary to the specification.
2758 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2759 restricted to single-key lookups, Exim was not diagnosing an error if
2760 * or *@ was used with a query-style lookup.
2762 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2764 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2765 long ago, but noone interested enough thought of it. Let's just say that
2766 the "1.1" means that there are some new functions that weren't there at
2767 some point in the past.
2769 PH/51 Error processing for expansion failure of helo_data from an smtp
2770 transport during callout processing was broken.
2772 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2773 tested/used via the -bh/-bhc/-bs options.
2775 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2776 bug, fixed in subsequent PCRE releases).
2778 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2779 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2781 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2787 SC/01 Use a glob alias rather than an array ref in eximstats generated
2788 parser. This improves both readability and performance.
2790 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2791 Don't display local sender or destination tables in eximstats unless
2792 there is data to show.
2793 Added average volumes into the eximstats top table text output.
2795 SC/03 Collect data on the number of addresses (recipients) as well
2796 as the number of messages in eximstats.
2798 TF/01 Correct an error in the documentation for the redirect router. Exim
2799 does (usually) call initgroups() when daemonizing.
2801 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2802 with consistent privilege compared to when running as a daemon.
2804 TF/03 Note in the spec that $authenticated_id is not set for local
2805 submissions from trusted users.
2807 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2808 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2810 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2811 by adding some example configuration directives to the default
2812 configuration file. A little bit of work is required to uncomment the
2813 directives and define how usernames and passwords are checked, but
2814 there is now a framework to start from.
2816 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2817 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2818 without this. I don't know how relevant this is to other LDAP libraries.
2820 PH/02 Add the verb name to the "unknown ACL verb" error.
2822 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2824 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2826 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2827 directory not expanded when it should be if an expanded home directory
2828 was set for the address (which is overridden by the transport).
2830 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2833 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2834 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2835 because it is too late at that time, and has no effect.
2837 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2838 security issue with \' (bugzilla #107). I could not use the
2839 PQescapeStringConn() function, because it needs a PGconn value as one of
2842 PH/08 When testing addresses using -bt, indicate those final addresses that
2843 are duplicates that would not cause an additional delivery. At least one
2844 person was confused, thinking that -bt output corresponded to deliveries.
2845 (Suppressing duplicates isn't a good idea as you lose the information
2846 about possibly different redirections that led to the duplicates.)
2848 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2849 systems where poll() doesn't work, in particular OS X.
2851 PH/10 Added more information to debugging output for retry time not reached.
2853 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2854 operations in malware.c.
2856 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2859 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2860 syslog_duplication set false, log lines that would normally be written
2861 both the the main log and to the reject log were not written to syslog at
2864 PH/14 In the default configuration, change the use of "message" in ACL warn
2865 statements to "add_header".
2867 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2868 not followed by a command (e.g. "seen endif").
2870 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2871 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2874 PH/17 Added extra conditions to the default value of delay_warning_condition
2878 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2879 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2880 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2883 The Auto-Submitted: and various List- headers are standardised, whereas I
2884 don't think Precedence: ever was.
2886 PH/18 Refactored debugging code in route_finduser() to show more information,
2887 in particular, the error code if getpwnam() issues one.
2889 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2890 This is apparently needed in addition to the PH/07 change above to avoid
2891 any possible encoding problems.
2893 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2894 but not after initializing Perl.
2896 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2897 output them only if debugging. By default they are written stderr,
2898 apparently, which is not desirable.
2900 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2903 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2906 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2908 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2909 authenticated or an ident call has been made. Suppress the default
2910 values for $authenticated_id and $authenticated_sender (but permit -oMai
2911 and -oMas) when testing with -bh.
2913 PH/24 Re-jigged the order of the tests in the default configuration so that the
2914 tests for valid domains and recipients precede the DNS black list and CSA
2915 tests, on the grounds that those ones are more expensive.
2917 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2918 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2919 This bug exists in every version of Exim that I still have, right back to
2922 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2923 However, an attempt to turn on case-sensitivity in a regex key by
2924 including (?-i) didn't work because the subject string was already
2925 lowercased, and the effects were non-intuitive. It turns out that a
2926 one-line patch can be used to allow (?-i) to work as expected.
2932 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2933 other effects) broke the use of negated acl sub-conditions.
2935 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2938 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2939 "Deny" causes Exim to reject the incoming connection with a 554 error.
2940 Unfortunately, if there is a major crisis, such as a disk failure,
2941 tcp-wrappers gives "deny", whereas what one would like would be some
2942 kind of temporary error. A kludge has been added to help with this.
2943 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2944 554 error is used if errno is still zero or contains ENOENT (which occurs
2945 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2948 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2950 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2951 errors. Otherwise a message that provokes a temporary error (when other
2952 messages do not) can cause a whole host to time out.
2954 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2955 addresses were routed directly to files or pipes from a redirect router.
2956 File deliveries just didn't batch; pipe deliveries might have suffered
2959 PH/06 A failure to get a lock for a hints database would erroneously always say
2960 "Failed to get write lock", even when it was really a read lock.
2962 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2963 of 0600. This has been changed to use the value of the lockfile_mode
2964 option (which defaults to 0600).
2966 PH/08 Applied small patch from the Sieve maintainer.
2968 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2969 folder from quota calculations, a direct delivery into this folder messed
2970 up the contents of the maildirsize file. This was because the regex was
2971 used only to exclude .Trash (or whatever) when the size of the mailbox
2972 was calculated. There was no check that a delivery was happening into an
2973 excluded directory. This bug has been fixed by ignoring all quota
2974 processing for deliveries into excluded directories.
2976 PH/10 Added the maildirfolder_create_regex option to appendfile.
2982 PH/01 The code for finding all the local interface addresses on a FreeBSD
2983 system running IPv6 was broken. This may well have applied to all BSD
2984 systems, as well as to others that have similar system calls. The broken
2985 code found IPv4 interfaces correctly, but gave incorrect values for the
2986 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2987 that it would not match correctly against @[] and not recognize the IPv6
2990 PH/02 The ipliteral router was not recognizing addresses of the form user@
2991 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2993 PH/03 Added disable_ipv6.
2995 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2996 decoded form, because it is most often used to construct To: headers
2997 lines in autoreplies, and the decoded form may well be syntactically
2998 invalid. However, $reply_address has leading white space removed, and all
2999 newlines turned into spaces so that the autoreply transport does not
3002 PH/05 If group was specified without a user on a router, and no group or user
3003 was specified on a transport, the group from the router was ignored.
3005 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3006 for visible compile-time settings that can be used to change these
3007 numbers, for those that want even more. Backwards compatibility with old
3008 spool files has been maintained. However, going back to a previous Exim
3009 release will lost any variables that are in spool files.
3011 PH/07 Two small changes when running in the test harness: increase delay when
3012 passing a TCP/IP connection to a new process, in case the original
3013 process has to generate a bounce, and remove special handling of
3014 127.0.0.2 (sic), which is no longer necessary.
3016 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3017 be the same on different OS.
3019 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3022 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3023 whether --show-vars was specified or not
3025 JJ/02 exipick: Added support for new ACL variable spool format introduced
3028 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3029 syntactically invalid From: or Reply-to: line, and a filter used this to
3030 generate an autoreply, and therefore failed to obtain an address for the
3031 autoreply, Exim could try to deliver to a non-existent relative file
3032 name, causing unrelated and misleading errors. What now happens is that
3033 it logs this as a hard delivery error, but does not attempt to create a
3036 PH/11 The exinext utility has a -C option for testing purposes, but although
3037 the given file was scanned by exinext itself; it wasn't being passed on
3038 when Exim was called.
3040 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3041 an end-of-file indication when reading a command response.
3043 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3044 compiled. In many other places in Exim, IPv6 addresses are always
3045 recognized, so I have changed this. It also means that IPv4 domain
3046 literals of the form [IPV4:n.n.n.n] are now always recognized.
3048 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3049 used if the router is not running as root, for example, when verifying at
3050 ACL time, or when using -bh. The debugging output from this situation was
3051 non-existent - all you got was a failure to exec. I have made two
3054 (a) Failures to set uid/gid, the current directory, or a process leader
3055 in a subprocess such as that created by queryprogram now generate
3056 suitable debugging output when -d is set.
3058 (b) The queryprogram router detects when it is not running as root,
3059 outputs suitable debugging information if -d is set, and then runs
3060 the subprocess without attempting to change uid/gid.
3062 PH/15 Minor change to Makefile for building test_host (undocumented testing
3065 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3066 additional section of a DNS packet that returns MX or SRV records.
3067 Instead, it always explicitly searches for A/AAAA records. This avoids
3068 major problems that occur when a DNS server includes only records of one
3069 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3070 fixed another bug: if SRV records were looked up and the corresponding
3071 address records were *not* found in the additional section, the port
3072 values from the SRV records were lost.
3074 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3075 using the correct key (the original address) when searching the retry
3076 rules in order to find which one to use for generating the retry hint.
3078 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3079 adding the default one. Similarly, if it contains a Reply-To: header, the
3080 errors_reply_to option, if set, is not used.
3082 PH/19 When calculating a retry time, Exim used to measure the "time since
3083 failure" by looking at the "first failed" field in the retry record. Now
3084 it does not use this if it is later than than the arrival time of the
3085 message. Instead it uses the arrival time. This makes for better
3086 behaviour in cases where some deliveries succeed, thus re-setting the
3087 "first failed" field. An example is a quota failure for a huge message
3088 when small messages continue to be delivered. Without this change, the
3089 "time since failure" will always be short, possible causing more frequent
3090 delivery attempts for the huge message than are intended.
3091 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3093 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3094 $1, $2, $3) because the numerical variables can be reset during some
3095 expansion items (e.g. "match"), thereby losing the authentication data.
3097 PH/21 Make -bV show the size of off_t variables so that the test suite can
3098 decide whether to run tests for quotas > 2G.
3100 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3101 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3102 filecount value is greater than 2G or if a quota value is greater than 2G
3103 on a system where the size of off_t is not greater than 4, a panic error
3106 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3107 never match. The debug and -bh output now contains an explicit error
3108 message indicating a malformed IPv4 address or mask.
3110 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3111 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3112 PH/23 above applies.
3114 PH/25 Do not write to syslog when running in the test harness. The only
3115 occasion when this arises is a failure to open the main or panic logs
3116 (for which there is an explicit test).
3118 PH/26 Added the /no_tell option to "control=freeze".
3120 PH/27 If a host name lookup failed very early in a connection, for example, if
3121 the IP address matched host_lookup and the reverse lookup yielded a name
3122 that did not have a forward lookup, an error message of the form "no IP
3123 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3124 could be logged. Now it outputs the IP address instead of "NULL".
3126 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3127 allows the sender and the authenticated sender to be set when
3128 submitting a message from within Exim. Since child_open_exim() is
3129 documented for local_scan(), the new function should be too.
3131 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3132 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3133 results in an empty string is now treated as unset.
3135 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3137 PH/31 Added sender_verify_fail logging option.
3139 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3140 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3141 tidied the source and removed it altogether.
3143 PH/33 When a queue run was abandoned because the load average was too high, a
3144 log line was always written; now it is written only if the queue_run log
3145 selector is set. In addition, the log line for abandonment now contains
3146 information about the queue run such as the pid. This is always present
3147 in "start" and "stop" lines but was omitted from the "abandon" line.
3149 PH/34 Omit spaces between a header name and the colon in the error message that
3150 is given when verify = headers_syntax fails (if there are lots of them,
3151 the message gets confusing).
3153 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3154 names, as there are now some PTR records that contain slashes. This check
3155 is only to protect against broken name servers that fall over on strange
3156 characters, so the fact that it applies to all lookups doesn't matter.
3158 PH/36 Now that the new test suite is complete, we can remove some of the
3159 special code in Exim that was needed for the old test suite. For example,
3160 sorting DNS records because real resolvers return them in an arbitrary
3161 order. The new test suite's fake resolver always returns records in the
3164 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3165 bounces) except when queue_only is set, to avoid logging races between
3166 the different processes.
3168 PH/38 Panic-die if .include specifies a non-absolute path.
3170 PH/39 A tweak to the "H" retry rule from its user.
3172 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3173 a label. They prevented compilation on older perls.
3175 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3176 a warning to be raised on newish perls.
3178 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3179 on queue. Changes to match documented behaviour of showing count of
3180 messages matching specified criteria.
3182 PH/40 Changed the default ident timeout from 30s to 5s.
3184 PH/41 Added support for the use of login_cap features, on those BSD systems
3185 that have them, for controlling the resources used by pipe deliveries.
3187 PH/42 The content-scanning code uses fopen() to create files in which to put
3188 message data. Previously it was not paying any attention to the mode of
3189 the files. Exim runs with umask(0) because the rest of the code creates
3190 files with open(), and sets the required mode explicitly. Thus, these
3191 files were ending up world-writeable. This was not a big issue, because,
3192 being within the spool directory, they were not world-accessible. I have
3193 created a function called modefopen, which takes an additional mode
3194 argument. It sets umask(777), creates the file, chmods it to the required
3195 mode, then resets the umask. All the relevant calls to fopen() in the
3196 content scanning code have been changed to use this function.
3198 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3199 to 24 hours. This avoids potential overflow problems when processing G
3200 and H retry rules. I suspect nobody ever tinkers with this value.
3202 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3204 PH/45 When the plaintext authenticator is running as a client, the server's
3205 challenges are checked to ensure they are valid base64 strings. By
3206 default, the authentication attempt is cancelled if an invalid string is
3207 received. Setting client_ignore_invalid_base64 true ignores these errors.
3208 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3209 they are received. Thus, the responses can be made to depend on the
3210 challenges. If an invalid string is ignored, an empty string is placed in
3213 PH/46 Messages that are created by the autoreply transport now contains a
3214 References: header, in accordance with RFCs 2822 and 3834.
3216 PH/47 Added authenticated_sender_force to the smtp transport.
3218 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3220 PH/49 Installed latest patch from the Sieve maintainer.
3222 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3223 was also set, the appendfile transport was unnecessarily scanning a
3224 directory of message files (e.g. for maildir delivery) to find the count
3225 of files (along with the size), even though it did not need this
3226 information. It now does the scan only if it needs to find either the
3227 size of the count of files.
3229 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3231 PH/52 Two bugs concerned with error handling when the smtp transport is
3234 (i) Exim was not creating retry information for temporary errors given
3235 for individual recipients after the DATA command when the smtp transport
3236 was used in LMTP mode. This meant that they could be retried too
3237 frequently, and not timed out correctly.
3239 (ii) Exim was setting the flag that allows error details to be returned
3240 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3241 recipients that were returned after the DATA command.
3243 PH/53 This is related to PH/52, but is more general: for any failing address,
3244 when detailed error information was permitted to be returned to the
3245 sender, but the error was temporary, then after the final timeout, only
3246 "retry timeout exceeded" was returned. Now it returns the full error as
3247 well as "retry timeout exceeded".
3249 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3250 do this, and (what is worse) MTAs that accept it.
3252 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3253 will now be deprecated.
3255 PH/56 New os.c-cygwin from the Cygwin maintainer.
3257 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3258 formats (previously only available in exim formats via -bpr, -bpru,
3259 and -bpra. Now also available in native and exiqgrep formats)
3261 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3262 with very large, slow to parse queues
3264 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3266 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3268 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3269 responses to authentication challenges, though it was showing the
3270 challenges; (ii) I've removed the CR characters from the debug output for
3273 PH/58 Allow for the insertion of a newline as well as a space when a string
3274 is turned into more than one encoded-word during RFC 2047 encoding. The
3275 Sieve code now uses this.
3277 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3278 data_4xx, lost_connection, tls_required.
3280 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3281 message was being sent as an SMTP response.
3283 PH/61 Add -l and -k options to exicyclog.
3285 PH/62 When verifying, if an address was redirected to one new address, so that
3286 verification continued, and the new address failed or deferred after
3287 having set something in $address_data, the value of $address_data was not
3288 passed back to the ACL. This was different to the case when no
3289 redirection occurred. The value is now passed back in both cases.
3291 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3292 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3293 use login_cap.h, so on its own it isn't the distinguishing feature. The
3294 new name refers directly to the setclassresources() function.
3296 PH/65 Added configuration files for NetBSD3.
3298 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3300 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3301 is preferred over IPv4.
3303 PH/68 The bounce_return_message and bounce_return_body options were not being
3304 honoured for bounces generated during the reception of non-SMTP messages.
3305 In particular, this applied to messages rejected by the ACL. This bug has
3306 been fixed. However, if bounce_return_message is true and bounce_return_
3307 body is false, the headers that are returned for a non-SMTP message
3308 include only those that have been read before the error was detected.
3309 (In the case of an ACL rejection, they have all been read.)
3311 PH/69 The HTML version of the specification is now built in a directory called
3312 spec_html instead of spec.html, because the latter looks like a path with
3313 a MIME-type, and this confuses some software.
3315 PH/70 Catch two compiler warnings in sieve.c.
3317 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3318 function verify_get_ident() calls ip_connect() to connect a socket, but
3319 if the "connect()" function timed out, ip_connect() used to close the
3320 socket. However, verify_get_ident() also closes the socket later, and in
3321 between Exim writes to the log, which may get opened at this point. When
3322 the socket was closed in ip_connect(), the log could get the same file
3323 descriptor number as the socket. This naturally causes chaos. The fix is
3324 not to close the socket in ip_connect(); the socket should be closed by
3325 the function that creates it. There was only one place in the code where
3326 this was missing, in the iplookup router, which I don't think anybody now
3327 uses, but I've fixed it anyway.
3329 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3330 well as to direct DNS lookups. Otherwise the handling of names in host
3331 lists is inconsistent and therefore confusing.
3337 PH/01 Two changes to the default runtime configuration:
3339 (1) Move the checks for relay_from_hosts and authenticated clients from
3340 after to before the (commented out) DNS black list checks.
3342 (2) Add control=submission to the relay_from_hosts and authenticated
3343 clients checks, on the grounds that messages accepted by these
3344 statements are most likely to be submissions.
3346 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3348 (1) Generate an error if the third argument for the ${prvs expansion is
3351 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3354 (3) Reset the variables that are obtained from the first argument of
3355 ${prvscheck and used in the second argument before leaving the code,
3356 because their memory is reclaimed, so using them afterwards may do
3359 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3360 one (it's much easier than Tom thought :-).
3362 (5) Because of (4), we can now allow for the use of $prvscheck_result
3363 inside the third argument.
3365 PH/03 For some reason, the default setting of PATH when running a command from
3366 a pipe transport was just "/usr/bin". I have changed it to
3369 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3370 anything to be listed in the output from -bV.
3372 PH/05 When a filter generated an autoreply, the entire To: header line was
3373 quoted in the delivery log line, like this:
3375 => >A.N.Other <ano@some.domain> <original@ddress> ...
3377 This has been changed so that it extracts the operative address. There
3378 may be more than one such address. If so, they are comma-separated, like
3381 => >ano@some.domain,ona@other.domain <original@ddress> ...
3383 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3384 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3385 not being looked up in the rDNS to get a host name, Exim was showing the
3386 IP address twice in Received: lines, even though the IP addresses were
3387 identical. For example:
3389 Received: from [1.2.3.4] (helo=[1.2.3.4])
3391 However, if the real host name was known, it was omitting the HELO data
3392 if it matched the actual IP address. This has been tidied up so that it
3393 doesn't show the same IP address twice.
3395 PH/07 When both +timestamp and +memory debugging was on, the value given by
3396 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3397 called by the debug printing, thereby overwriting the timestamp buffer.
3398 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3400 PH/08 When the original message was included in an autoreply transport, it
3401 always said "this is a copy of the message, including all the headers",
3402 even if body_only or headers_only was set. It now gives an appropriate
3405 PH/09 Applied a patch from the Sieve maintainer which:
3407 o fixes some comments
3408 o adds the (disabled) notify extension core
3409 o adds some debug output for the result of if/elsif tests
3410 o points to the current vacation draft in the documentation
3411 and documents the missing references header update
3415 o fixes a bug in processing the envelope test (when testing
3416 multiple envelope elements, the last element determined the
3419 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3420 Electronic Mail") by including:
3422 Auto-submitted: auto-generated
3424 in the messages that it generates (bounce messages and others, such as
3425 warnings). In the case of bounce messages for non-SMTP messages, there was
3426 also a typo: it was using "Auto_submitted" (underscore instead of
3427 hyphen). Since every message generated by Exim is necessarily in response
3428 to another message, thes have all been changed to:
3430 Auto-Submitted: auto-replied
3432 in accordance with these statements in the RFC:
3434 The auto-replied keyword:
3436 - SHOULD be used on messages sent in direct response to another
3437 message by an automatic process,
3439 - MUST NOT be used on manually-generated messages,
3441 - MAY be used on Delivery Status Notifications (DSNs) and Message
3442 Disposition Notifications (MDNs),
3444 - MUST NOT be used on messages generated by automatic or periodic
3445 processes, except for messages which are automatic responses to
3448 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3449 to the default Received: header definition.
3451 PH/12 Added log selector acl_warn_skipped (default on).
3453 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3454 variables because (a) they are in the wrong storage pool and (b) even if
3455 they were copied, it wouldn't work properly because of the caching.
3457 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3458 checking when decoding. Apparently there are clients that generate
3459 overlong encoded strings. Why am I not surprised?
3461 PH/15 If the first argument of "${if match_address" was not empty, but did not
3462 contain an "@" character, Exim crashed. Now it writes a panic log message
3463 and treats the condition as false.
3465 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3467 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3468 extension "envelope-auth". The code is finished and in agreement with
3469 other implementations, but there is no documentation so far and in fact,
3470 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3471 not changing the active code.
3473 Print executed "if" and "elsif" statements when debugging is used. This
3474 helps a great deal to understand what a filter does.
3476 Document more things not specified clearly in RFC3028. I had all this
3477 sorted out, when out of a sudden new issues came to my mind. Oops."
3479 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3482 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3483 canonical form (as documented). However, after a host name lookup from
3484 the IP address, check_host() was doing a simple string comparison with
3485 addresses acquired from the DNS when checking that the found name did
3486 have the original IP as one of its addresses. Since any found IPv6
3487 addresses are likely to be in abbreviated form, the comparison could
3488 fail. Luckily, there already exists a function for doing the comparison
3489 by converting both addresses to binary, so now that is used instead of
3490 the text comparison.
3492 PH/20 There was another similar case to PH/19, when a complete host name was
3493 given in a host list; looking up its IP address could give an abbreviated
3494 form, whereas the current host's name might or might not be abbreviated.
3495 The same fix has been applied.
3501 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3502 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3505 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3506 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3508 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3510 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3511 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3512 Also, for "Auto-Submitted", treat anything other than "no" as
3513 non-personal, in accordance with RFC 3834. (Previously it treated
3514 anything starting "auto-" as non-personal.)
3516 TF/01 The control=submission/name=... option had a problem with syntax
3517 errors if the name included a slash character. The /name= option
3518 now slurps the rest of the string, so it can include any characters
3519 but it must come last in the list of options (after /sender_retain
3522 PH/05 Some modifications to the interface to the fake nameserver for the new
3530 TK/01 Added the "success_on_redirect" address verification option. See
3531 NewStuff for rationale and an example.
3533 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3535 PH/02 Patch to exigrep to allow it to work on syslog lines.
3537 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3538 fread() to skip over the body file's header line, because in Cygwin the
3539 header line is locked and is inaccessible.
3541 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3542 co-exist for some time) to make it clear that it is the Exim ID that is
3543 referenced, not the Message-ID: header line.
3545 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3546 string_format() function, because snprintf() does not exist on all
3549 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3550 ${stat: expansion item.
3552 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3553 protocol synchronization error", to keep the pedants happy.
3555 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3556 well as for IRIX systems, when gcc is being used. See the host.c source
3559 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3561 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3564 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3565 transport and to the smtp transport in LMTP mode.
3567 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3569 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3570 run a filter in a subprocess. This could lead to confusion in subsequent
3571 lookups in the parent process. There should also be a search_tidyup() at
3572 the end of the subprocess.
3574 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3575 only if the host matched helo_try_verify_hosts, which caused the
3576 verification to occur when the EHLO/HELO command was issued. The ACL just
3577 tested the remembered result. Now, if a previous verification attempt has
3578 not happened, "verify = helo" does it there and then.
3580 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3582 TK/03 Fix log output including CR from clamd.
3584 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3585 exist provoked a memory error which could cause a segfault.
3587 PH/15 Installed PCRE 6.2
3589 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3591 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3592 of the problem. Specifically, suggested +O2 rather than +O1 for the
3595 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3597 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3598 record, the port was not correctly logged when the outgoing_port log
3599 selector was set (it logged the transort's default port).
3601 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3602 fallback_hosts, and "hosts" in the smtp transport.
3604 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3605 host errors such as "Connection refused".
3607 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3608 authentication with radiusclient 0.4.9:
3610 - Error returned from rc_read_config was caught wrongly
3611 - Username/password not passed on to radius server due to wrong length.
3613 The presumption is that some radiusclient API changes for 4.51/PH/17
3614 were not taken care of correctly. The code is still untested by me (my
3615 Linux distribution still has 0.3.2 of radiusclient), but it was
3616 contributed by a Radius user.
3618 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3619 expanding the "port" option of the smtp transport.
3621 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3622 while reading a MIME header. Thanks to Tom Hughes for a patch.
3624 PH/24 Include config.h inside local_scan.h so that configuration settings are
3627 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3628 that in an ACL for RCPT (for example), you can examine exactly what was
3631 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3632 commands, but it was not correctly comparing the address with the actual
3633 client host address. Thus, it would show the EHLO address in Received:
3634 header lines when this was not necessary.
3636 PH/27 Added the % operator to ${eval:}.
3638 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3639 it should be ignoring failures (because with -C, for example, it has lost
3640 privilege). It wasn't ignoring creation failures other than "already
3643 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3646 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3647 because an input line was too long, either on its own, or by virtue of
3648 too many continuations, the temporary file was not being removed, and the
3649 return code was incorrect.
3651 PH/31 Missing "BOOL" in function definition in filtertest.c.
3653 PH/32 Applied Sieve patches from the maintainer.
3655 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3657 PH/33 Added "verify = not_blind".
3659 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3660 Local/Makefile (with some defaults set). These are used in built scripts
3661 such as exicyclog, but they have never been used in the exim_install
3662 script (though there are many overriding facilities there). I have
3663 arranged that the exim_install script now takes note of these two
3666 PH/35 Installed configuration files for Dragonfly.
3668 PH/36 When a locally submitted message by a trusted user did not contain a
3669 From: header, and the sender address was obtained from -f or from an SMTP
3670 MAIL command, and the trusted user did not use -F to supply a sender
3671 name, $originator_name was incorrectly used when constructing a From:
3672 header. Furthermore, $originator_name was used for submission mode
3673 messages from external hosts without From: headers in a similar way,
3674 which is clearly wrong.
3676 PH/37 Added control=suppress_local_fixups.
3678 PH/38 When log_selector = +received_sender was set, and the addition of the
3679 sender made the log line's construction buffer exactly full, or one byte
3680 less than full, an overflow happened when the terminating "\n" was
3683 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3684 when the result of a list match is failure because a DNS lookup failed.
3686 PH/40 RM_COMMAND is now used in the building process.
3688 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3689 the "build-* directories that it finds.
3691 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3692 address in a domain literal was a prefix of an interface address.
3694 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3695 when verifying a sender address, unless rewrite_headers is false.
3697 PH/44 Wrote a long comment about why errors_to addresses are verified as
3698 recipients, not senders.
3700 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3701 the ratelimit ACL was added.
3703 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3705 PH/46 Added extra information about PostgreSQL errors to the error string.
3707 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3708 avoiding the need to install special zones in a real server. This is
3709 backwards compatible; if it can't find the fake resolver, it drops back.
3710 Thus, both old and new test suites can be run.
3712 TF/02 Added util/ratelimit.pl
3714 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3715 clock is set back in time.
3717 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3718 Candler <B.Candler@pobox.com>.
3720 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3721 OK for addresses that are the result of redirections.
3723 PH/48 A number of further additions for the benefit of the new test suite,
3724 including a fake gethostbyname() that interfaces to the fake DNS resolver
3727 TF/06 The fix for widen_domains has also been applied to qualify_single and
3728 search_parents which are the other dnslookup options that can cause
3731 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3734 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3736 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3737 service" if it encountered an MX record with an empty target hostname.
3738 The message is now "an MX or SRV record indicated no SMTP service".
3740 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3741 if the DNS of the sending site is misconfigured. This is quite a
3742 common situation. This change restores the behaviour of treating a
3743 helo verification defer as a failure.
3745 PH/51 If self=fail was set on a router, the bounce message did not include the
3746 actual error message.
3752 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3754 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3755 command itself ended in error, the underlying message about the transport
3756 filter timeout was being overwritten with the pipe command error. Now the
3757 underlying error message should be appended to the second error message.
3759 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3761 PH/02 Reduce the amount of output that "make" produces by default. Full output
3762 can still be requested.
3764 PH/03 The warning log line about a condition test deferring for a "warn" verb
3765 was being output only once per connection, rather than after each
3766 occurrence (because it was using the same function as for successful
3767 "warn" verbs). This seems wrong, so I have changed it.
3769 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3770 it should not have, which might have caused a crash in the right
3771 circumstances, but probably never did.
3773 PH/04 Installed a modified version of Tony Finch's patch to make submission
3774 mode fix the return path as well as the Sender: header line, and to
3775 add a /name= option so that you can make the user's friendly name appear
3778 TF/03 Added the control = fakedefer ACL modifier.
3780 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3781 Mark Lowes for thorough testing.
3783 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3785 TK/03 Merged latest SRS patch from Miles Wilton.
3787 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3788 with the definition in sysexits.h (which is #included earlier).
3789 Fortunately, Exim does not actually use EX_OK. The code used to try to
3790 preserve the sysexits.h value, by assuming that macro definitions were
3791 scanned for macro replacements. I have been disabused of this notion,
3792 so now the code just undefines EX_OK before #including unistd.h.
3794 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3795 in the smtp transport. When a block could not be written in a single
3796 write() function, the timeout was being re-applied to each part-write.
3797 This seems wrong - if the receiver was accepting one byte at a time it
3798 would take for ever. The timeout is now adjusted when this happens. It
3799 doesn't have to be particularly precise.
3801 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3802 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3804 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3805 without a domain if the check to <postmaster@domain> fails.
3807 SC/01 Eximstats: added -xls and the ability to specify output files
3808 (patch written by Frank Heydlauf).
3810 SC/02 Eximstats: use FileHandles for outputting results.
3812 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3814 SC/04 Eximstats: fixed display of large numbers with -nvr option
3816 SC/05 Eximstats: fixed merging of reports with empty tables.
3818 SC/06 Eximstats: added the -include_original_destination flag
3820 SC/07 Eximstats: removed tabs and trailing whitespace.
3822 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3824 TK/06 MBOX spool code: Add real "From " MBOX separator line
3825 so the .eml file is really in mbox format (even though
3826 most programs do not really care). Patch from Alex Miller.
3828 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3829 The latter is generated from $received_to and is only set if the
3830 message has one envelope recipient. SA can use these headers,
3831 obviously out-of-the-box. Patch from Alex Miller.
3833 PH/08 The ${def test on a variable was returning false if the variable's
3834 value was "0", contrary to what the specification has always said!
3835 The result should be true unless the variable is empty.
3837 PH/09 The syntax error of a character other than { following "${if
3838 def:variable_name" (after optional whitespace) was not being diagnosed.
3839 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3840 accidental colon was present, for example, could give incorrect results.
3842 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3843 result is used (not including appendfile, where other changes are about
3846 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3847 This involved changing a lot of size variables from int to off_t. It
3848 should work with maildirs and everything.
3850 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3851 spamd dying while we are connected to it.
3853 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3856 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3857 function for BDB 4.3.
3859 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3861 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3862 the sources for PCRE have been re-arranged and more files are now
3865 PH/15 The code I had for printing potentially long long variables in PH/11
3866 above was not the best (it lost precision). The length of off_t variables
3867 is now inspected at build time, and an appropriate printing format (%ld
3868 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3869 to be "long long int" or "long int". This is needed for the internal
3870 formatting function string_vformat().
3872 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3873 the configuration file to be ":syslog", then the script "guesses" where
3874 the logs files are, rather than using the compiled in default. In our
3875 case the guess is not the same as the compiled default, so the script
3876 suddenly stopped working when I started to use syslog. The patch checks
3877 to see if log_file_path is "". If so, it attempts to read it from exim
3878 with no configuration file to get the compiled in version, before it
3879 falls back to the previous guessing code."
3881 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3882 implementing BATV in an Exim configuration. See NewStuff for the gory
3885 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3886 Makefile that are specific to HP-UX.
3888 PH/18 If the "use_postmaster" option was set for a recipient callout together
3889 with the "random" option, the postmaster address was used as the MAIL
3890 FROM address for the random test, but not for the subsequent recipient
3891 test. It is now used for both.
3893 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3894 patch removes a few documentation additions to RFC 3028, because the
3895 latest draft now contains them. It adds the new en;ascii-case comparator
3896 and a new error check for 8bit text in MIME parts. Comparator and
3897 require names are now matched exactly. I enabled the subaddress
3898 extension, but it is not well tested yet (read: it works for me)."
3900 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3901 rework some of the code of TK/09 above to avoid the hardwired use of
3902 "%lld" and "long long". Replaced the call to snprintf() with a call to
3905 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3906 records point to non-existent hosts", "retry timeout exceeded", and
3907 "retry time not reached for any host after a long failure period".
3909 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3910 experimental DomainKeys support:
3912 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3913 (2) On an error such as an illegally used "control", the wrong name for
3914 the control was given.
3916 These problems did NOT occur unless DomainKeys support was compiled.
3918 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3920 PH/24 Added ${if match_ip condition.
3922 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3923 databases so that it will be absolutely obvious if a crash occurs in the
3924 DB library. This is a regular occurrence (often caused by mis-matched
3927 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3928 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3929 picked up on a user's system that detects such things. There doesn't seem
3930 to be a gcc warning option for this - only an attribute that has to be
3931 put on the function's prototype. It seems that in Fedora Core 4 they have
3932 set this on a number of new functions. No doubt there will be more in due
3935 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3936 specify a transport. However, if an address that was verified by such a
3937 router was the subject of a callout, Exim crashed because it tried to
3938 read the rcpt_include_affixes from the non-existent transport. Now it
3939 just assumes that the setting of that option is false. This bug was
3940 introduced by 4.51/PH/31.
3942 PH/28 Changed -d+all to exclude +memory, because that information is very
3943 rarely of interest, but it makes the output a lot bigger. People tend to
3944 do -d+all out of habit.
3946 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3947 code in os-type was giving problems when libc.so lives in lib64, like on
3950 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3951 aren't the modern standard, and it seems that some systems' include files
3952 don't always have them. Exim was already checking for some of the newer
3953 ones like T_AAAA, and defining it itself. I've added checks for all the
3954 record types that Exim uses.
3956 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3957 not automatically generating a new one, as it is supposed to. This
3958 prevented TLS from working. If the file did exist, but contained invalid
3959 data, a new version was generated, as expected. It was only the case of a
3960 non-existent file that was broken.
3962 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3963 with a change in libdomainkeys > 0.64.
3965 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3966 from DNS. If the selector record carries the flag, it now has
3967 precedence over the domain-wide flag.
3969 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3971 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3972 the use of a port name that isn't defined in /etc/services) occurred, the
3973 message was deferred as in a normal delivery, and thus remained on the
3974 spool, instead of being failed because of the mua_wrapper setting. This
3975 is now fixed, and I tidied up some of the mua_wrapper messages at the
3978 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3979 the messages in a hash of arrays rather than using individual hashes.
3980 This is a bit cleaner and results in dramatic memory savings, albeit
3981 at a slight CPU cost.
3983 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3984 as requested by Marc Sherman.
3986 SC/10 Eximstats: added histograms for user specified patterns as requested
3989 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3991 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3992 fopen() in the content-scanning modules that did not already have it.
3998 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3999 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4001 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4003 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4005 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4006 to test Sieve filters that use "vacation".
4008 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4009 that changes the way the GnuTLS parameters are stored in the cache file.
4010 The new format can be generated externally. For backward compatibility,
4011 if the data in the cache doesn't make sense, Exim assumes it has read an
4012 old-format file, and it generates new data and writes a new file. This
4013 means that you can't go back to an older release without removing the
4016 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4017 work if there are any delivery delays because "one_time" forces the
4018 parent to be marked "delivered", so its unseen clone is never tried
4019 again. For this reason, Exim now forbids the simultaneous setting of
4022 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4023 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4024 entry if you want to know the details. The fix, however, neglected to
4025 consider the case when local delivery batching is involved. The test for
4026 "previously delivered" was not happening when checking to see if an
4027 address could be batched with a previous (undelivered) one; under
4028 certain circumstances this could lead to multiple deliveries to the same
4031 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4032 in its include files, and this causes problems building Exim.
4034 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4035 header_syntax) but Exim was just ignoring anything given after a slash.
4036 In particular, this caused confusion with an attempt to use "verify =
4037 reverse_host_lookup/defer_ok". An error is now given when options are
4038 supplied for verify items that do not have them. (Maybe reverse_host_
4039 lookup should have a defer_ok option, but that's a different point.)
4041 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4042 defined by RFC 821) to 2048, because there were problems with some AUTH
4043 commands, and RFC 1869 says the size should be increased for extended
4044 SMTP commands that take arguments.
4046 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4049 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4050 "unknown" error; now it says that the functionality isn't in the binary.
4052 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4053 an address' error message when a string expansion fails (syntax or
4054 whatever). Otherwise the password may appear in the log. Following change
4055 PH/42 below, there is no longer a chance of it appearing in a bounce
4058 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4060 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4061 of its addresses was ever tried. (Bugzilla bug #2.)
4063 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4064 the result incorrectly in the debug output. (It correctly added a newline
4065 to what was transported.)
4067 TF/01 Added $received_time.
4069 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4070 commented out examples of how to interface to a virus scanner and to
4071 SpamAssassin. Also added commented examples of av_scanner and
4072 spamd_address settings.
4074 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4075 and controls are allowed in which ACLs. There were a couple of minor
4076 errors. Some of the entries in the conditions table (which is a table of
4077 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4078 negation of where the condition IS allowed.
4080 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4082 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4083 header file does not have a version number, so I've had to invent a new
4084 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4085 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4086 radiusclient), but it was contributed by a Radius user.
4088 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4089 files or directories, for OpenSSL.
4091 PH/19 When an Exim process that is running as root has to create an Exim log
4092 file, it does so in a subprocess that runs as exim:exim so as to get the
4093 ownership right at creation (otherwise, other Exim processes might see
4094 the file with the wrong ownership). There was no test for failure of this
4095 fork() call, which would lead to the process getting stuck as it waited
4096 for a non-existent subprocess. Forks do occasionally fail when resources
4097 run out. I reviewed all the other calls to fork(); they all seem to check
4100 PH/20 When checking for unexpected SMTP input at connect time (before writing
4101 the banner), Exim was not dealing correctly with a non-positive return
4102 from the read() function. If the client had disconnected by this time,
4103 the result was a log entry for a synchronization error with an empty
4104 string after "input=" when read() returned zero. If read() returned -1
4105 (an event I could not check), uninitialized data bytes were printed.
4106 There were reports of junk text (parts of files, etc) appearing after
4109 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4111 PH/22 Added support for macro redefinition, and (re)definition in between
4112 driver and ACL definitions.
4114 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4115 forgetting to use the resulting value; it was using the unexpanded value.
4117 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4118 hadn't been configured. The fix is from Juergen Kreileder, who
4119 understands it better than I do:
4121 "Here's what I see happening with three configured cyrus_sasl
4122 authenticators configured (plain, login, cram-md5):
4124 On startup auth_cyrus_sasl_init() gets called for each of these.
4125 This means three calls to sasl_listmech() without a specified mech_list.
4126 => SASL tests which mechs of all available mechs actually work
4127 => three warnings about OTP not working
4128 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4130 With the patch, sasl_listmech() also gets called three times. But now
4131 SASL's mech_list option is set to the server_mech specified in the the
4132 authenticator. Or in other words, the answer from sasl_listmech()
4133 gets limited to just the mech you're testing for (which is different
4135 => the return list contains just 'plain' or 'login', 'cram-md5' or
4136 nothing depending on the value of ob->server_mech.
4138 I've just tested the patch: Authentication still works fine,
4139 unavailable mechs specified in the exim configuration are still
4140 caught, and the auth.log warnings about OTP are gone."
4142 PH/25 When debugging is enabled, the contents of the command line are added
4143 to the debugging output, even when log_selector=+arguments is not
4146 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4147 answer is "GNU", and only if the return is "GNU/something" is the answer
4150 PH/27 $acl_verify_message is now set immediately after the failure of a
4151 verification in an ACL, and so is available in subsequent modifiers. In
4152 particular, the message can be preserved by coding like this:
4154 warn !verify = sender
4155 set acl_m0 = $acl_verify_message
4157 Previously, $acl_verify_message was set only while expanding "message"
4158 and "log_message" when a very denied access.
4160 PH/28 Modified OS/os.c-Linux with
4162 -#ifndef OS_LOAD_AVERAGE
4163 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4165 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4166 nomenclature these days.)
4168 PH/29 Installed patch from the Sieve maintainer that adds the options
4169 sieve_useraddress and sieve_subaddress to the redirect router.
4171 PH/30 In these circumstances:
4172 . Two addresses routed to the same list of hosts;
4173 . First host does not offer TLS;
4174 . First host accepts first address;
4175 . First host gives temporary error to second address;
4176 . Second host offers TLS and a TLS session is established;
4177 . Second host accepts second address.
4178 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4179 and peerdn, if requested) that were in fact used only for the second
4182 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4183 attention to any local part prefix or suffix that was matched by the
4184 router that accepted the address. It now behaves in the same way as it
4185 does for delivery: the affixes are removed from the local part unless
4186 rcpt_include_affixes is set on the transport.
4188 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4189 timeout during the DATA phase of an incoming message.
4191 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4192 applied a patch sanctioned by the Sieve maintainer.
4194 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4195 the uid or gid is negative. A case of a negative gid caused this to be
4196 noticed. The fix allows for either to be negative.
4198 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4199 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4202 PH/36 The change PH/12 above was broken. Fixed it.
4204 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4205 the grounds that routing the same address twice would always produce the
4206 same answer. This might have been true once, but it is certainly no
4207 longer true now. Routing a child address may depend on the previous
4208 routing that produced that child. Some complicated redirection strategies
4209 went wrong when messages had multiple recipients, and made Exim's
4210 behaviour dependent on the order in which the addresses were given.
4212 I have moved the duplicate checking until after the routing is complete.
4213 Exim scans the addresses that are assigned to local and remote
4214 transports, and removes any duplicates. This means that more work will be
4215 done, as duplicates will always all be routed, but duplicates are
4216 presumably rare, so I don't expect this is of any significance.
4218 For deliveries to pipes, files, and autoreplies, the duplicate checking
4219 still happens during the routing process, since they are not going to be
4222 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4223 It corrects a timeout issue with spamd. This is Ian's comment: "The
4224 background is that sometimes spamd either never reads data from a
4225 connection it has accepted, or it never writes response data. The exiscan
4226 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4227 blindly assumes that writes won't block so it may never time out."
4229 PH/39 Allow G after quota size as well as K and M.
4231 PH/40 The value set for $authenticated_id in an authenticator may not contain
4232 binary zeroes or newlines because the value is written to log lines and
4233 to spool files. There was no check on this. Now the value is run through
4234 the string_printing() function so that such characters are converted to
4235 printable escape sequences.
4237 PH/41 $message_linecount is a new variable that contains the total number of
4238 lines in the message. Compare $body_linecount, which is the count for the
4241 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4242 bounce and delay warning messages, except in certain special cases, which
4245 (a) An SMTP error message from a remote host;
4246 (b) A message specified in a :fail: redirection;
4247 (c) A message specified in a "fail" command in a system filter;
4248 (d) A message specified in a FAIL return from the queryprogram router;
4249 (e) A message specified by the cannot_route_message router option.
4251 In these cases only, Exim does include the error details in bounce and
4252 warning messages. There are also a few cases where bland messages such
4253 as "unrouteable address" or "local delivery error" are given.
4255 PH/43 $value is now also set for the "else" part of a ${run expansion.
4257 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4258 being worked on, but at least Exim now implements the latest version to
4261 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4262 process to complete was treated as a delivery failure, a timeout while
4263 writing the message to the pipe was logged, but erroneously treated as a
4264 successful delivery. Such timeouts include transport filter timeouts. For
4265 consistency with the overall process timeout, these timeouts are now
4266 treated as errors, giving rise to delivery failures by default. However,
4267 there is now a new Boolean option for the pipe transport called
4268 timeout_defer, which, if set TRUE, converts the failures into defers for
4269 both kinds of timeout. A transport filter timeout is now identified in
4272 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4273 systems where "make" and "gmake" are different, calling "gmake" at top
4274 level broke things. I've arranged for the value of $(MAKE) to be passed
4275 from the Makefile to this script so that it can call the same version of
4279 A note about Exim versions 4.44 and 4.50
4280 ----------------------------------------
4282 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4283 changes of various kinds. As a consequence, a big documentation update was
4284 needed. This delayed the release for rather longer than seemed good, especially
4285 in the light of a couple of (minor) security issues. Therefore, the changes
4286 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4287 release. So 4.44 and 4.50 are in effect two different branches that both start
4290 I have left the 4.50 change log unchanged; it contains all the changes since
4291 4.43. The change log for 4.44 is below; many of its items are identical to
4292 those for 4.50. This seems to be the most sensible way to preserve the
4293 historical information.
4299 1. Minor wording change to the doc/README.SIEVE file.
4301 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4302 computation of the current number of files was incorrect.
4304 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4305 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4308 4. Give more explanation in the error message when the command for a transport
4309 filter fails to execute.
4311 5. There are several places where Exim runs a non-Exim command in a
4312 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4313 was being done only for the command run by the queryprogram router. It is
4314 now done for all such subprocesses. The other cases are: ${run, transport
4315 filters, and the commands run by the lmtp and pipe transports.
4317 6. Added CONFIGURE_GROUP build-time option.
4319 7. Some older OS have a limit of 256 on the maximum number of file
4320 descriptors. Exim was using setrlimit() to set 1000 as a large value
4321 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4322 systems. I've change it so that if it can't get 1000, it tries for 256.
4324 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4325 was an oversight, and furthermore, ever since the addition of extra
4326 controls (e.g. 4.43/32), the checks on when to allow different forms of
4327 "control" were broken. There should now be diagnostics for all cases when a
4328 control that does not make sense is encountered.
4330 9. Added the /retain_sender option to "control=submission".
4332 10. $recipients is now available in the predata ACL (oversight).
4334 11. Tidy the search cache before the fork to do a delivery from a message
4335 received from the command line. Otherwise the child will trigger a lookup
4336 failure and thereby defer the delivery if it tries to use (for example) a
4337 cached ldap connection that the parent has called unbind on.
4339 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4340 of $address_data from the recipient verification was clobbered by the
4341 sender verification.
4343 13. The value of address_data from a sender verification is now available in
4344 $sender_address_data in subsequent conditions in the ACL statement.
4346 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4348 15. Added a new option "connect=<time>" to callout options, to set a different
4351 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4352 was its contents. (It was OK if the option was not defined at all.)
4354 17. A "Completed" log line is now written for messages that are removed from
4355 the spool by the -Mrm option.
4357 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4358 information about exactly what failed.
4360 19. Added -dd to debug only the daemon process.
4362 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4363 handles timeouts, both on the server side and network timeouts. Renamed the
4364 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4366 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4367 It is now set to "smtps".
4369 22. $host_address is now set to the target address during the checking of
4370 ignore_target_hosts.
4372 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4373 being passed; this would have caused $sender_host_name to have been used if
4374 matching the list had actually called for a host name (not very likely,
4375 since this list is usually IP addresses). A host name is now passed as
4378 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4379 code that specifies a non-restarting handler (typically sigaction() in
4380 modern systems) in an attempt to fix a rare and obscure crash bug.
4382 25. Narrowed the window for a race in the daemon that could cause it to ignore
4383 SIGCHLD signals. This is not a major problem, because they are used only to
4384 wake it up if nothing else does.
4386 26. A malformed maildirsize file could cause Exim to calculate negative values
4387 for the mailbox size or file count. Odd effects could occur as a result.
4388 The maildirsize information is now recalculated if the size or filecount
4391 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4392 support for a long time. Removed HAVE_SYS_VFS_H.
4394 28. Installed the latest version of exipick from John Jetmore.
4396 29. In an address list, if the pattern was not a regular expression, an empty
4397 subject address (from a bounce message) matched only if the pattern was an
4398 empty string. Non-empty patterns were not even tested. This was the wrong
4399 because it is perfectly reasonable to use an empty address as part of a
4400 database query. An empty address is now tested by patterns that are
4401 lookups. However, all the other forms of pattern expect the subject to
4402 contain a local part and a domain, and therefore, for them, an empty
4403 address still always fails if the pattern is not itself empty.
4405 30. Exim went into a mad DNS loop when attempting to do a callout where the
4406 host was specified on an smtp transport, and looking it up yielded more
4407 than one IP address.
4409 31. Re-factored the code for checking spool and log partition space into a
4410 function that finds that data and another that does the check. The former
4411 is then used to implement four new variables: $spool_space, $log_space,
4412 $spool_inodes, and $log_inodes.
4414 32. The RFC2047 encoding function was originally intended for short strings
4415 such as real names; it was not keeping to the 75-character limit for
4416 encoded words that the RFC imposes. It now respects the limit, and
4417 generates multiple encoded words if necessary. To be on the safe side, I
4418 have increased the buffer size for the ${rfc2047: expansion operator from
4421 33. It is now permitted to omit both strings after an "if" condition; if the
4422 condition is true, the result is "true". As before, when the second string
4423 is omitted, a false condition yields an empty string. This makes it less
4424 cumbersome to write custom ACL and router conditions.
4426 34. Failure to deliver a bounce message always caused it to be frozen, even if
4427 there was an errors_to setting on the router. The errors_to setting is now
4430 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4431 canonical form (fully expanded) before being placed in
4432 $sender_host_address.
4434 36. The table in the code that translates DNS record types into text (T_A to
4435 "A" for instance) was missing entries for NS and CNAME. It is just possible
4436 that this could have caused confusion if both these types were looked up
4437 for the same domain, because the text type is used as part of Exim's
4438 per-process caching. But the chance of anyone hitting this buglet seems
4441 37. The dnsdb lookup has been extended in a number of ways.
4443 (1) There is a new type, "zns", which walks up the domain tree until it
4444 finds some nameserver records. It should be used with care.
4446 (2) There is a new type, "mxh", which is like "mx" except that it returns
4447 just the host names, not the priorities.
4449 (3) It is now possible to give a list of domains (or IP addresses) to be
4450 looked up. The behaviour when one of the lookups defers can be
4451 controlled by a keyword.
4453 (4) It is now possible to specify the separator character for use when
4454 multiple records are returned.
4456 38. The dnslists ACL condition has been extended: it is now possible to supply
4457 a list of IP addresses and/or domains to be looked up in a particular DNS
4460 39. Added log_selector=+queue_time_overall.
4462 40. When running the queue in the test harness, wait just a tad after forking a
4463 delivery process, to get repeatability of debugging output.
4465 41. Include certificate and key file names in error message when GnuTLS fails
4466 to set them up, because the GnuTLS error message doesn't include the name
4467 of the failing file when there is a problem reading it.
4469 42. Allow both -bf and -bF in the same test run.
4471 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4473 44. The "Exiscan patch" is now merged into the mainline Exim source.
4475 45. Sometimes the final signoff response after QUIT could fail to get
4476 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4477 < 0 before doing a fflush(). This bug looks as though it goes back to the
4478 introduction of TLS in release 3.20, but "sometimes" must have been rare
4479 because the tests only now provoked it.
4481 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4482 (this can affect the format of dates).
4484 47. exim_tidydb, when checking for the continued existence of a message for
4485 which it has found a message-specific retry record, was not finding
4486 messages that were in split spool directories. Consequently, it was
4487 deleting retry records that should have stayed in existence.
4489 48. Steve fixed some bugs in eximstats.
4491 49. The SPA authentication driver was not abandoning authentication and moving
4492 on to the next authenticator when an expansion was forced to fail,
4493 contradicting the general specification for all authenticators. Instead it
4494 was generating a temporary error. It now behaves as specified.
4496 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4497 (the order specifies the preference for clients). The order is now AES256,
4498 AES128, 3DES, ARCFOUR128.
4500 51. Small patch to Sieve code - explicitly set From: when generating an
4503 52. Exim crashed if a remote delivery caused a very long error message to be
4504 recorded - for instance if somebody sent an entire SpamAssassin report back
4505 as a large number of 550 error lines. This bug was coincidentally fixed by
4506 increasing the size of one of Exim's internal buffers (big_buffer) that
4507 happened as part of the Exiscan merge. However, to be on the safe side, I
4508 have made the code more robust (and fixed the comments that describe what
4511 53. Now that there can be additional text after "Completed" in log lines (if
4512 the queue_time_overall log selector is set), a one-byte patch to exigrep
4513 was needed to allow it to recognize "Completed" as not the last thing in
4516 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4517 patch that reportedly fixes this has been added. I am not expert enough to
4518 create a test for it. This is what the patch creator wrote:
4520 "I found a little strange behaviour of ldap code when working with
4521 Windows 2003 AD Domain, where users was placed in more than one
4522 Organization Units. When I tried to give exim partial DN, the exit code
4523 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4524 But simultaneously result of request was absolutely normal ldap result,
4525 so I produce this patch..."
4527 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4528 REFERENCE, so I have modified the code to exclude the patch when that macro
4531 55. Some experimental protocols are using DNS PTR records for new purposes. The
4532 keys for these records are domain names, not reversed IP addresses. The
4533 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4534 leaves it alone. Component reversal etc. now happens only for IP addresses.
4537 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4539 57. Double the size of the debug message buffer (to 2048) so that more of very
4540 long debug lines gets shown.
4542 58. The exicyclog utility now does better if the number of log files to keep
4543 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4545 59. Two changes related to the smtp_active_hostname option:
4547 (1) $smtp_active_hostname is now available as a variable.
4548 (2) The default for smtp_banner uses $smtp_active_hostname instead
4549 of $primary_hostname.
4551 60. The host_aton() function is supposed to be passed a string that is known
4552 to be a valid IP address. However, in the case of IPv6 addresses, it was
4553 not checking this. This is a hostage to fortune. Exim now panics and dies
4554 if the condition is not met. A case was found where this could be provoked
4555 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4556 components; fortuitously, this particular loophole had already been fixed
4557 by change 4.50/55 above.
4559 If there are any other similar loopholes, the new check in host_aton()
4560 itself should stop them being exploited. The report I received stated that
4561 data on the command line could provoke the exploit when Exim was running as
4562 exim, but did not say which command line option was involved. All I could
4563 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4564 running as the user.
4567 61. There was a buffer overflow vulnerability in the SPA authentication code
4568 (which came originally from the Samba project). I have added a test to the
4569 spa_base64_to_bits() function which I hope fixes it.
4572 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4573 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4575 63. The daemon start-up calls getloadavg() while still root for those OS that
4576 need the first call to be done as root, but it missed one case: when
4577 deliver_queue_load_max is set with deliver_drop_privilege. This is
4578 necessary for the benefit of the queue runner, because there is no re-exec
4579 when deliver_drop_privilege is set.
4581 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4582 This has been fixed.
4584 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4585 was in use, was not putting the data itself into the right store pool;
4586 consequently, it could be overwritten for a subsequent message in the same
4587 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4590 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4592 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4593 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4594 the function were treating the return as a boolean value, which happened to
4595 work because 0=false and not-0=true, but is not correct code.
4597 68. The host_aton() function was not handling scoped IPv6 addresses (those
4598 with, for example, "%eth0" on the end) correctly.
4600 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4601 negated items (that is, ~something) in unsigned ints. Some compilers
4602 apparently mutter when there is no cast.
4604 70. If an address verification called from an ACL failed, and did not produce a
4605 user-specific message (i.e. there was only a "system" message), nothing was
4606 put in $acl_verify_message. In this situation, it now puts the system
4609 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4610 session; change 4.31/43 added the unwanted input to the log line - except
4611 that it did not do this in the start of session case. It now does.
4613 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4614 This is wrong and can cause the other end to generate a synchronization
4615 error if it is another Exim or anything else that does the synchronization
4616 check. A QUIT command is no longer sent after a timeout.
4618 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4619 during host lookups.
4621 74. The defer_ok option of callout verification was not working if it was used
4622 when verifying addresses in header lines, that is, for this case:
4624 verify = header_sender/callout=defer_ok
4626 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4627 those file descriptors could be used for SMTP connections. If anything
4628 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4629 could be sent to the SMTP client, causing chaos. The daemon now opens
4630 stdin, stdout, and stderr to /dev/null when it puts itself into the
4633 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4634 log by default. The user can override this with suitable Perl magic.
4636 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4637 the log message when discard triggers, was not working for the DATA ACL or
4638 for the non-SMTP ACL.
4640 78. Error message wording change in sieve.c.
4642 79. If smtp_accept_max_per_host was set, the number of connections could be
4643 restricted to fewer than expected, because the daemon was trying to set up
4644 a new connection before checking whether the processes handling previous
4645 connections had finished. The check for completed processes is now done
4646 earlier. On busy systems, this bug wouldn't be noticed because something
4647 else would have woken the daemon, and it would have reaped the completed
4650 80. If a message was submitted locally by a user whose login name contained one
4651 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4652 It caused a spool format error. I have fixed the spool reading code. A
4653 related problem was that the "from" clause in the Received: line became
4654 illegal because of the space(s). It is now covered by ${quote_local_part.
4656 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4659 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4661 83. Patch from Sieve maintainer to fix unterminated string problem in
4662 "vacation" handling.
4664 84. Some minor changes to the Linux configuration files to help with other
4665 OS variants using glibc.
4667 85. One more patch for Sieve to update vacation handling to latest spec.
4670 ----------------------------------------------------
4671 See the note above about the 4.44 and 4.50 releases.
4672 ----------------------------------------------------
4678 1. Change 4.43/35 introduced a bug that caused file counts to be
4679 incorrectly computed when quota_filecount was set in an appendfile
4682 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4683 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4686 3. Give more explanation in the error message when the command for a transport
4687 filter fails to execute.
4689 4. There are several places where Exim runs a non-Exim command in a
4690 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4691 was being done only for the command run by the queryprogram router. It is
4692 now done for all such subprocesses. The other cases are: ${run, transport
4693 filters, and the commands run by the lmtp and pipe transports.
4695 5. Some older OS have a limit of 256 on the maximum number of file
4696 descriptors. Exim was using setrlimit() to set 1000 as a large value
4697 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4698 systems. I've change it so that if it can't get 1000, it tries for 256.
4700 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4701 was an oversight, and furthermore, ever since the addition of extra
4702 controls (e.g. 4.43/32), the checks on when to allow different forms of
4703 "control" were broken. There should now be diagnostics for all cases when a
4704 control that does not make sense is encountered.
4706 7. $recipients is now available in the predata ACL (oversight).
4708 8. Tidy the search cache before the fork to do a delivery from a message
4709 received from the command line. Otherwise the child will trigger a lookup
4710 failure and thereby defer the delivery if it tries to use (for example) a
4711 cached ldap connection that the parent has called unbind on.
4713 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4714 of $address_data from the recipient verification was clobbered by the
4715 sender verification.
4717 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4718 was its contents. (It was OK if the option was not defined at all.)
4720 11. A "Completed" log line is now written for messages that are removed from
4721 the spool by the -Mrm option.
4723 12. $host_address is now set to the target address during the checking of
4724 ignore_target_hosts.
4726 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4727 being passed; this would have caused $sender_host_name to have been used if
4728 matching the list had actually called for a host name (not very likely,
4729 since this list is usually IP addresses). A host name is now passed as
4732 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4733 code that specifies a non-restarting handler (typically sigaction() in
4734 modern systems) in an attempt to fix a rare and obscure crash bug.
4736 15. Narrowed the window for a race in the daemon that could cause it to ignore
4737 SIGCHLD signals. This is not a major problem, because they are used only to
4738 wake it up if nothing else does.
4740 16. A malformed maildirsize file could cause Exim to calculate negative values
4741 for the mailbox size or file count. Odd effects could occur as a result.
4742 The maildirsize information is now recalculated if the size or filecount
4745 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4746 support for a long time. Removed HAVE_SYS_VFS_H.
4748 18. Updated exipick to current release from John Jetmore.
4750 19. Allow an empty sender to be matched against a lookup in an address list.
4751 Previously the only cases considered were a regular expression, or an
4754 20. Exim went into a mad DNS lookup loop when doing a callout where the
4755 host was specified on the transport, if the DNS lookup yielded more than
4758 21. The RFC2047 encoding function was originally intended for short strings
4759 such as real names; it was not keeping to the 75-character limit for
4760 encoded words that the RFC imposes. It now respects the limit, and
4761 generates multiple encoded words if necessary. To be on the safe side, I
4762 have increased the buffer size for the ${rfc2047: expansion operator from
4765 22. Failure to deliver a bounce message always caused it to be frozen, even if
4766 there was an errors_to setting on the router. The errors_to setting is now
4769 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4770 canonical form (fully expanded) before being placed in
4771 $sender_host_address.
4773 24. Updated eximstats to version 1.33
4775 25. Include certificate and key file names in error message when GnuTLS fails
4776 to set them up, because the GnuTLS error message doesn't include the name
4777 of the failing file when there is a problem reading it.
4779 26. Expand error message when OpenSSL has problems setting up cert/key files.
4782 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4783 (this can affect the format of dates).
4785 28. exim_tidydb, when checking for the continued existence of a message for
4786 which it has found a message-specific retry record, was not finding
4787 messages that were in split spool directories. Consequently, it was
4788 deleting retry records that should have stayed in existence.
4790 29. eximstats updated to version 1.35
4791 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4792 1.35 - bugfix such that pie charts by volume are generated correctly
4794 30. The SPA authentication driver was not abandoning authentication and moving
4795 on to the next authenticator when an expansion was forced to fail,
4796 contradicting the general specification for all authenticators. Instead it
4797 was generating a temporary error. It now behaves as specified.
4799 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4800 (the order specifies the preference for clients). The order is now AES256,
4801 AES128, 3DES, ARCFOUR128.
4803 31. Small patch to Sieve code - explicitly set From: when generating an
4806 32. Exim crashed if a remote delivery caused a very long error message to be
4807 recorded - for instance if somebody sent an entire SpamAssassin report back
4808 as a large number of 550 error lines. This bug was coincidentally fixed by
4809 increasing the size of one of Exim's internal buffers (big_buffer) that
4810 happened as part of the Exiscan merge. However, to be on the safe side, I
4811 have made the code more robust (and fixed the comments that describe what
4814 33. Some experimental protocols are using DNS PTR records for new purposes. The
4815 keys for these records are domain names, not reversed IP addresses. The
4816 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4817 leaves it alone. Component reversal etc. now happens only for IP addresses.
4820 34. The host_aton() function is supposed to be passed a string that is known
4821 to be a valid IP address. However, in the case of IPv6 addresses, it was
4822 not checking this. This is a hostage to fortune. Exim now panics and dies
4823 if the condition is not met. A case was found where this could be provoked
4824 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4825 components; fortuitously, this particular loophole had already been fixed
4826 by change 4.50/55 or 4.44/33 above.
4828 If there are any other similar loopholes, the new check in host_aton()
4829 itself should stop them being exploited. The report I received stated that
4830 data on the command line could provoke the exploit when Exim was running as
4831 exim, but did not say which command line option was involved. All I could
4832 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4833 running as the user.
4836 35. There was a buffer overflow vulnerability in the SPA authentication code
4837 (which came originally from the Samba project). I have added a test to the
4838 spa_base64_to_bits() function which I hope fixes it.
4841 36. The daemon start-up calls getloadavg() while still root for those OS that
4842 need the first call to be done as root, but it missed one case: when
4843 deliver_queue_load_max is set with deliver_drop_privilege. This is
4844 necessary for the benefit of the queue runner, because there is no re-exec
4845 when deliver_drop_privilege is set.
4847 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4848 was in use, was not putting the data itself into the right store pool;
4849 consequently, it could be overwritten for a subsequent message in the same
4850 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4853 38. Sometimes the final signoff response after QUIT could fail to get
4854 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4855 < 0 before doing a fflush(). This bug looks as though it goes back to the
4856 introduction of TLS in release 3.20, but "sometimes" must have been rare
4857 because the tests only now provoked it.
4863 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4864 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4865 Now it returns an integer. A number of calls were still expecting a T/F
4866 return. Fortuitously, in all cases, the tests worked in OK situations,
4867 which is the norm. However, things would have gone wrong on any write
4868 failures on the smtp file descriptor. This function is used when sending
4869 messages over SMTP and also when doing verify callouts.
4871 2. When Exim is called to do synchronous delivery of a locally submitted
4872 message (the -odf or -odi options), it no longer closes stderr before doing
4875 3. Implemented the mua_wrapper option.
4877 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4879 5. Implemented the functions header_remove(), header_testname(),
4880 header_add_at_position(), and receive_remove_recipient(), and exported them
4883 6. If an ACL "warn" statement specified the addition of headers, Exim already
4884 inserted X-ACL-Warn: at the start if there was no header name. However, it
4885 was not making this test for the second and subsequent header lines if
4886 there were newlines in the string. This meant that an invalid header could
4887 be inserted if Exim was badly configured.
4889 7. Allow an ACL "warn" statement to add header lines at the start or after all
4890 the Received: headers, as well as at the end.
4892 8. Added the rcpt_4xx retry error code.
4894 9. Added postmaster_mailfrom=xxx to callout verification option.
4896 10. Added mailfrom=xxxx to the callout verification option, for verify=
4899 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4900 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4902 12. Inserted some casts to stop certain compilers warning when using pointer
4903 differences as field lengths or precisions in printf-type calls (mostly
4904 affecting debugging statements).
4906 13. Added optional readline() support for -be (dynamically loaded).
4908 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4909 same clock tick as a message's arrival, so that its received time was the
4910 same as the "first fail" time on the retry record, and that message
4911 remained on the queue past the ultimate address timeout, every queue runner
4912 would try a delivery (because it was past the ultimate address timeout) but
4913 after another failure, the ultimate address timeout, which should have then
4914 bounced the address, did not kick in. This was a "< instead of <=" error;
4915 in most cases the first failure would have been in the next clock tick
4916 after the received time, and all would be well.
4918 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4919 being recognized when the domain list was tested by the match_domain
4920 condition in an expansion string.
4922 16. Added the ${str2b64: operator.
4924 17. Exim was always calling setrlimit() to set a large limit for the number of
4925 processes, without checking whether the existing limit was already
4926 adequate. (It did check for the limit on file descriptors.) Furthermore,
4927 errors from getrlimit() and setrlimit() were being ignored. Now they are
4928 logged to the main and panic logs, but Exim does carry on, to try to do its
4929 job under whatever limits there are.
4931 18. Imported PCRE 5.0.
4933 19. Trivial typo in log message " temporarily refused connection" (the leading
4936 20. If the log selector return_path_on_delivery was set and an address was
4937 redirected to /dev/null, the delivery process crashed because it assumed
4938 that a return path would always be set for a "successful" delivery. In this
4939 case, the whole delivery is bypassed as an optimization, and therefore no
4942 21. Internal re-arrangement: the function for sending a challenge and reading
4943 a response while authentication was assuming a zero-terminated challenge
4944 string. It's now changed to take a pointer and a length, to allow for
4945 binary data in such strings.
4947 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4949 23. Exim was not respecting finduser_retries when seeking the login of the
4950 uid under which it was called; it was always trying 10 times. (The default
4951 setting of finduser_retries is zero.) Also, it was sleeping after the final
4952 failure, which is pointless.
4954 24. Implemented tls_on_connect_ports.
4956 25. Implemented acl_smtp_predata.
4958 26. If the domain in control=submission is set empty, Exim assumes that the
4959 authenticated id is a complete email address when it generates From: or
4960 Sender: header lines.
4962 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4963 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4964 chown and chgrp in /bin and hostname in /usr/bin.
4966 28. Exim was keeping the "process log" file open after each use, just as it
4967 does for the main log. This opens the possibility of it remaining open for
4968 long periods when the USR1 signal hits a daemon. Occasional processlog
4969 errors were reported, that could have been caused by this. Anyway, it seems
4970 much more sensible not to leave this file open at all, so that is what now
4973 29. The long-running daemon process does not normally write to the log once it
4974 has entered its main loop, and it closes the log before doing so. This is
4975 so that log files can straightforwardly be renamed and moved. However,
4976 there are a couple of unusual error situations where the daemon does write
4977 log entries, and I had neglected to close the log afterwards.
4979 30. The text of an SMTP error response that was received during a remote
4980 delivery was being truncated at 512 bytes. This is too short for some of
4981 the long messages that one sometimes sees. I've increased the limit to
4984 31. It is now possible to make retry rules that apply only when a message has a
4985 specific sender, in particular, an empty sender.
4987 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4988 it possible to be selective about when SMTP synchronization is enforced.
4990 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4992 32. Implemented hosts_connection_nolog.
4994 33. Added an ACL for QUIT.
4996 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4999 35. Added mailbox_size and mailbox_filecount to appendfile.
5001 36. Added control = no_multiline_responses to ACLs.
5003 37. There was a bug in the logic of the code that waits for the clock to tick
5004 in the case where the clock went backwards by a substantial amount such
5005 that the microsecond fraction of "now" was more than the microsecond
5006 fraction of "then" (but the whole seconds number was less).
5008 38. Added support for the libradius Radius client library this is found on
5009 FreeBSD (previously only the radiusclient library was supported).
5015 1. When certain lookups returned multiple values in the form name=value, the
5016 quoting of the values was not always being done properly. Specifically:
5017 (a) If the value started with a double quote, but contained no whitespace,
5019 (b) If the value contained whitespace other than a space character (i.e.
5020 tabs or newlines or carriage returns) it was not quoted.
5021 This fix has been applied to the mysql and pgsql lookups by writing a
5022 separate quoting function and calling it from the lookup code. The fix
5023 should probably also be applied to nisplus, ibase and oracle lookups, but
5024 since I cannot test any of those, I have not disturbed their existing code.
5026 2. A hit in the callout cache for a specific address caused a log line with no
5027 reason for rejecting RCPT. Now it says "Previous (cached) callout
5028 verification failure".
5030 3. There was an off-by-one bug in the queryprogram router. An over-long
5031 return line was truncated at 256 instead of 255 characters, thereby
5032 overflowing its buffer with the terminating zero. As well as fixing this, I
5033 have increased the buffer size to 1024 (and made a note to document this).
5035 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5036 when Exim is waiting for an SMTP response from a remote server, Exim
5037 restarts its select() call on the socket, thereby resetting its timeout.
5038 This is not a problem when such interrupts are rare. Somebody set up a cron
5039 job to run exiwhat every 2 minutes, which is less than the normal select()
5040 timeout (5 or 10 minutes). This meant that the select() timeout never
5041 kicked in because it was always reset. I have fixed this by comparing the
5042 time when an interrupt arrives with the time at the start of the first call
5043 to select(). If more time than the timeout has elapsed, the interrupt is
5044 treated as a timeout.
5046 5. Some internal re-factoring in preparation for the addition of Sieve
5047 extensions (by MH). In particular, the "personal" test is moved to a
5048 separate function, and given an option for scanning Cc: and Bcc: (which is
5049 not set for Exim filters).
5051 6. When Exim created an email address using the login of the caller as the
5052 local part (e.g. when creating a From: or Sender: header line), it was not
5053 quoting the local part when it contained special characters such as @.
5055 7. Installed new OpenBSD configuration files.
5057 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5058 try to make them clearer.
5060 9. Callout options, other than the timeout value, were being ignored when
5061 verifying sender addresses in header lines. For example, when using
5063 verify = header_sender/callout=no_cache
5065 the cache was (incorrectly) being used.
5067 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5068 only the Cygwin environment.
5070 11. When return_path_on_delivery was set as a log selector, if different remote
5071 addresses in the same message used different return paths and parallel
5072 remote delivery occurred, the wrong values would sometimes be logged.
5073 (Whenever a remote delivery process finished, the return path value from
5074 the most recently started remote delivery process was logged.)
5076 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5077 lines when AUTH and/or TLS are in use. This is the "received protocol"
5078 field. Exim used to use "asmtp" for authenticated SMTP, without any
5079 indication (in the protocol name) for TLS use. Now it follows the RFC and
5080 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5081 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5082 names appear in log lines as well as in Received: header lines.
5084 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5085 extensions, and comparison tests, and to fix some bugs.
5087 14. Changes to the "personal" filter test:
5089 (1) The test was buggy in that it was just doing the equivalent of
5090 "contains" tests on header lines. For example, if a user's address was
5091 anne@some.where, the "personal" test would incorrectly be true for
5093 To: susanne@some.where
5095 This test is now done by extracting each address from the header in turn,
5096 and checking the entire address. Other tests that are part of "personal"
5097 are now done using regular expressions (for example, to check local parts
5098 of addresses in From: header lines).
5100 (2) The list of non-personal local parts in From: addresses has been
5101 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5102 taken from the Sieve specification recommendations.
5104 (3) If the message contains any header line starting with "List-" it is
5105 treated as non-personal.
5107 (4) The test for "circular" in the Subject: header line has been removed
5108 because it now seems ill-conceived.
5110 15. Minor typos in src/EDITME comments corrected.
5112 16. Installed latest exipick from John Jetmore.
5114 17. If headers_add on a router specified a text string that was too long for
5115 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5116 of string_sprintf() is now avoided.
5118 18. $message_body_size was not set (it was always zero) when running the DATA
5119 ACL and the local_scan() function.
5121 19. For the "mail" command in an Exim filter, no default was being set for
5122 the once_repeat time, causing a random time value to be used if "once" was
5123 specified. (If the value happened to be <= 0, no repeat happened.) The
5124 default is now 0s, meaning "never repeat". The "vacation" command was OK
5125 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5126 (I found it when inspecting the code).
5128 20. There is now an overall timeout for performing a callout verification. It
5129 defaults to 4 times the callout timeout, which applies to individual SMTP
5130 commands during the callout. The overall timeout applies when there is more
5131 than one host that can be tried. The timeout is checked before trying the
5132 next host. This prevents very long delays if there are a large number of
5133 hosts and all are timing out (e.g. when the network connections are timing
5134 out). The value of the overall timeout can be changed by specifying an
5135 additional sub-option for "callout", called "maxwait". For example:
5137 verify = sender/callout=5s,maxwait=20s
5139 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5140 to the end before writing, but this should make it even safer).
5142 22. Exim was forgetting that it had advertised PIPELINING for the second and
5143 subsequent messages on an SMTP connection. It was also not resetting its
5144 memory on STARTTLS and an internal HELO.
5146 23. When Exim logs an SMTP synchronization error within a session, it now
5147 records whether PIPELINING has been advertised or not.
5149 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5150 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5151 rather than long int.
5153 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5155 26. Added the never_mail option to autoreply.
5161 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5162 crash if the getsockname() call failed; this can happen if a connection is
5163 closed very soon after it is established. The problem was simply in the
5164 order in which certain operations were done, causing Exim to try to write
5165 to the SMTP stream before it had set up the file descriptor. The bug has
5166 been fixed by making things happen in the correct order.
5172 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5173 before the connection was closed, thus losing the rejection response.
5175 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5176 some early Solaris releases, but causes trouble in current releases where
5177 socklen_t is defined.
5179 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5182 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5185 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5186 that releasing the top of it at the end releases what was used for sub-
5187 expansions (unless the block got too big). However, discard this block if
5188 the first thing is a variable or header, so that we can use its block when
5189 it is dynamic (useful for very large $message_headers, for example).
5191 6. Lookups now cache *every* query, not just the most recent. A new, separate
5192 store pool is used for this. It can be recovered when all lookup caches are
5193 flushed. Lookups now release memory at the end of their result strings.
5194 This has involved some general refactoring of the lookup sources.
5196 7. Some code has been added to the store_xxx() functions to reduce the amount
5197 of flapping under certain conditions.
5199 8. log_incoming_interface used to affect only the <= reception log lines. Now
5200 it causes the local interface and port to be added to several more SMTP log
5201 lines, for example "SMTP connection from", and rejection lines.
5203 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5205 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5207 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5208 could be overwritten at the end of the current message (or the start of a
5209 new message if it was set in a HELO ACL). The value is now preserved for
5210 the duration of the SMTP connection.
5212 12. If a transport had a headers_rewrite setting, and a matching header line
5213 contained an unqualified address, that address was qualified, even if it
5214 did not match any rewriting rules. The underlying bug was that the values
5215 of the flags that permit the existence of unqualified sender and recipient
5216 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5217 non-local messages, and by -bnq for local messages) were not being
5218 preserved with the message after it was received.
5220 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5221 "next input=" as part of the text comprising the host identity instead of
5222 the correct text. The code was using the same buffer for two different
5223 strings. However, depending on which order the printing function evaluated
5224 its arguments, the bug did not always show up. Under Linux, for example, my
5225 test suite worked just fine.
5227 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5228 doesn't work with some older versions of Perl. It has been changed to "my",
5229 which in any case is probably the better facility to use.
5231 15. A really picky compiler found some instances of statements for creating
5232 error messages that either had too many or two few arguments for the format
5235 16. The size of the buffer for calls to the DNS resolver has been increased
5236 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5237 for addresses that have a lot of PTR records. This alleviates a problem; it
5238 does not fully solve it.
5240 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5241 buffer now truncates the list and logs the incident, which is the same
5242 action as happens when Exim is looking up a host name and its aliases.
5243 Previously in this situation something unpredictable would happen;
5244 sometimes it was "internal error: store_reset failed".
5246 18. If a server dropped the connection unexpectedly when an Exim client was
5247 using GnuTLS and trying to read a response, the client delivery process
5248 crashed while trying to generate an error log message.
5250 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5251 string, for example:
5253 warn message = H1: something\nH2: something
5255 the text was added as a single header line from Exim's point of view
5256 though it ended up OK in the delivered message. However, searching for the
5257 second and subsequent header lines using $h_h2: did not work. This has been
5258 fixed. Similarly, if a system filter added multiple headers in this way,
5259 the routers could not see them.
5261 20. Expanded the error message when iplsearch is called with an invalid key to
5262 suggest using net-iplsearch in a host list.
5264 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5265 ACLs are no longer actually imposed (and a message to that effect is
5268 22. If a "gecos" field in a passwd entry contained escaped characters, in
5269 particular, if it contained a \" sequence, Exim got it wrong when building
5270 a From: or a Sender: header from that name. A second bug also caused
5271 incorrect handling when an unquoted " was present following a character
5272 that needed quoting.
5274 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5275 was not being matched caselessly.
5277 24. Arranged for all hyphens in the exim.8 source to be escaped with
5280 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5281 properly. Recipient callout cache records were still being keyed to include
5282 the sender, even when use_sender was set false. This led to far more
5283 callouts that were necessary. The sender is no longer included in the key
5284 when use_sender is false.
5286 26. Added "control = submission" modifier to ACLs.
5288 27. Added the ${base62d: operator to decode base 62 numbers.
5290 28. dnsdb lookups can now access SRV records.
5292 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5293 the configuration file.
5295 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5296 (-v) mode. This makes the output for a verbose queue run more intelligible.
5298 31. Added a use_postmaster feature to recipient callouts.
5300 32. Added the $body_zerocount variable, containing the number of binary zero
5301 bytes in the message body.
5303 33. The time of last modification of the "new" subdirectory is now used as the
5304 "mailbox time last read" when there is a quota error for a maildir
5307 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5309 35. Added +ignore_unknown as a special item in host lists.
5311 36. Code for decoding IPv6 addresses in host lists is now included, even if
5312 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5313 address was recognized as an IP address, but was then not correctly decoded
5314 into binary, causing unexpected and incorrect effects when compared with
5321 1. Very minor rewording of debugging text in manualroute to say "list of
5322 hosts" instead of "hostlist".
5324 2. If verify=header_syntax was set, and a header line with an unqualified
5325 address (no domain) and a large number of spaces between the end of the
5326 name and the colon was received, the reception process suffered a buffer
5327 overflow, and (when I tested it) crashed. This was caused by some obsolete
5328 code that should have been removed. The fix is to remove it!
5330 3. When running in the test harness, delay a bit after writing a bounce
5331 message to get a bit more predictability in the log output.
5333 4. Added a call to search_tidyup() just before forking a reception process. In
5334 theory, someone could use a lookup in the expansion of smtp_accept_max_
5335 per_host which, without the tidyup, could leave open a database connection.
5337 5. Added the variables $recipient_data and $sender_data which get set from a
5338 lookup success in an ACL "recipients" or "senders" condition, or a router
5339 "senders" option, similar to $domain_data and $local_part_data.
5341 6. Moved the writing of debug_print from before to after the "senders" test
5344 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5345 problems for message scanning, either using a data ACL, or using
5346 local_scan() because the Received: header was not generated till after they
5347 were called (in order to set the time as the time of reception completion).
5348 I have revised the way this works. The header is now generated after the
5349 body is received, but before the ACL or local_scan() are called. After they
5350 are run, the timestamp in the header is updated.
5356 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5357 before starting a queue runner without re-exec. This happened only when
5358 deliver_drop_privilege was set or when the Exim user was set to root. The
5359 effect of the bug was that timeouts during subsequent deliveries caused
5360 crashes instead of being properly handled. The handler is now left at its
5361 default (and expected) setting.
5363 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5364 message, again when deliver_drop_privilege is set or Exim is run as root.
5365 The bug described in (1) was not present in this case, but the tidying up
5366 of the other signals was missing. I have made the two cases consistent.
5368 3. The ignore_target_hosts setting on a manualroute router was being ignored
5369 for hosts that were looked up using the /MX notation.
5371 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5374 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5375 operated on the sender address. After changing the $sender_address to <>
5376 for the sender address verify, Exim was re-instated it as the original
5377 (before rewriting) address, but remembering that it had rewritten it, so it
5378 wasn't rewriting it again. This bug also had the effect of breaking the
5379 sender address verification caching when the sender address was rewritten.
5381 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5382 This has been changed so that if the ip literal address matches
5383 ignore_target_hosts, the router declines.
5385 7. Added expansion conditions match_domain, match_address, and match_local_
5386 part (NOT match_host).
5388 8. The placeholder for the Received: header didn't have a length field set.
5390 9. Added code to Exim itself and to exim_lock to test for a specific race
5391 condition that could lead to file corruption when using MBX delivery. The
5392 issue is with the lockfile that is created in /tmp. If this file is removed
5393 after a process has opened it but before that process has acquired a lock,
5394 there is the potential for a second process to recreate the file and also
5395 acquire a lock. This could lead to two Exim processes writing to the file
5396 at the same time. The added code performs the same test as UW imapd; it
5397 checks after acquiring the lock that its file descriptor still refers to
5398 the same named file.
5400 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5401 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5404 11. Added the smtp_active_hostname option. If used, this will typically be made
5405 to depend on the incoming interface address. Because $interface_address is
5406 not set up until the daemon has forked a reception process, error responses
5407 that can happen earlier (such as "too many connections") no longer contain
5410 12. If an expansion in a condition on a "warn" statement fails because a lookup
5411 defers, the "warn" statement is abandoned, and the next ACL statement is
5412 processed. Previously this caused the whole ACL to be aborted.
5414 13. Added the iplsearch lookup type.
5416 14. Added ident_timeout as a log selector.
5418 15. Added tls_certificate_verified as a log selector.
5420 16. Added a global option tls_require_ciphers (compare the smtp transport
5421 option of the same name). This controls incoming TLS connections.
5423 17. I finally figured out how to make tls_require_ciphers do a similar thing
5424 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5425 before starting the TLS session.
5427 18. Tabs are now shown as \t in -bP output.
5429 19. If the log selector return_path_on_delivery was set, Exim crashed when
5430 bouncing a message because it had too many Received: header lines.
5432 20. If two routers both had headers_remove settings, and the first one included
5433 a superfluous trailing colon, the final name in the first list and the
5434 first name in the second list were incorrectly joined into one item (with a
5435 colon in the middle).
5441 1. Added -C and -D options to the exinext utility, mainly to make it easier
5442 to include in the automated testing, but these could be helpful when
5443 multiple configurations are in use.
5445 2. The exinext utility was not formatting the output nicely when there was
5446 an alternate port involved in the retry record key, nor when there was a
5447 message id as well (for retries that were specific to a specific message
5448 and a specific host). It was also confused by IPv6 addresses, because of
5449 the additional colons they contain. I have fixed the IPv4 problem, and
5450 patched it up to do a reasonable job for IPv6.
5452 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5453 delivery, the log line now contains "pipelined" if PIPELINING was used.
5455 4. An SMTP transport process used to panic and die if the bind() call to set
5456 an explicit outgoing interface failed. This has been changed; it is now
5457 treated in the same way as a connect() failure.
5459 5. A reference to $sender_host_name in the part of a conditional expansion
5460 that was being skipped was still causing a DNS lookup. This no longer
5463 6. The def: expansion condition was not recognizing references to header lines
5464 that used bh_ and bheader_.
5466 7. Added the _cache feature to named lists.
5468 8. The code for checking quota_filecount in the appendfile transport was
5469 allowing one more file than it should have been.
5471 9. For compatibility with Sendmail, the command line option
5479 and sets the incoming protocol and host name (for trusted callers). The
5480 host name and its colon can be omitted when only the protocol is to be set.
5481 Note the Exim already has two private options, -pd and -ps, that refer to
5482 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5483 "s", but I don't think that's a major issue.
5485 10. A number of refactoring changes to the code, none of which should affect
5488 (a) The number of logging options was getting close to filling up the
5489 32-bit word that was used as a bit map. I have split them into two classes:
5490 those that are passed in the argument to log_write(), and those that are
5491 only ever tested independently outside of that function. These are now in
5492 separate 32-bit words, so there is plenty of room for expansion again.
5493 There is no change in the user interface or the logging behaviour.
5495 (b) When building, for example, log lines, the code previously used a
5496 macro that called string_cat() twice, in order to add two strings. This is
5497 not really sufficiently general. Furthermore, there was one instance where
5498 it was actually wrong because one of the argument was used twice, and in
5499 one call a function was used. (As it happened, calling the function twice
5500 did not affect the overall behaviour.) The macro has been replaced by a
5501 function that can join an arbitrary number of extra strings onto a growing
5504 (c) The code for expansion conditions now uses a table and a binary chop
5505 instead of a serial search (which was left over from when there were very
5506 few conditions). Also, it now recognizes conditions like "pam" even when
5507 the relevant support is not compiled in: a suitably worded error message is
5508 given if an attempt is made to use such a condition.
5510 11. Added ${time_interval:xxxxx}.
5512 12. A bug was causing one of the ddress fields not to be passed back correctly
5513 from remote delivery subprocesses. The field in question was not being
5514 subsequently used, so this caused to problems in practice.
5516 13. Added new log selectors queue_time and deliver_time.
5518 14. Might have fixed a bug in maildirsizefile handling that threw up
5519 "unexpected character" debug warnings, and recalculated the data
5520 unnecessarily. In any case, I expanded the warning message to give more
5523 15. Added the message "Restricted characters in address" to the statements in
5524 the default ACL that block characters like @ and % in local parts.
5526 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5527 Three changes have been made:
5529 (a) There was a serious bug; a negative response to MAIL caused the whole
5530 recipient domain to be cached as invalid, thereby blocking all messages
5531 to all local parts at the same domain, from all senders. This bug has
5532 been fixed. The domain is no longer cached after a negative response to
5533 MAIL if the sender used is not empty.
5535 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5538 (c) A new callout option, "use_sender" has been added for people who want
5539 the modified behaviour.
5545 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5548 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5549 indeed breaks things for older releases.
5551 3. Added additional logging to the case where there is a problem reading data
5552 from a filter that is running in a subprocess using a pipe, in order to
5553 try to track down a specific problem.
5555 4. Testing facility fudge: when running in the test harness and attempting
5556 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5557 getting "No route to host". Convert this to a timeout.
5559 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5562 6. Some OS don't have socklen_t but use size_t instead. This affects the
5563 fifth argument of getsockopt() amongst other things. This is now
5564 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5565 can be set for individual OS. I have set it for SunOS5, OSF1, and
5566 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5567 some earlier ones do not.
5569 7. Change 4.30/15 was not doing the test caselessly.
5571 8. The standard form for an IPv6 address literal was being rejected by address
5572 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5573 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5574 this, as well as the form without the "IPv6" on the front (but only when
5575 address literals are enabled, of course).
5577 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5579 10. Exim crashed if a message with an empty sender address specified by -f
5580 encountered a router with an errors_to setting. This could be provoked only
5581 by a command such as
5585 where an empty string was supplied; "<>" did not hit this bug.
5587 11. Installed PCRE release 4.5.
5589 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5590 remained set. It is now erased.
5592 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5593 times from message ids (which are base 36 rather than the normal 62).
5595 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5596 were being counted as actual protocol errors, and logged if the log
5597 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5598 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5599 and DATA following a set of rejected RCPTs do not count as protocol errors.
5600 In other words, Exim assumes they were pipelined, though this may not
5601 actually be the case. Of course, in all cases the client gets an
5602 appropriate error code.
5604 15. If a lookup fails in an ACL condition, a message about the failure may
5605 be available; it is used if testing the ACL cannot continue, because most
5606 such messages specify what the cause of the deferral is. However, some
5607 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5608 that caused an old message to be retained and used if a later statement
5609 caused a defer, replacing the real cause of the deferral.
5611 16. If an IP address had so many PTR records that the DNS lookup buffer
5612 was not large enough to hold them, Exim could crash while trying to process
5613 the truncated data. It now detects and logs this case.
5615 17. Further to 4.21/58, another change has been made: if (and only if) the
5616 first line of a message (the first header line) ends with CRLF, a bare LF
5617 in a subsequent header line has a space inserted after it, so as not to
5618 terminate the header.
5620 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5621 unnecessarily, used atoi() instead of strtol(), and didn't check the
5622 termination when getting file sizes from file names by regex.
5624 19. Completely re-implemented the support for maildirsize files, in the light
5625 of a number of problems with the previous contributed implementation
5626 (4.30/29). In particular:
5628 . If the quota is zero, the maildirsize file is maintained, but no quota is
5631 . If the maildir directory does not exist, it is created before any attempt
5632 to write a maildirsize file.
5634 . The quota value in the file is just a cache; if the quota is changed in
5635 the transport, the new value overrides.
5637 . A regular expression is available for excluding directories from the
5640 20. The autoreply transport checks the characters in options that define the
5641 message's headers; it allows continued headers, but it was checking with
5642 isspace() after an embedded newline instead of explicitly looking for a
5645 21. If all the "regular" hosts to which an address was routed had passed their
5646 expiry times, and had not reached their retry times, the address was
5647 bounced, even if fallback hosts were defined. Now Exim should go on to try
5650 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5651 equivalent code in the SMTP transport. Some hosts send humungous responses
5652 to HELO/EHLO, more than 1024 it seems.
5654 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5655 gives compiler warnings in some environments. I've now done it "properly",
5658 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5659 (because of problems with the built-in one) was declared to return uschar *
5660 instead of char *, causing compiler failure.
5662 25. Fixed a file descriptor leak when processing alias/forward files.
5664 26. Fixed a minor format string issue in dbfn.c.
5666 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5668 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5669 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5672 29. When a sender address is verified, it is cached, to save repeating the test
5673 when there is more than one recipient in a message. However, when the
5674 verification involves a callout, it is possible for different callout
5675 options to be set for different recipients. It is too complicated to keep
5676 track of this in the cache, so now Exim always runs a verification when a
5677 callout is required, relying on the callout cache for the optimization.
5678 The overhead is duplication of the address routing, but this should not be
5681 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5682 to be verified with callout=postmaster, and the main callout worked but the
5683 postmaster check failed, the verification correctly failed. However, if a
5684 subsequent RCPT command asked for sender verification *without* the
5685 postmaster check, incorrect caching caused this verification also to fail,
5688 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5689 it was not caching the DNS options (qualify_single, search_parents) that
5690 were used when the lookup failed. A subsequent lookup with different
5691 options therefore always gave the same answer, though there were cases
5692 where it should not have. (Example: a "domains = !$mx_any" option on a
5693 dnslookup router: the "domains" option is always processed without any
5694 widening, but the router might have qualify_single set.) Now Exim uses the
5695 cached value only when the same options are set.
5697 32. Added John Jetmore's "exipick" utility to the distribution.
5699 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5700 than a timeout (e.g. a certificate is required, and is not provided), an
5701 Exim server now closes the connection immediately. Previously it waited for
5702 the client to close - but if the client is SSL, it seems that they each
5703 wait for each other, leading to a delay before one of them times out.
5705 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5706 maintained 0.8.x compatibility because I don't think many are using it, and
5707 it is clearly obsolete.
5709 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5712 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5713 client certificate was expired. A simple patch fixes this, though I don't
5714 understand the full logic of why the verify callback is called multiple
5717 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5718 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5719 which causes problems with some clients (such as the Certicom SSL Plus
5720 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5721 disables the coutermeasure allowing Eudora to connect."
5723 38. Exim was not checking that a write() to a log file succeeded. This could
5724 lead to Bad Things if a log got too big, in particular if it hit a file
5725 size limit. Exim now panics and dies if it cannot write to a log file, just
5726 as it does if it cannot open a log file.
5728 39. Modified OS/Makefile-Linux so that it now contains
5730 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5732 The two -D definitions ensure that Exim is compiled with large file
5733 support, which makes it possible to handle log files that are bigger than
5736 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5737 instance) a domain was checked against a named list that involved a lookup,
5738 causing $domain_data to be set, then another domain was checked against the
5739 same list, then the first domain was re-checked, the value of $domain_data
5740 after the final check could be wrong. In particular, if the second check
5741 failed, it could be set empty. This bug probably also applied to
5744 41. The strip_trailing_dot option was not being applied to the address given
5745 with the -f command-line option.
5747 42. The code for reading a message's header from the spool was incrementing
5748 $received_count, but never initializing it. This meant that the value was
5749 incorrect (doubled) while delivering a message in the same process in which
5750 it was received. In the most common configuration of Exim, this never
5751 happens - a fresh exec is done - but it can happen when
5752 deliver_drop_privilege is set.
5754 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5755 it now includes up to 150 characters of the unexpected data in the log
5758 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5759 and building data strings. The size of both of these buffers was 10 000
5760 bytes - far larger than anybody would *ever* want, thought I. Needless to
5761 say, somebody hit the limit. I have increased the maximum line length to
5762 20 000 and the maximum data length of concatenated lines to 100 000. I have
5763 also fixed two bugs, because there was no checking on these buffers. Tsk,
5764 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5765 buffer is too small.
5767 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5768 lsearch lookups. Now it does.
5770 46. When parsing a route_list item in a manualroute router, a fixed-length
5771 buffer was used for the list of hosts. I made this 1024 bytes long,
5772 thinking that nobody would ever have a list of hosts that long. Wrong.
5773 Somebody had a whole pile of complicated expansion conditions, and the
5774 string was silently truncated, leading to an expansion error. It turns out
5775 that it is easier to change to an unlimited length (owing to other changes
5776 that have happened since this code was originally written) than to build
5777 structure for giving a limitation error. The length of the item that
5778 expands into the list of hosts is now unlimited.
5780 47. The lsearch lookup could not handle data where the length of text line was
5781 more than 4095 characters. Such lines were truncated, leading to shortened
5782 data being returned. It should now handle lines of any length.
5784 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5785 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5788 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5789 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5790 no longer happens in comment lines. A list of replacements is now placed
5791 at the head of all of the source files, except those whose only change is
5792 to replace PERL_COMMAND in the very first #! line.
5794 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5795 messages on the queue, with a bottom-up merge sort, using code contributed
5796 by Michael Haardt. This should make operations like -bp somewhat faster on
5797 large queues. It won't affect queue runners, except when queue_run_in_order
5800 51. Installed eximstats 1.31 in the distribution.
5802 52. Added support for SRV lookups to the dnslookup router.
5804 53. If an ACL referred to $message_body or $message_body_end, the value was not
5805 reset for any messages that followed in the same SMTP session.
5807 54. The store-handling optimization for building very long strings was not
5808 differentiating between the different store pools. I don't think this
5809 actually made any difference in practice, but I've tidied it.
5811 55. While running the routers to verify a sender address, $sender_address
5812 was still set to the sender address. This is wrong, because when routing to
5813 send a bounce to the sender, it would be empty. Therefore, I have changed
5814 it so that, while verifying a sender address, $sender_address is set to <>.
5815 (There is no change to what happens when verifying a recipient address.)
5817 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5818 target A or AAAA records (if not already returned) without resetting the
5819 qualify_single or search_parents options of the DNS resolver. These are
5820 inappropriate in this case because the targets of MX and SRV records must
5821 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5822 target that, when qualified, matched something in the local domain. These
5823 two options are now turned off when doing these lookups.
5825 57. It seems that at least some releases of Reiserfs (which does not have the
5826 concept of a fixed number of inodes) returns zero and not -1 for the
5827 number of available inodes. This interacted badly with check_spool_inodes,
5828 which assumed that -1 was the "no such thing" setting. What I have done is
5829 to check that the total number of inodes is greater than zero before doing
5830 the test of how many are available.
5832 58. When a "warn" ACL statement has a log_message modifier, the message is
5833 remembered, and not repeated. This is to avoid a lot of repetition when a
5834 message has many recipients that cause the same warning to be written.
5835 However, Exim was preserving the list of already written lines for an
5836 entire SMTP session, which doesn't seem right. The memory is now reset if a
5837 new message is started.
5839 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5840 debugging output unless log_rewrite was also set.
5842 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5843 of (int)(handle) when we know that handle contains (void *)(-1).
5845 61. The Exim daemon panic-logs an error return when it closes the incoming
5846 connection. However "connection reset by peer" seems to be common, and
5847 isn't really an error worthy of noting specially, so that particular error
5850 62. When Exim is trying to find all the local interfaces, it used to panic and
5851 die if the ioctl to get the interface flags failed. However, it seems that
5852 on at least one OS (Solaris 9) it is possible to have an interface that is
5853 included in the list of interfaces, but for which you get a failure error
5854 for this call. This happens when the interface is not "plumbed" into a
5855 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5856 failure of the "get flags" call assumes that the interface is down.
5858 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5859 makes life easier for people who are doing arithmetic on fields extracted
5860 from dates, where you often get leading zeros that should not be
5861 interpreted as octal.
5863 64. Added qualify_domain to the redirect router, to override the global
5866 65. If a pathologically long header line contained very many addresses (the
5867 report of this problem mentioned 10 000) and each of them was rewritten,
5868 Exim could use up a very large amount of memory. (It kept on making new
5869 copies of the header line as it rewrote, and never released the old ones.)
5870 At the expense of a bit more processing, the header rewriting function has
5871 been changed so that it no longer eats memory in this way.
5873 66. The generation of the Received: header has been moved from the time that a
5874 message starts to be received, to the time that it finishes. The timestamp
5875 in the Received: header should now be very close to that of the <= log
5876 line. There are two side-effects of this change:
5878 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5879 logged header lines no longer include the local Received: line, because
5880 it has not yet been created. The same applies to a copy of the message
5881 that is returned to a non-SMTP sender when a message is rejected.
5883 (b) When a filter file is tested using -bf, no additional Received: header
5884 is added to the test message. After some thought, I decided that this
5887 This change does not affect the value of $received_for. It is still set
5888 after address rewriting, but before local_scan() is called.
5890 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5892 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5893 gave an unhelpful panic error message, and a defer error. I have managed to
5894 change this behaviour so that it now rejects any supplied certificate,
5895 which seems right, as the list of acceptable certificates is empty.
5897 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5898 gave an unhelpful defer error. I have not managed to make this reject any
5899 supplied certificates, but the error message it gives is "no certificate
5900 supplied", which is not helpful.
5902 70. exigrep's output now also includes lines that are not associated with any
5903 message, but which match the given pattern. Implemented by a patch from
5904 Martin Sluka, which also tidied up the Perl a bit.
5906 71. Recipient callout verification, like sender verification, was using <> in
5907 the MAIL FROM command. This isn't really the right thing, since the actual
5908 sender may affect whether the remote host accepts the recipient or not. I
5909 have changed it to use the actual sender in the callout; this means that
5910 the cache record is now keyed on a recipient/sender pair, not just the
5911 recipient address. There doesn't seem to be a real danger of callout loops,
5912 since a callout by the remote host to check the sender would use <>.
5913 [SEE ABOVE: changed after hitting problems.]
5915 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5916 temporary errors. However, in the case of such a code being given after
5917 the end of a data transmission (i.e. after ".") Exim was failing to write
5918 a retry record for the message. (Yes, there was some broken host that was
5919 actually sending 8xx at this point.)
5921 73. An unknown lookup type in a host list could cause Exim to panic-die when
5922 the list was checked. (An example that provoked this was putting <; in the
5923 middle of a list instead of at the start.) If this happened during a DATA
5924 ACL check, a -D file could be left lying around. This kind of configuration
5925 error no longer causes Exim to die; instead it causes a defer error. The
5926 incident is still logged to the main and panic logs.
5928 74. Buglet left over from Exim 3 conversion. The message "too many messages
5929 in one connection" was written to the rejectlog but not the mainlog, except
5930 when address rewriting (yes!) was being logged.
5932 75. Added write_rejectlog option.
5934 76. When a system filter was run not as root (that is, when system_filter_user
5935 was set), the values of the $n variables were not being returned to the
5936 main process; thus, they were not subsequently available in the $sn
5939 77. Added +return_path_on_delivery log selector.
5941 78. A connection timeout was being treated differently from recipients deferred
5942 when testing hosts_max_try with a message that was older than the host's
5943 retry timeout. (The host should not be counted, thus allowing all hosts to
5944 be tried at least once before bouncing.) This may have been the cause of an
5945 occasionally reported bug whereby a message would remain on the queue
5946 longer than the retry timeout, but would be bounced if a delivery was
5947 forced. I say "may" because I never totally pinned down the problem;
5948 setting up timeout/retry tests is difficult. See also the next item.
5950 79. The ultimate address timeout was not being applied to errors that involved
5951 a combination of host plus message (for example, a timeout on a MAIL
5952 command). When an address resolved to a number of possible hosts, and they
5953 were not all tried for each delivery (e.g. because of hosts_max_try), a
5954 message could remain on the queue longer than the retry timeout.
5956 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5959 81. Fixed an obscure SMTP outgoing bug which required at least the following
5960 conditions: (a) there was another message waiting for the same server;
5961 (b) the server returned 5xx to all RCPT commands in the first message so
5962 that the message was not completed; (c) the server dropped the connection
5963 or gave a negative response to the RSET that Exim sends to abort the
5964 transaction. The observed case was a dropped connection after DATA that had
5965 been sent in pipelining mode. That is, the server had advertised PIPELINING
5966 but was not implementing it correctly. The effect of the bug was incorrect
5967 behaviour, such as trying another host, and this could lead to a crash.
5973 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5974 and daemon.c were passed as pointers to ints; they should have been
5975 pointers to socklen_t variables (which are typically unsigned ints).
5977 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5980 3. Fixed a really odd bug that affected only the testing scheme; patching a
5981 certain fixed string in the binary changed the value of another string that
5982 happened to be identical to the end of the original first string.
5984 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5985 name", it returns that address as the IP address. On some operating
5986 systems (e.g. Solaris), it also passes back the IP address string as the
5987 "host name". However, on others (e.g. Linux), it passes back an empty
5988 string. Exim wasn't checking for this, and was changing the host name to an
5989 empty string, assuming it had been canonicalized.
5991 5. Although rare, it is permitted to have more than one PTR record for a given
5992 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5993 all the names associated with an address, because they do in Solaris.
5994 However, it seems that they do not in Linux for data that comes from the
5995 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5996 I found this out when I moved to a new Linux workstation and tried to run
5997 the Exim test suite.
5999 To get round this problem I have changed the code so that it now does its
6000 own call to the DNS to look up PTR records when searching for a host name.
6001 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6002 addresses that are only in /etc/hosts are still found.
6004 This behaviour is, however, controlled by an option called host_lookup_
6005 order, which defaults to "bydns:byaddr". If people want to use the other
6006 order, or indeed, just use one or the other means of lookup, they can
6007 specify it in this variable.
6009 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6010 some operating systems, this comes back from gethostbyaddr() as an empty
6011 string, and this is what Exim used to test for. However, it seems that in
6012 other systems, "." is yielded. Exim now tests for this case too.
6014 7. The values of check_spool_space and check_log_space are now held internally
6015 as a number of kilobytes instead of an absolute number of bytes. If a
6016 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6017 kilobyte. This means that much larger values can be stored.
6019 8. Exim monitor: an attempt to get the action menu when not actually pointing
6020 at a message produces an empty menu entitled "No message selected". This
6021 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6022 no entries in it ("Shell widget menu has zero width and/or height"). So I
6023 have added a single, blank menu entry in this case.
6025 9. Added ${quote_local_part.
6027 10. MIME decoding is now applied to the contents of Subject: header lines when
6030 11. Now that a reference to $sender_host_address automatically causes a reverse
6031 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6032 host lookup before query-style lookups in lists that might use this
6033 variable. This has therefore been abolished, and the "net-" prefix is no
6034 longer necessary for query-style lookups.
6036 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6037 have been a typo for LFLAGS, so it has been changed.
6039 13. The install script calls Exim with "-C /dev/null" in order to find the
6040 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6041 to be output. However, since Exim outputs its version number before the
6042 error, it didn't break the script. It just looked ugly. I fixed this by
6043 always allowing "-C /dev/null" if the caller is root.
6045 14. Ignore overlarge ACL variable number when reading spool file - insurance
6046 against a later release with more variables having written the file.
6048 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6049 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6050 this, as well as the form without the "IPv6" on the front.
6052 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6053 OS/Makefile-Darwin file.
6055 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6056 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6058 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6059 variable. However, some OS use ints for inodes. I've added cast to long int
6060 to get rid of the compiler warning.
6062 19. I had forgotten to lock out "/../" in configuration file names when
6063 ALT_CONFIG_PREFIX was set.
6065 20. Routers used for verification do not need to specify transports. However,
6066 if such a router generated a host list, and callout was configured, Exim
6067 crashed, because it could not find a port number from the (non-existent)
6068 transport. It now assumes port 25 in this circumstance.
6070 21. Added the -t option to exigrep.
6072 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6073 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6074 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6075 them is set, LOOKUP_LSEARCH is forced.
6077 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6079 24. Added sender and host information to the "rejected by local_scan()" log
6080 line; previously there was no indication of these.
6082 25. Added .include_if_exists.
6084 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6085 Linux. It turns out that not all file systems support this. Apparently some
6086 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6087 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6088 not supported on the file descriptor, is now ignored when Exim is trying to
6089 sync a directory. This applies only to Linux.
6091 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6093 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6094 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6097 29. Added support for maildirsize files from supplied patch (modified a bit).
6099 30. The use of :fail: followed by an empty string could lead Exim to respond to
6100 sender verification failures with (e.g.):
6102 550 Verification failed for <xxx>
6103 550 Sender verify failed
6105 where the first response line was missing the '-' that indicates it is not
6106 the final line of the response.
6108 31. The loop for finding the name of the user that called Exim had a hardwired
6109 limit of 10; it now uses the value of finduser_retries, which is used for
6110 all other user lookups.
6112 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6115 33. Exim was neglecting to zero errno before one call of strtol() when
6116 expanding a string and expecting an integer value. On some systems this
6117 resulted in spurious "integer overflow" errors. Also, it was casting the
6118 result into an int without checking.
6120 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6121 did not work. The code looks as if it has *never* worked, though it appears
6122 to have been documented since at least release 1.62. I have made it work.
6124 35. The "timeout_DNS" error in retry rules, also documented since at least
6125 1.62, also never worked. As it isn't clear exactly what this means, and
6126 clearly it isn't a major issue, I have abolished the feature by treating it
6127 as "timeout", and writing a warning to the main and panic logs.
6129 36. The display of retry rules for -brt wasn't always showing the error code
6132 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6133 timeout_connect_A, timeout_connect_MX.
6135 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6136 to the empty sender.
6138 39. The daemon was not analysing the content of -oX till after it had closed
6139 stderr and disconnected from the controlling terminal. This meant that any
6140 syntax errors were only noted on the panic log, and the return code from
6141 the command was 0. By re-arranging the code a little, I've made the
6142 decoding happen first, so such errors now appear on stderr, and the return
6143 code is 1. However, the actual setting up of the sockets still happens in
6144 the disconnected process, so errors there are still only recorded on the
6147 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6148 connections (as happens on some IP stacks) was logged at start up time as
6149 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6150 it from "IPv6 and IPv4", which means that two separate sockets are being
6153 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6154 says whether AF_INET or AF_INET6 was passed as an argument.
6156 42. Exiwhat output was messed up when time zones were included in log
6159 43. Exiwhat now gives more information about the daemon's listening ports,
6160 and whether -tls-on-connect was used.
6162 44. The "port" option of the smtp transport is now expanded.
6164 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6165 silently ignored. Now an error message is written to the main and panic
6168 46. There's a new ACL modifier called "logwrite" which writes to a log file
6169 as soon as it is encountered.
6171 47. Added $local_user_uid and $local_user_gid at routing time.
6173 48. Exim crashed when trying to verify a sender address that was being
6176 49. Exim was recognizing only a space character after ".include". It now also
6177 recognizes a tab character.
6179 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6180 extracting the relevant information from the specification. The man page no
6181 longer contains scrambled data for the -d option, and I've added a section
6182 at the front about calling Exim under different names.
6184 51. Added "extra_headers" argument to the "mail" command in filter files.
6186 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6189 53. Installed eximstats 1.29.
6191 54. Added transport_filter_timeout as a generic transport option.
6193 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6194 Cc: header lines. This was required by RFC 822, but it not required by RFC
6197 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6198 incoming messages that did not have them. Now it does so only if the
6199 message originates locally, that is, if there is no associated remote host
6200 address. When Resent- header lines are present, this applies to the Resent-
6201 lines rather than the non-Resent- lines.
6203 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6204 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6206 58. Messages for configuration errors now include the name of the main
6207 configuration file - useful now that there may be more than one file in a
6208 list (.included file names were always shown).
6210 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6211 for those rare installations that do not start the daemon as root or run it
6212 setuid root. I've cut out the call to initgroups() if the daemon is not
6215 60. The Exim user and group can now be bound into the binary as text strings
6216 that are looked up at the start of Exim's processing.
6218 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6220 62. Added $mailstore_basename variable.
6222 63. Installed patch to sieve.c from Michael Haardt.
6224 64. When Exim failed to open the panic log after failing to open the main log,
6225 the original message it was trying to log was written to stderr and debug
6226 output, but if they were not available (the usual case in production), it
6227 was lost. Now it is written to syslog before the two lines that record the
6228 failures to open the logs.
6230 65. Users' Exim filters run in subprocesses under the user's uid. It is
6231 possible for a "deliver" command or an alias in a "personal" command to
6232 provoke an address rewrite. If logging of address rewriting is configured,
6233 this fails because the process is not running as root or exim. There may be
6234 a better way of dealing with this, but for the moment (because 4.30 needs
6235 to be released), I have disabled address rewrite logging when running a
6236 filter in a non-root, non-exim process.
6242 1. The buildconfig auxiliary program wasn't quoting the value set for
6243 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6244 not defined. This bug was masked in 4.22 by the effect that was fixed in
6247 2. Some messages that were rejected after a message id was allocated were
6248 shown as "incomplete" by exigrep. It no longer does this for messages that
6249 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6251 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6252 have allow_domain_literals set, the ID did not get logged in the <= line.
6253 Domain literals are now always recognized in Message-ID: header lines.
6255 4. The first argument for a ${extract expansion item is the key name or field
6256 number. Leading and trailing spaces in this item were not being ignored,
6257 causing some misleading effects.
6259 5. When deliver_drop_privilege was set, single queue runner processes started
6260 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6261 same command in the process it spins off) were not dropping privilege.
6263 6. When the daemon running as "exim" started a queue runner, it always
6264 re-executed Exim in the spun-off process. This is a waste of effort when
6265 deliver_drop_privilege is set. The new process now just calls the
6266 queue-runner function directly.
6272 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6275 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6276 ignored. Though the use of -oP was forcing the writing of a pid file, it
6277 was always written to the default place.
6279 3. If the message "no IP address found for host xxxx" is generated during
6280 incoming verification, it is now followed by identification of the incoming
6281 connection (so you can more easily find what provoked it).
6283 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6285 5. Added some features to "harden" Exim a bit more against certain attacks:
6287 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6288 be put in Local/Makefile. This is like the never_users runtime option,
6289 but it cannot be overridden. The default setting is "root".
6291 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6292 prefix string with which any file named in a -C command line option
6295 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6296 is retained for -C and -D only if the caller of Exim is root. Without
6297 it, the exim user may also use -C and -D and retain privilege.
6299 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6300 command line option is disabled.
6302 6. Macro names set by the -D option must start with an upper case letter, just
6303 like macro names defined in the configuration file.
6305 7. Added "dereference=" facility to LDAP.
6307 8. Two instances of the typo "uknown" in the source files are fixed.
6309 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6310 the Configure-Makefile script screwed up while processing it.
6312 10. Incorporated PCRE 4.4.
6314 11. The SMTP synchronization check was not operating right at the start of an
6315 SMTP session. For example, it could not catch a HELO sent before the client
6316 waited for the greeting. There is now a check for outstanding input at the
6317 point when the greeting is written. Because of the duplex, asynchronous
6318 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6319 way, but not yet received, when the check is performed.
6321 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6322 on TCP/IP sockets, because this apparently causes some broken clients to
6325 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6326 unchanged) from the Cygwin maintainer.
6328 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6329 was supported instead of testing for mbx. Effectively a typo.
6331 15. The spa authenticator server code was not checking that the input it
6332 received was valid base64.
6334 16. The debug output line for the "set" modifier in ACLs was not showing the
6335 name of the variable that was being set.
6337 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6339 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6340 lookup on its own. Something else was needed to trigger the lookup. For
6341 example, a match in host_lookup or the need for a host name in a host list.
6342 Now, if $sender_host_name is referenced and the host name has not yet been
6343 looked up, a lookup is performed. If the lookup fails, the variable remains
6344 empty, and $host_lookup_failed is set to "1".
6346 19. Added "eqi" as a case-independent comparison operator.
6348 20. The saslauthd authentication condition could segfault if neither service
6349 nor realm was specified.
6351 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6352 error message that was logged was misleading, and incoming SMTP
6353 connections were dropped. The message is now more accurate, and temporary
6354 errors are given to SMTP connections.
6356 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6357 (except RSET) with a 421 error, until QUIT is received. However, it was
6358 failing to send a response to QUIT.
6360 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6361 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6362 ACL is now run after the helo_try_verify_hosts code.
6364 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6365 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6366 case-independent, so other case variants are also recognized). Apparently
6367 some systems use these upper case variants.
6369 25. If more than two messages were waiting for the same host, and a transport
6370 filter was specified for the transport, Exim sent two messages over the
6371 same TCP/IP connection, and then failed with "socket operation on non-
6372 socket" when it tried to send the third.
6374 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6376 27. The extern definition of crypt16() in expand.c was not being excluded when
6377 the OS had its own crypt16() function.
6379 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6380 as a preferred synonym for return_size_limit, both as an option and as an
6383 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6385 30. Changed the default configuration ACL to relax the local part checking rule
6386 for addresses that are not in any local domains. For these addresses,
6387 slashes and pipe symbols are allowed within local parts, but the sequence
6388 /../ is explicitly forbidden.
6390 31. SPA server authentication was not clearing the challenge buffer before
6393 32. log_message in a "warn" ACL statement was writing to the reject log as
6394 well as to the main log, which contradicts the documentation and doesn't
6395 seem right (because no rejection is happening). So I have stopped it.
6397 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6398 However, I am unable to do any testing of this.
6400 34. Fixed an infelicity in the appendfile transport. When checking directories
6401 for a mailbox, to see if any needed to be created, it was accidentally
6402 using path names with one or more superfluous leading slashes; tracing
6403 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6405 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6406 contents are added to the log line that is written for every discarded
6407 recipient. (Previously a log_message setting was ignored.)
6409 36. The ${quote: operator now quotes the string if it is empty.
6411 37. The install script runs exim in order to find its version number. If for
6412 some reason other than non-existence or emptiness, which it checks, it
6413 could not run './exim', it was installing it with an empty version number,
6414 i.e. as "exim-". This error state is now caught, and the installation is
6417 38. An argument was missing from the function that creates an error message
6418 when Exim fails to connect to the socket for saslauthd authentication.
6419 This could cause Exim to crash, or give a corrupted message.
6421 39. Added isip, isip4, and isip6 to ${if conditions.
6423 40. The ACL variables $acl_xx are now saved with the message, and can be
6424 accessed later in routers, transports, and filters.
6426 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6427 strings in the file are not string-expanded.
6429 42. If a MAIL command specified a SIZE value that was too large to fit into an
6430 int variable, the check against message_size_limit failed. Such values are
6431 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6432 day this will have to be increased, but I don't think I want to be around
6433 when emails are that large.
6440 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6441 iconv() is not standard in FreeBSD.
6443 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6444 IPv6 enabled. The observed symptom was a segmentation fault on return from
6445 the function os_common_find_running_interfaces() in src/os.c.
6447 3. In the check_special_case() function in daemon.c I had used "errno" as an
6448 argument name, which causes warnings on some systems. This was basically a
6449 typo, since it was named "eno" in the comments!
6451 4. The code that waits for the clock to tick (at a resolution of some fraction
6452 of a second) so as to ensure message-id uniqueness was always waiting for
6453 at least one whole tick, when it could have waited for less. [This is
6454 almost certainly not relevant at current processor speeds, where it is
6455 unlikely to ever wait at all. But we try to future-proof.]
6457 5. The function that sleeps for a time interval that includes fractions of a
6458 second contained a race. It did not block SIGALRM between setting the
6459 timer, and suspending (a couple of lines later). If the interval was short
6460 and the sigsuspend() was delayed until after it had expired, the suspension
6461 never ended. On busy systems this could lead to processes getting stuck for
6464 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6465 process, before it forks any delivery processes. The open lookup caching
6466 mechanism meant that the open file or database connection was passed into
6467 the delivery process. The problem was that delivery processes always tidy
6468 up cached lookup data. This could cause a problem for the next delivery
6469 process started by the queue runner, because the external queue runner
6470 process does not know about the closure. So the next delivery process
6471 still has data in the lookup cache. In the case of a file lookup, there was
6472 no problem because closing a file descriptor in a subprocess doesn't affect
6473 the parent. However, if the lookup was caching a connection to a database,
6474 the connection was closed, and the second delivery process was likely to
6475 see errors such as "PGSQL: query failed: server closed the connection
6476 unexpectedly". The problem has been fixed by closing all cached lookups
6477 in a queue runner before running a delivery process.
6479 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6480 seem to have the "const" qualifier which it has on other OS. I've
6483 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6484 *of the same type* (client or server) with the same public name that an
6485 error should be diagnosed.
6487 9. When Exim looked up a host name for an IP address, but failed to find the
6488 original IP address when looking up the host name (a safety check), it
6489 output the message "<ip address> does not match any IP for NULL", which was
6490 confusing, to say the least. The bug was that the host name should have
6491 appeared instead of "NULL".
6493 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6494 user that is built into the binary, and the -C or -D options is used, root
6495 privilege is dropped before the configuration file is read. In addition,
6496 logging is switched to stderr instead of the normal log files. If the
6497 configuration then re-defines the Exim user, the unprivileged environment
6498 is probably not what is expected, so Exim logs a panic warning message (but
6501 However, if deliver_drop_privilege is set, the unprivileged state may well
6502 be exactly what is intended, so the warning has been cut out in that case,
6503 and Exim is allowed to try to write to its normal log files.
6509 1. smtp_return_error_details was not giving details for temporary sender
6510 or receiver verification errors.
6512 2. Diagnose a configuration error if two authenticators have the same public
6515 3. Exim used not to create the message log file for a message until the first
6516 delivery attempt. This could be confusing when incoming messages were held
6517 for policy or load reasons. The message log file is now created at the time
6518 the message is received, and an initial "Received" line is written to it.
6520 4. The automatically generated man page for command line options had a minor
6521 bug that caused no ill effects; however, a more serious problem was that
6522 the procedure for building the man page automatically didn't always
6523 operate. Consequently, release 4.20 contains an out-of-date version. This
6524 shouldn't happen again.
6526 5. When building Exim with embedded Perl support, the script that builds the
6527 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6528 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6530 6. The freeze_tell option was not being used for messages that were frozen on
6531 arrival, either by an ACL or by local_scan().
6533 7. Added the smtp_incomplete_transaction log selector.
6535 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6536 was accepting AUTH without a new EHLO.
6538 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6539 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6542 10. Logging of TCP/IP connections (when configured) now happens in the main
6543 daemon process instead of the child process, so that the TCP/IP connection
6544 count is more accurate (but it can never be perfect).
6546 11. The use of "drop" in a nested ACL was not being handled correctly in the
6547 outer ACL. Now, if condition failure induced by the nested "drop" causes
6548 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6549 or "require"), the connection is dropped.
6551 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6552 that yield "discard" can now be used with an "accept" or a "discard" verb,
6553 but an error is generated for any others (because I can't see a useful way
6554 to define what should happen).
6556 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6557 are now processed in the same way as lines in the Exim configuration file.
6558 In particular, continuation lines are supported.
6560 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6562 15. Added -ti meaning -t -i.
6564 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6565 domains, and warn by logging if others are found.
6567 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6568 returned from GIFCONF when Exim is trying to find the list of interfaces on
6569 a host. The code in os.c has been modified to copy each ifreq to an aligned
6570 structure in all cases.
6572 Also, in some cases, the returned ifreq's were being copied to a 'struct
6573 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6574 means the last couple of bytes of an IPv6 address could be chopped if the
6575 ifreq contained only a normal sockaddr (14 bytes storage).
6577 18. Named domain lists were not supported in the hosts_treat_as_local option.
6578 An entry such as +xxxx was not recognized, and was treated as a literal
6581 19. Ensure that header lines added by a DATA ACL are included in the reject log
6582 if the ACL subsequently rejects the message.
6584 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6585 MD5 (which is deprecated).
6587 21. When testing a filter file using -bf, Exim was writing a message when it
6588 took the sender from a "From " line in the message, but it was not doing so
6589 when it took $return_path from a Return-Path: header line. It now does.
6591 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6592 with a valid header line field name (a series of printing characters
6593 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6595 23. Changed "disc" in the source to "disk" to conform to the documentation and
6596 the book and for uniformity.
6598 24. Ignore Sendmail's -Ooption=value command line item.
6600 25. When execve() failed while trying to run a command in a pipe transport,
6601 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6602 could be confused with a return value of 69 from the command itself. This
6603 has been changed to 127, the value the shell returns if it is asked to run
6604 a non-existent command. The wording for the related log line suggests a
6605 non-existent command as the problem.
6607 26. If received_header_text expands to an empty string, do not add a Received:
6608 header line to the message. (Well, it adds a token one on the spool, but
6609 marks it "old" so that it doesn't get used or transmitted.)
6611 27. Installed eximstats 1.28 (addition of -nt option).
6613 28. There was no check for failure on the call to getsockname() in the daemon
6614 code. This can fail if there is a shortage of resources on the system, with
6615 ENOMEM, for example. A temporary error is now given on failure.
6617 29. Contrary to the C standard, it seems that in some environments, the
6618 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6619 program. Exim now does this explicitly; it affects the formatting of
6620 timestamps using strftime().
6622 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6623 complaints. I've now initialized all the variables, to avoid this.
6625 32. Header lines added by a system filter were not being "seen" during
6626 transport-time rewrites.
6628 33. The info_callback() function passed to OpenSSL is set up with type void
6629 (*)(SSL *, int, int), as described somewhere. However, when calling the
6630 function (actually a macro) that sets it up, the type void(*)() is
6631 expected. I've put in a cast to prevent warnings from picky compilers.
6633 34. If a DNS black list lookup found a CNAME record, but there were no A
6634 records associated with the domain it pointed at, Exim crashed.
6636 35. If a DNS black list lookup returned more than one A record, Exim ignored
6637 all but the first. It now scans all returned addresses if a particular IP
6638 value is being sought. In this situation, the contents of the
6639 $dnslist_value variable are a list of all the addresses, separated by a
6642 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6643 to accept a host name and all its aliases if the forward lookup for any of
6644 them yielded the IP address of the incoming connection. Now it accepts only
6645 those names whose forward lookup yields the correct IP address. Any other
6646 names are discarded. This closes a loophole whereby a rogue DNS
6647 administrator could create reverse DNS records to break through a
6648 wildcarded host restriction in an ACL.
6650 37. If a user filter or a system filter that ran in a subprocess used any of
6651 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6652 the wrong values were passed to the pipe command ($thisaddress had the
6653 value of $0, $0 had the value of $1, etc). This bug was introduced by
6654 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6656 38. Improved the line breaking for long SMTP error messages from ACLs.
6657 Previously, if there was no break point between 40 and 75 characters, Exim
6658 left the rest of the message alone. Two changes have been made: (a) I've
6659 reduced the minimum length to 35 characters; (b) if it can't find a break
6660 point between 35 and 75 characters, it looks ahead and uses the first one
6661 that it finds. This may give the occasional overlong line, but at least the
6662 remaining text gets split now.
6664 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6665 file descriptors might be low, and that setting 1000 would always raise it.
6666 It turns out that in some environments, the limit is already over 1000 and
6667 that lowering it causes trouble. So now Exim takes care not to decrease it.
6669 40. When delivering a message, the value of $return_path is set to $sender_
6670 address at the start of routing (routers may change the value). By an
6671 oversight, this default was not being set up when an address was tested by
6672 -bt or -bv, which affected the outcome if any router or filter referred to
6675 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6676 log only when "message" or "log_message" are set. However, if one of the
6677 conditions was an address verification, or a call to a nested ACL, the
6678 messages generated by the underlying test were being passed through. This
6679 no longer happens. The underlying message is available in $acl_verify_
6680 message for both "message" and "log_message" expansions, so it can be
6681 passed through if needed.
6683 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6684 new expansion $bh_ to give the encoded byte string without charset
6685 translation. Translation happens only if iconv() is available; HAVE_ICONV
6686 indicates this at build time. HEADERS_CHARSET gives the charset to
6687 translate to; headers_charset can change it in the configuration, and
6688 "headers charset" can change it in an individual filter file.
6690 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6691 that creates RFC 2047 encoded "words" labels them as that charset instead
6692 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6693 expansion operator; (ii) when Exim creates a From: line for a local
6694 message; (iii) when a header line is rewritten to include a "phrase" part.
6696 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6697 buggy, causing it to skip the first lines of messages whose message ID
6698 ended in 'D'. This would not have bitten before Exim release 4.14, because
6699 message IDs were unlikely to end in 'D' before then. The effect was to have
6700 incorrect size information for certain domains.
6702 45. #include "config.h" was missing at the start of the crypt16.c module. This
6703 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6706 46. If there was a timeout during a "random" callout check, Exim treated it as
6707 a failure of the random address, and carried on sending RSET and the real
6708 address. If the delay was just some slowness somewhere, the response to the
6709 original RCPT would be taken as a response to RSET and so on, causing
6710 mayhem of various kinds.
6712 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6713 when I implemented it. It didn't allow for the fact that some option values
6714 may legitimately be negative (e.g. size_addition), and it didn't even do
6715 the right test for positive values.
6717 48. Domain names in DNS records are case-independent. Exim always looks them up
6718 in lower case. Some resolvers return domain names in exactly the case they
6719 appear in the zone file, that is, they may contain uppercase letters. Not
6720 all resolvers do this - some return always lower case. Exim was treating a
6721 change of case by a resolver as a change of domain, similar to a widening
6722 of a domain abbreviation. This triggered its re-routing code and so it was
6723 trying to route what was effectively the same domain again. This normally
6724 caused routing to fail (because the router wouldn't handle the domain
6725 twice). Now Exim checks for this case specially, and just changes the
6726 casing of the domain that it ultimately uses when it transmits the message
6729 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6732 50. If a filter generated a file delivery with a non-absolute name (possible if
6733 no home directory exists for the router), the forbid_file option was not
6736 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6737 the existing equality matching.
6739 52. Exim was using ints instead of ino_t variables in some places where it was
6740 dealing with inode numbers.
6742 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6743 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6744 called TMPDIR, and if it finds it is different, it changes its value.
6746 54. The smtp_printf() function is now made available to local_scan() so
6747 additional output lines can be written before returning. There is also an
6748 smtp_fflush() function to enable the detection of a dropped connection.
6749 The variables smtp_input and smtp_batched_input are exported to
6752 55. Changed the default runtime configuration: the message "Unknown user"
6753 has been removed from the ACL, and instead placed on the localuser router,
6754 using the cannot_route_message feature. This means that any verification
6755 failures that generate their own messages won't get overridden. Similarly,
6756 the "Unrouteable address" message that was in the ACL for unverifiable
6757 relay addresses has also been removed.
6759 56. Added hosts_avoid_esmtp to the smtp transport.
6761 57. The exicyclog script was not checking for the esoteric option
6762 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6763 will work only if exicyclog is run under the appropriate euid.
6765 58. Following a discussion on the list, the rules by which Exim recognises line
6766 endings on incoming messages have been changed. The -dropcr and drop_cr
6767 options are now no-ops, retained only for backwards compatibility. The
6768 following line terminators are recognized: LF CRLF CR. However, special
6769 processing applies to CR:
6771 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6772 nor a local message in the state where . is a terminator.
6774 (ii) If a bare CR is encountered in a header line, an extra space is added
6775 after the line terminator so as not to end the header. The reasoning
6776 behind this is that bare CRs in header lines are most likely either
6777 to be mistakes, or people trying to play silly games.
6779 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6780 was being incorrectly given as 18 bytes larger than it should have been.
6781 This is a VOB (very old bug).
6783 60. This may never have affected anything current, but just in case it has:
6784 When the local host is found other than at the start of a list of hosts,
6785 the local host, those with the same MX, and any that follow, are discarded.
6786 When the list in question was part of a longer list of hosts, the following
6787 hosts (not currently being processed) were also being discarded. This no
6788 longer happens. I'm not sure if this situation could ever has previously
6791 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6794 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6795 header. This is something that is recommended in a new Internet Draft, and
6796 is something that is documented as being done by Sendmail. There are two
6797 possible values. For messages generated by the autoreply transport, Exim
6800 Auto-Submitted: auto-replied
6802 whereas for all other generated messages (e.g. bounces) it adds
6804 Auto-Submitted: auto-generated
6806 63. The "personal" condition in filters now includes a test for the
6807 Auto-Submitted: header. If it contains the string "auto-" the message it
6808 not considered personal.
6810 64. Added rcpt_include_affixes as a generic transport option.
6812 65. Added queue_only_override (default true).
6814 66. Added the syslog_duplication option.
6816 67. If what should have been the first header line of a message consisted of
6817 a space followed by a colon, Exim was mis-interpreting it as a header line.
6818 It isn't of course - it is syntactically invalid and should therefore be
6819 treated as the start of the message body. The misbehaviour could have
6820 caused a number of strange effects, including loss of data in subsequent
6821 header lines, and spool format errors.
6823 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6824 client host had authenticated. This control can now be exercised by an ACL
6825 for more flexibility.
6827 69. By default, callouts do not happen when testing with -bh. There is now a
6828 variant, -bhc, which does actually run the callout code, including
6829 consulting and updating the callout cache.
6831 70. Added support for saslauthd authentication, courtesy of Alexander
6834 71. If statvfs() failed on the spool or log directories while checking their
6835 size for availability, Exim confusingly gave the error "space shortage".
6836 Furthermore, in debugging mode it crashed with a floating point exception.
6837 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6838 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6839 serious problem, Exim now writes to the main and panic logs when this
6840 happens, with details of the failure. It then refuses to accept the
6841 incoming message, giving the message "spool directory problem" or "log
6842 directory problem" with a 421 code for SMTP messages.
6844 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6845 0, 1, and 2 exist, because some OS complain for execs without them (see
6846 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6847 descriptors. However, the code omitted to check that the open succeeded,
6848 causing mysterious errors if for some reason the permissions on /dev/null
6849 got screwed. Now Exim writes a message to the main and panic logs, and
6850 bombs out if it can't open /dev/null.
6852 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6853 interact so that it is all more flexible. It is supposed to remain
6854 backwards compatible. Also added extra_local_interfaces.
6856 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6857 to bomb out with an assertion failure - to the client this appears as a
6858 connection drop. This problem occurs in the part of the code that was taken
6859 from the Samba project. Fortunately, the assertion is in a very simple
6860 function, so I have fixed this by reproducing the function inline in the
6861 one place where it is called, and arranging for authentication to fail
6862 instead of killing the process with assert().
6864 75. The SPA client code was not working when the server requested OEM rather
6865 than Unicode encoding.
6867 76. Added code to make require_files with a specific uid setting more usable in
6868 the case where statting the file as root fails - usually a non-root-mounted
6869 NFS file system. When this happens and the failure is EACCES, Exim now
6870 forks a subprocess and does the per-uid checking as the relevant uid.
6872 77. Added process_log_path.
6874 78. If log_file_path was not explicitly set, a setting of check_log_space or
6875 check_log_inodes was ignored.
6877 79. If a space check for the spool or log partitions fails, the incident is now
6878 logged. Of course, in the latter case the data may get lost...
6880 80. Added the %p formatting code to string_format() so that it can be used to
6881 print addresses in debug_print(). Adjusted all the address printing in the
6882 debugging in store.c to use %p rather than %d.
6884 81. There was a concern that a line of code in smtp_in.c could overflow a
6885 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6886 initially expressed, the concern was not well-founded, because trailing
6887 spaces are removed early. However, if the trailing spaces were followed by
6888 a NULL, they did not get removed, so the overflow was possible. Two fixes
6891 (a) I re-wrote the offending code in a cleaner fashion.
6892 (b) If an incoming SMTP command contains a NULL character, it is rejected
6895 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6896 runs initgroups(), so that if the Exim user is in any additional groups,
6897 they will be used during message reception.
6903 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob