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).
147 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
148 if was a fixed-sized field and bitmask ops via macros; it is now more
151 PP/07 GitHub PR 56: Apply MariaDB build fix.
152 Patch provided by Jaroslav Škarvada.
154 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
155 during Coverity cleanups [4.87 JH/47]
156 Diagnosis and fix provided by Michael Fischer v. Mollard.
158 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
159 the right size to place the terminating semicolon on its own folded
160 line, the header hash was calculated to an incorrect value thanks to
161 the (relaxed) space the fold became.
163 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
164 and confused the parent.
166 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
167 which could crash as a result. This could lead to undeliverable messages.
169 JH/28 Logging: "next input sent too soon" now shows where input was truncated
172 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
173 matters on fast-turnover and PID-randomising systems, which were getting
174 out-of-order delivery.
180 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
181 than -2003 did; needs libidn2 in addition to libidn.
183 JH/02 The path option on a pipe transport is now expanded before use.
185 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
186 Patch provided by "Björn", documentation fix added too.
188 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
189 missing a wire-to-host endian conversion.
191 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
192 close after a BDAT command line could be taken as a following command,
193 giving a synch failure. Fix by only checking for synch immediately
194 before acknowledging the chunk.
196 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
197 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
198 macro. Patches provided by Josh Soref.
200 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
201 Previously we did not; the RFC seems ambiguous and VRFY is not listed
202 by IANA as a service extension. However, John Klensin suggests that we
205 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
206 the dkim code may be unix-mode line endings rather than smtp wire-format
207 CRLF, so prepend a CR to any bare LF.
209 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
210 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
212 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
213 the first were themselves being wrongly included in the feed into dkim
214 processing; with most chunk sizes in use this resulted in an incorrect
215 body hash calculated value.
217 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
218 DKIM signature block, for verification. Although advised against by
219 standards it is specifically not ruled illegal.
221 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
223 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
224 missing a body hash (the bh= tag).
226 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
227 It seems that HAProxy sends the Proxy Protocol information in clear and
228 only then does a TLS startup, so do the same.
230 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
231 TCP connections (such as for Spamd) unless the daemon successfully set
232 Fast Open mode on its listening sockets. This fixes breakage seen on
233 too-old kernels or those not configured for Fast Open, at the cost of
234 requiring both directions being enabled for TFO, and TFO never being used
235 by non-daemon-related Exim processes.
237 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
238 endings, at least on the first header line. Try to canonify any that get
239 past that check, despite the cost.
241 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
242 now limited to an arbitrary five deep, while parsing addresses with the
243 strip_excess_angle_brackets option enabled.
245 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
246 instead leave the unprompted TLS handshake in socket buffer for the
247 TLS library to consume.
249 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
251 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
253 JH/16 Drop variables when they go out of scope. Memory management drops a whole
254 region in one operation, for speed, and this leaves assigned pointers
255 dangling. Add checks run only under the testsuite which checks all
256 variables at a store-reset and panics on a dangling pointer; add code
257 explicitly nulling out all the variables discovered. Fixes one known
258 bug: a transport crash, where a dangling pointer for $sending_ip_address
259 originally assigned in a verify callout, is re-used.
261 PP/06 Drop '.' from @INC in various Perl scripts.
263 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
265 PP/08 Reduce a number of compilation warnings under clang; building with
266 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
267 should be warning-free.
269 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
271 HS/01 Fix portability problems introduced by PP/08 for platforms where
272 realloc(NULL) is not equivalent to malloc() [SunOS et al].
274 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
275 chunk. This allows us to accept broken chunked messages. We need a more
276 general solution here.
278 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
279 already-broken messages in the queue.
281 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
283 JH/19 Fix reference counting bug in routing-generated-address tracking.
289 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
290 supports it and a size is available (ie. the sending peer gave us one).
292 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
293 years of being deprecated). The replacements are the ACLs
294 acl_smtp_mime and acl_not_smtp_mime.
296 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
297 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
298 or even an in-clear connection were permitted. Now, if the host lookup
299 was dnssec and dane was requested then the host is only used if the
300 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
301 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
302 if one fails this test.
303 This means that a poorly-configured remote DNS will make it incommunicado;
304 but it protects against a DNS-interception attack on it.
306 JH/04 Bug 1810: make continued-use of an open smtp transport connection
307 non-noisy when a race steals the message being considered.
309 JH/05 If main configuration option tls_certificate is unset, generate a
310 self-signed certificate for inbound TLS connections.
312 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
313 in rewrites and routers.
315 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
316 and logged a warning sing 4.83; now they are a configuration file error.
318 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
319 (lacking @domain). Apply the same qualification processing as RCPT.
321 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
323 JH/10 Support ${sha256:} applied to a string (as well as the previous
326 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
327 a cutthrough deliver is pending, as we always want to make a connection.
328 This also avoids re-routing the message when later placing the cutthrough
329 connection after a verify cache hit.
330 Do not update it with the verify result either.
332 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
333 when routing results in more than one destination address.
335 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
336 signing (which inhibits the cutthrough capability). Previously only
337 the presence of an option was tested; now an expansion evaluating as
338 empty is permissible (obviously it should depend only on data available
339 when the cutthrough connection is made).
341 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
342 the relevant preceding SMTP command did not note the pipelining mode.
344 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
345 Previously they were not counted.
347 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
348 as one having no matching records. Previously we deferred the message
349 that needed the lookup.
351 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
352 distinguished as "(=".
354 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
355 for missing MX records. Previously it only worked for missing A records.
357 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
359 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
360 after the data-go-ahead and data-ack. Patch from Jason Betts.
362 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
363 even for a "none" policy. Patch from Tony Meyer.
365 JH/22 Fix continued use of a connection for further deliveries. If a port was
366 specified by a router, it must also match for the delivery to be
369 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
370 When one of the recipients of a message was unsuitable for the connection
371 (has no matching addresses), we lost track of needing to mark it
372 deferred. As a result mail would be lost.
374 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
376 JH/25 Decoding ACL controls is now done using a binary search; the source code
377 takes up less space and should be simpler to maintain. Merge the ACL
378 condition decode tables also, with similar effect.
380 JH/26 Fix problem with one_time used on a redirect router which returned the
381 parent address unchanged. A retry would see the parent address marked as
382 delivered, so not attempt the (identical) child. As a result mail would
385 JH/27 Fix a possible security hole, wherein a process operating with the Exim
386 UID can gain a root shell. Credit to http://www.halfdog.net/ for
387 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
390 JH/28 Enable {spool,log} filesystem space and inode checks as default.
391 Main config options check_{log,spool}_{inodes,space} are now
392 100 inodes, 10MB unless set otherwise in the configuration.
394 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
395 Previously it only applied to the main-section connection policy
398 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
400 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
401 by me. Added RFC7919 DH primes as an alternative.
403 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
404 are not in the system include path.
406 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
407 GnuTLS, when a session startup failed (eg because the client disconnected)
408 Exim did stdio operations after fclose. This was exposed by a recent
409 change which nulled out the file handle after the fclose.
411 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
412 signed directly by the cert-signing cert, rather than an intermediate
413 OCSP-signing cert. This is the model used by LetsEncrypt.
415 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
417 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
418 an incoming connection.
420 HS/02 Bug 1802: Do not half-close the connection after sending a request
423 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
424 fallback to "prime256v1".
426 JH/34 SECURITY: Use proper copy of DATA command in error message.
427 Could leak key material. Remotely exploitable. CVE-2016-9963.
433 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
434 and 3.4.4 - once the server is enabled to respond to an OCSP request
435 it does even when not requested, resulting in a stapling non-aware
436 client dropping the TLS connection.
438 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
439 support variable-length bit vectors. No functional change.
441 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
442 The I= interface field on outgoing lines is now after the H= remote
443 host field, same as incoming lines. There is a separate
444 outgoing_interface log selector which allows you to disable the
447 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
448 If not running log_selector +smtp_connection the mainlog would be held
449 open indefinitely after a "too many connections" event, including to a
450 deleted file after a log rotate. Leave the per net connection logging
451 leaving it open for efficiency as that will be quickly detected by the
452 check on the next write.
454 HS/01 Bug 1671: Fix post transport crash.
455 Processing the wait-<transport> messages could crash the delivery
456 process if the message IDs didn't exist for some reason. When
457 using 'split_spool_directory=yes' the construction of the spool
458 file name failed already, exposing the same netto behaviour.
460 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
461 mime_regex ACL conditions.
463 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
464 to DSN fail messages (bounces): remote IP, remote greeting, remote response
465 to HELO, local diagnostic string.
467 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
468 log line to debug. Even when configured with a tls authenticator many
469 client connections are expected to not authenticate in this way, so
470 an authenticate fail is not an error.
472 HS/02 Add the Exim version string to the process info. This way exiwhat
473 gives some more detail about the running daemon.
475 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
476 matter for fast-change records such as DNSBLs.
478 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
479 retry record, even if constant. There may be multiple transports with
480 different interface settings and the retry behaviour needs to be kept
483 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
485 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
487 JH/10 Bug 840: fix log_defer_output option of pipe transport
489 JH/11 Bug 830: use same host for all RCPTS of a message, even under
490 hosts_randomize. This matters a lot when combined with mua_wrapper.
492 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
493 ${quote_pgsql:<string>} operator.
495 JH/13 Bug 1708: avoid misaligned access in cached lookup.
497 JH/14 Change header file name for freeradius-client. Relevant if compiling
498 with Radius support; from the Gentoo tree and checked under Fedora.
500 JH/15 Bug 1712: Introduce $prdr_requested flag variable
502 JH/16 Bug 1714: Permit an empty string as expansion result for transport
503 option transport_filter, meaning no filtering.
505 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
507 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
508 defaults to "*" (all hosts). The variable is now available when not built
509 with TLS, default unset, mainly to enable keeping the testsuite sane.
510 If a server certificate is not supplied (via tls_certificate) an error is
511 logged, and clients will find TLS connections fail on startup. Presumably
512 they will retry in-clear.
513 Packagers of Exim are strongly encouraged to create a server certificate
514 at installation time.
516 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
517 with the $config_file variable.
519 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
520 in transport context, after the attempt, and per-recipient. The latter type
521 is per host attempted. The event data is the error message, and the errno
522 information encodes the lookup type (A vs. MX) used for the (first) host,
523 and the trailing two digits of the smtp 4xx response.
525 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
526 to write to mainlog (or rejectlog, paniclog) in the window between file
527 creation and permissions/ownership being changed. Particularly affects
528 installations where exicyclog is run as root, rather than exim user;
529 result is that the running daemon panics and dies.
531 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
533 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
534 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
535 "pri" and "weight". Note that the previous implicit priority given by the
536 list order is no longer honoured.
538 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
541 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
542 by defining SUPPORT_SOCKS.
544 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
545 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
546 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
547 variables are renamed to proxy_{local,external}_{address,port}.
549 JH/27 Move Internationalisation support from Experimental to mainline, enabled
550 for a build by defining SUPPORT_I18N
552 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
553 of the query string, and make ${quote_redis:} do that quoting.
555 JH/29 Move Events support from Experimental to mainline, enabled by default
556 and removable for a build by defining DISABLE_EVENT.
558 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
560 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
561 cached by the daemon.
563 JH/32 Move Redis support from Experimental to mainline, enabled for a build
564 by defining LOOKUP_REDIS. The libhiredis library is required.
566 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
567 keys are given for lookup.
569 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
570 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
571 only supported when built with TLS support. The PolarSSL SHA routines
572 are still used when the TLS library is too old for convenient support.
574 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
575 openssl_options), for security. OpenSSL forces this from version 1.1.0
576 server-side so match that on older versions.
578 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
579 allocation for $value could be released as the expansion processing
580 concluded, but leaving the global pointer active for it.
582 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
583 and to use the domains and local_parts ACL conditions.
585 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
586 incorrectly not doubled on cutthrough transmission, hence seen as a
587 body-termination at the receiving system - resulting in truncated mails.
588 Commonly the sender saw a TCP-level error, and retransmitted the message
589 via the normal store-and-forward channel. This could result in duplicates
590 received - but deduplicating mailstores were liable to retain only the
591 initial truncated version.
593 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
595 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
597 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
598 we're in there, support oversigning also; bug 1309.
600 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
602 HS/04 Add support for keep_environment and add_environment options.
604 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
605 either intentional arithmetic overflow during PRNG, or testing config-
608 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
609 delivery resulted in actual delivery. Cancel cutthrough before DATA
612 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
613 rejects a recipient: pass the reject to the originator.
615 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
616 Many were false-positives and ignorable, but it's worth fixing the
619 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
620 for the new environment-manipulation done at startup. Move the routines
621 from being local to tls.c to being global via the os.c file.
623 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
624 an extract embedded as result-arg for a map, the first arg for extract
625 is unavailable so we cannot tell if this is a numbered or keyed
626 extraction. Accept either.
632 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
635 JH/02 The smtp transport option "multi_domain" is now expanded.
637 JH/03 The smtp transport now requests PRDR by default, if the server offers
640 JH/04 Certificate name checking on server certificates, when exim is a client,
641 is now done by default. The transport option tls_verify_cert_hostnames
642 can be used to disable this per-host. The build option
643 EXPERIMENTAL_CERTNAMES is withdrawn.
645 JH/05 The value of the tls_verify_certificates smtp transport and main options
646 default to the word "system" to access the system default CA bundle.
647 For GnuTLS, only version 3.0.20 or later.
649 JH/06 Verification of the server certificate for a TLS connection is now tried
650 (but not required) by default. The verification status is now logged by
651 default, for both outbound TLS and client-certificate supplying inbound
654 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
657 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
658 Status Notification (bounce) messages are now MIME format per RFC 3464.
659 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
660 under the control of the dsn_advertise_hosts option, and routers may
661 have a dsn_lasthop option.
663 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
664 default, modifiable by a malware= option. The list separator for
665 the options can now be changed in the usual way. Bug 68.
667 JH/10 The smtp_receive_timeout main option is now expanded before use.
669 JH/11 The incoming_interface log option now also enables logging of the
670 local interface on delivery outgoing connections.
672 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
673 if the interface and destination host and port all match.
675 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
678 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
679 Patch from Andrew Lewis.
681 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
682 now supports optional time-restrictions, weighting, and priority
683 modifiers per server. Patch originally by <rommer@active.by>.
685 JH/16 The spamd_address main option now supports a mixed list of local
686 and remote servers. Remote servers can be IPv6 addresses, and
687 specify a port-range.
689 JH/17 Bug 68: The spamd_address main option now supports an optional
690 timeout value per server.
692 JH/18 Bug 1581: Router and transport options headers_add/remove can
693 now have the list separator specified.
695 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
698 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
701 JH/21 Support for the A6 type of dns record is withdrawn.
703 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
704 rather than the verbs used.
706 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
707 from 255 to 1024 chars.
709 JH/24 Verification callouts now attempt to use TLS by default.
711 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
712 are generic router options now. The defaults didn't change.
714 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
715 Original patch from Alexander Shikoff, worked over by JH.
717 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
718 files if ZCAT_COMMAND is not executable.
720 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
722 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
724 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
725 Normally benign, it bites when the pair was led to by a CNAME;
726 modern usage is to not canonicalize the domain to a CNAME target
727 (and we were inconsistent anyway for A-only vs AAAA+A).
729 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
731 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
732 when evaluating $sender_host_dnssec.
734 JH/31 Check the HELO verification lookup for DNSSEC, adding new
735 $sender_helo_dnssec variable.
737 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
739 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
741 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
743 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
744 documented as working, but never had. Support all but $spam_report.
746 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
747 added for tls authenticator.
749 HS/03 Add perl_taintmode main config option
755 TL/01 When running the test suite, the README says that variables such as
756 no_msglog_check are global and can be placed anywhere in a specific
757 test's script, however it was observed that placement needed to be near
758 the beginning for it to behave that way. Changed the runtest perl
759 script to read through the entire script once to detect and set these
760 variables, reset to the beginning of the script, and then run through
761 the script parsing/test process like normal.
763 TL/02 The BSD's have an arc4random API. One of the functions to induce
764 adding randomness was arc4random_stir(), but it has been removed in
765 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
766 function when detected.
768 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
769 cause callback expansion.
771 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
772 syntax errors in an expansion can be treated as a string instead of
773 logging or causing an error, due to the internal use of bool_lax
774 instead of bool when processing it.
776 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
777 server certificates when making smtp deliveries.
779 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
781 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
783 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
785 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
786 Merged patch from Sebastian Wiedenroth.
788 JH/05 Fix results-pipe from transport process. Several recipients, combined
789 with certificate use, exposed issues where response data items split
790 over buffer boundaries were not parsed properly. This eventually
791 resulted in duplicates being sent. This issue only became common enough
792 to notice due to the introduction of connection certificate information,
793 the item size being so much larger. Found and fixed by Wolfgang Breyha.
795 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
796 size buffer was used, resulting in syntax errors when an expansion
799 JH/07 Add support for directories of certificates when compiled with a GnuTLS
800 version 3.3.6 or later.
802 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
803 is EXPERIMENTAL_EVENT, the main-configuration and transport options
804 both become "event_action", the variables become $event_name, $event_data
805 and $event_defer_errno. There is a new variable $verify_mode, usable in
806 routers, transports and related events. The tls:cert event is now also
807 raised for inbound connections, if the main configuration event_action
810 TL/06 In test suite, disable OCSP for old versions of openssl which contained
811 early OCSP support, but no stapling (appears to be less than 1.0.0).
813 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
814 server certificate names available under the smtp transport option
815 "tls_verify_cert_hostname" now do not permit multi-component wildcard
818 JH/10 Time-related extraction expansions from certificates now use the main
819 option "timezone" setting for output formatting, and are consistent
820 between OpenSSL and GnuTLS compilations. Bug 1541.
822 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
823 encoded parameter in the incoming message. Bug 1558.
825 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
826 include certificate info, eximon was claiming there were spoolfile
829 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
831 JH/14 Log delivery-related information more consistently, using the sequence
832 "H=<name> [<ip>]" wherever possible.
834 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
835 are problematic for Debian distribution, omit them from the release
838 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
840 JH/16 Fix string representation of time values on 64bit time_t architectures.
843 JH/17 Fix a null-indirection in certextract expansions when a nondefault
844 output list separator was used.
849 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
850 checkers that were complaining about end of non-void function with no
853 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
854 This was a regression introduced in 4.83 by another bugfix.
856 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
858 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
859 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
865 TF/01 Correctly close the server side of TLS when forking for delivery.
867 When a message was received over SMTP with TLS, Exim failed to clear up
868 the incoming connection properly after forking off the child process to
869 deliver the message. In some situations the subsequent outgoing
870 delivery connection happened to have the same fd number as the incoming
871 connection previously had. Exim would try to use TLS and fail, logging
872 a "Bad file descriptor" error.
874 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
875 utilities have not been installed.
877 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
878 temporary space as the ACL may create new global variables.
880 TL/01 LDAP support uses per connection or global context settings, depending
881 upon the detected version of the libraries at build time.
883 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
884 to extract and use the src ip:port in logging and expansions as if it
885 were a direct connection from the outside internet. PPv2 support was
886 updated based on HAProxy spec change in May 2014.
888 JH/02 Add ${listextract {number}{list}{success}{fail}}.
890 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
891 Properly escape header and check for NULL return.
893 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
896 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
898 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
899 characters in header names, implemented as a verify condition.
900 Contributed by Michael Fischer v. Mollard.
902 TL/05 Rename SPF condition results err_perm and err_temp to standardized
903 results permerror and temperror. Previous values are deprecated but
904 still accepted. In a future release, err_perm and err_temp will be
905 completely removed, which will be a backward incompatibility if the
906 ACL tests for either of these two old results. Patch contributed by
907 user bes-internal on the mailing list.
909 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
911 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
912 selectors, in both main and reject logs.
914 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
917 JH/07 Add malware type "sock" for talking to simple daemon.
919 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
921 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
922 routers/transports under cutthrough routing.
924 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
925 numbers. Touch up "bool" conditional to keep the same definition.
927 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
929 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
931 JH/12 Expand items in router/transport headers_add or headers_remove lists
932 individually rather than the list as a whole. Bug 1452.
934 Required for reasonable handling of multiple headers_ options when
935 they may be empty; requires that headers_remove items with embedded
936 colons must have them doubled (or the list-separator changed).
938 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
939 view the policy declared in the DMARC record. Currently, $dmarc_status
940 is a combined value of both the record presence and the result of the
943 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
945 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
946 dnslookup router and the smtp transport (applying to the forward
949 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
950 of ldap servers used for a specific lookup. Patch provided by Heiko
953 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
954 New variable $lookup_dnssec_authenticated for observability.
956 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
957 Patch submitted by Lars Timman.
959 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
961 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
962 Requires trusted mode and valid format message id, aborts otherwise.
963 Patch contributed by Heiko Schlichting.
965 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
966 certextract with support for various fields. Bug 1358.
968 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
969 is requested by default, modifiable by smtp transport option
972 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
973 operate on certificate variables to give certificate fingerprints
974 Also new ${sha256:cert_variable}.
976 JH/23 The PRDR feature is moved from being Experimental into the mainline.
978 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
979 Christian Aistleitner.
981 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
983 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
984 file. Patch from Wolfgang Breyha.
986 JH/25 Expand the coverage of the delivery $host and $host_address to
987 client authenticators run in verify callout. Bug 1476.
989 JH/26 Port service names are now accepted for tls_on_connect_ports, to
990 align with daemon_smtp_ports. Bug 72.
992 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
993 support and error reporting did not work properly.
995 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
996 and is readable. Patch from Andrew Colin Kissa.
998 TL/14 Enhance documentation of ${run expansion and how it parses the
999 commandline after expansion, particularly in the case when an
1000 unquoted variable expansion results in an empty value.
1002 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1004 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1005 by renaming away from it. Addresses GH issue 15, reported by
1008 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1010 TL/15 SECURITY: prevent double expansion in math comparison functions
1011 (can expand unsanitized data). Not remotely exploitable.
1018 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1020 PP/02 Make -n do something, by making it not do something.
1021 When combined with -bP, the name of an option is not output.
1023 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1026 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1027 $sender_host_name and config options to manage this, and basic check
1030 PP/05 DSCP support for outbound connections and control modifier for inbound.
1032 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1033 (Only plugin which currently uses this is kerberos4, which nobody should
1034 be using, but we should make it available and other future plugins might
1035 conceivably use it, even though it would break NAT; stuff *should* be
1036 using channel bindings instead).
1038 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1039 name; added for Sendmail compatibility; requires admin caller.
1040 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1041 just ignore it); requires trusted caller.
1042 Also parse but ignore: -Ac -Am -X<logfile>
1045 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1047 TL/02 Add +smtp_confirmation as a default logging option.
1049 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1050 Patch by Magnus Holmgren from 2007-02-20.
1052 TL/04 Bugzilla 1281 - Spec typo.
1053 Bugzilla 1283 - Spec typo.
1054 Bugzilla 1290 - Spec grammar fixes.
1056 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1058 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1060 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1061 mailing list by by Dmitry Isaikin.
1063 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1065 JH/02 Support "G" suffix to numbers in ${if comparisons.
1067 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1069 NM/01 Bugzilla 1197 - Spec typo
1070 Bugzilla 1196 - Spec examples corrections
1072 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1074 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1075 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1078 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1079 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1081 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1082 "acl {{name}{arg}...}", and optional args on acl condition
1085 JH/05 Permit multiple router/transport headers_add/remove lines.
1087 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1089 JH/07 Avoid using a waiting database for a single-message-only transport.
1090 Performance patch from Paul Fisher. Bugzilla 1262.
1092 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1095 JH/09 Add $headers_added variable, with content from use of ACL modifier
1096 add_header (but not yet added to the message). Bugzilla 199.
1098 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1099 Pulled from Bugzilla 817 by Wolfgang Breyha.
1101 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1103 (nb: this is the same fix as in Exim 4.80.1)
1105 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1108 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1111 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1113 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1114 advertise SMTP AUTH mechanism to us, instead of a generic
1115 protocol violation error. Also, make Exim more robust to bad
1116 data from the Dovecot auth socket.
1118 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1120 When a queue runner is handling a message, Exim first routes the
1121 recipient addresses, during which it prunes them based on the retry
1122 hints database. After that it attempts to deliver the message to
1123 any remaining recipients. It then updates the hints database using
1126 So if a recipient address works intermittently, it can get repeatedly
1127 deferred at routing time. The retry hints record remains fresh so the
1128 address never reaches the final cutoff time.
1130 This is a fairly common occurrence when a user is bumping up against
1131 their storage quota. Exim had some logic in its local delivery code
1132 to deal with this. However it did not apply to per-recipient defers
1133 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1135 This change adds a proper retry rule check during routing so that the
1136 final cutoff time is checked against the message's age. We only do
1137 this check if there is an address retry record and there is not a
1138 domain retry record; this implies that previous attempts to handle
1139 the address had the retry_use_local_parts option turned on. We use
1140 this as an approximation for the destination being like a local
1141 delivery, as in LMTP.
1143 I suspect this new check makes the old local delivery cutoff check
1144 redundant, but I have not verified this so I left the code in place.
1146 TF/02 Correct gecos expansion when From: is a prefix of the username.
1148 Test 0254 submits a message to Exim with the header
1152 When I ran the test suite under the user fanf2, Exim expanded
1153 the header to contain my full name, whereas it should have added
1154 a Resent-Sender: header. It erroneously treats any prefix of the
1155 username as equal to the username.
1157 This change corrects that bug.
1159 GF/01 DCC debug and logging tidyup
1160 Error conditions log to paniclog rather than rejectlog.
1161 Debug lines prefixed by "DCC: " to remove any ambiguity.
1163 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1165 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1166 Bug spotted by Jeremy Harris; was flawed since initial commit.
1167 Would have resulted in OCSP responses post-SNI triggering an Exim
1168 NULL dereference and crash.
1170 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1172 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1173 Bug detection, analysis and fix by Samuel Thibault.
1174 Bugzilla 1331, Debian bug #698092.
1176 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1178 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1179 Server implementation by Todd Lyons, client by JH.
1180 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1181 config variable "prdr_enable" controls whether the server
1182 advertises the facility. If the client requests PRDR a new
1183 acl_data_smtp_prdr ACL is called once for each recipient, after
1184 the body content is received and before the acl_smtp_data ACL.
1185 The client is controlled by both of: a hosts_try_prdr option
1186 on the smtp transport, and the server advertisement.
1187 Default client logging of deliveries and rejections involving
1188 PRDR are flagged with the string "PRDR".
1190 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1191 fclose(). Diagnosis by Todd Lyons.
1193 PP/17 Update configure.default to handle IPv6 localhost better.
1194 Patch by Alain Williams (plus minor tweaks).
1197 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1198 This is now consistent with GnuTLS, and is now documented: the
1199 previous undocumented portable approach to treating the option as
1200 unset was to force an expansion failure. That still works, and
1201 an empty string is now equivalent.
1203 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1204 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1205 not performing validation itself.
1207 PP/20 Added force_command boolean option to pipe transport.
1208 Patch from Nick Koston, of cPanel Inc.
1210 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1213 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1215 PP/21 Fix eximon continuous updating with timestamped log-files.
1216 Broken in a format-string cleanup in 4.80, missed when I repaired the
1217 other false fix of the same issue.
1218 Report and fix from Heiko Schlichting.
1221 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1222 Report from Prashanth Katuri.
1224 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1225 It's SecureTransport, so affects any MacOS clients which use the
1226 system-integrated TLS libraries, including email clients.
1228 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1229 using a MIME ACL for non-SMTP local injection.
1230 Report and assistance in diagnosis by Warren Baker.
1232 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1234 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1236 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1237 last id that failed so it may be referenced in subsequent ACL's.
1239 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1242 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1243 ldap library initialization, allowing self-signed CA's to be
1244 used. Also properly sets require_cert option later in code by
1245 using NULL (global ldap config) instead of ldap handle (per
1246 session). Bug diagnosis and testing by alxgomz.
1248 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1249 the src/util/ subdirectory.
1251 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1252 renamed to Transport Post Delivery Action by Jeremy Harris, as
1255 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1256 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1257 redis_servers = needs to be configured which will be used by the redis
1258 lookup. Patch from Warren Baker, of The Packet Hub.
1260 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1262 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1263 hostname or reverse DNS when processing a host list. Used suggestions
1264 from multiple comments on this bug.
1266 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1268 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1269 Missed a few lines, added it to make the runtest require no keyboard
1272 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1273 contains upper case chars. Make router use caseful_local_part.
1275 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1276 support when GnuTLS has been built with p11-kit.
1282 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1284 This, or similar/improved, will also be change PP/11 of 4.82.
1290 PP/01 Handle short writes when writing local log-files.
1291 In practice, only affects FreeBSD (8 onwards).
1292 Bugzilla 1053, with thanks to Dmitry Isaikin.
1294 NM/01 Bugzilla 949 - Documentation tweak
1296 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1299 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1301 PP/02 Implemented gsasl authenticator.
1303 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1305 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1306 `pkg-config foo` for cflags/libs.
1308 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1309 with rest of GSASL and with heimdal_gssapi.
1311 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1312 `pkg-config foo` for cflags/libs for the TLS implementation.
1314 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1315 properties get this fed in as external SSF. A number of robustness
1316 and debugging improvements to the cyrus_sasl authenticator.
1318 PP/08 cyrus_sasl server now expands the server_realm option.
1320 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1321 Patch by Jeremy Harris.
1323 PP/10 Added dbmjz lookup type.
1325 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1327 PP/12 MAIL args handles TAB as well as SP, for better interop with
1328 non-compliant senders.
1329 Analysis and variant patch by Todd Lyons.
1331 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1332 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1333 Patch from Dirk Mueller <dmueller@suse.com>
1335 PP/13 tls_peerdn now print-escaped for spool files.
1336 Observed some $tls_peerdn in wild which contained \n, which resulted
1337 in spool file corruption.
1339 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1340 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1341 or write after TLS renegotiation, which otherwise led to messages
1344 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1345 as a tracking header (ie: a signed header comes before the signature).
1346 Patch from Wolfgang Breyha.
1348 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1349 comma-sep list; embedded commas doubled.
1351 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1353 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1355 Report and patch from Dmitry Banschikov.
1357 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1358 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1359 failures. We appear to now support TLS1.1+ with Exim.
1361 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1362 lets Exim select keys and certificates based upon TLS SNI from client.
1363 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1364 before an outbound SMTP session. New log_selector, +tls_sni.
1366 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1367 NULL dereference. Report and patch from Alun Jones.
1369 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1370 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1371 Not seeing resolver debug output on NetBSD, but suspect this is a
1372 resolver implementation change.
1374 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1375 Left warnings. Added "eximon gdb" invocation mode.
1377 PP/21 Defaulting "accept_8bitmime" to true, not false.
1379 PP/22 Added -bw for inetd wait mode support.
1381 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1382 locate the relevant includes and libraries. Made this the default.
1384 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1385 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1387 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1388 This may cause build issues on older platforms.
1390 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1391 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1392 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1393 Added SNI support via GnuTLS too.
1394 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1396 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1398 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1399 Applied second patch from Janne, implementing suggestion to default
1400 multiple-strings-in-record handling to match SPF spec.
1402 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1404 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1405 read-only, out of scope).
1406 Patch from Wolfgang Breyha, report from Stuart Northfield.
1408 PP/29 Fix three issues highlighted by clang analyser static analysis.
1409 Only crash-plausible issue would require the Cambridge-specific
1410 iplookup router and a misconfiguration.
1411 Report from Marcin Mirosław.
1413 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1415 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1416 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1417 As part of this, removing so much warning spew let me fix some minor
1418 real issues in debug logging.
1420 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1421 assignment on my part. Fixed.
1423 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1424 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1425 Janne Snabb (who went above and beyond: thank you).
1427 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1428 string otherwise requires a connection and a bunch more work and it's
1429 relatively easy to get wrong. Should also expose TLS library linkage
1432 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1433 64-bit ${eval} (JH/03).
1435 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1436 GNU libc to support some of the 64-bit stuff, should not lead to
1437 conflicts. Defined before os.h is pulled in, so if a given platform
1438 needs to override this, it can.
1440 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1441 protection layer was required, which is not implemented.
1442 Bugzilla 1254, patch from Wolfgang Breyha.
1444 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1445 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1446 tls_dhparam take prime identifiers. Also unbreak combination of
1447 OpenSSL+DH_params+TLSSNI.
1449 PP/39 Disable SSLv2 by default in OpenSSL support.
1455 PP/01 Solaris build fix for Oracle's LDAP libraries.
1456 Bugzilla 1109, patch from Stephen Usher.
1458 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1460 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1463 TF/02 Fix a couple more cases where we did not log the error message
1464 when unlink() failed. See also change 4.74-TF/03.
1466 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1467 lock up or crash if it happened to be inside a call to libc when it
1468 got a SIGUSR1 from exiwhat.
1470 The SIGUSR1 handler appends the current process status to the process
1471 log which is later printed by exiwhat. It used to use the general
1472 purpose logging code to do this, but several functions it calls are
1473 not safe for signals.
1475 The new output code in the SIGUSR1 handler is specific to the process
1476 log, and simple enough that it's easy to inspect for signal safety.
1477 Removing some special cases also simplifies the general logging code.
1478 Removing the spurious timestamps from the process log simplifies
1481 TF/04 Improved ratelimit ACL condition.
1483 The /noupdate option has been deprecated in favour of /readonly which
1484 has clearer semantics. The /leaky, /strict, and /readonly update modes
1485 are mutually exclusive. The update mode is no longer included in the
1486 database key; it just determines when the database is updated. (This
1487 means that when you upgrade Exim will forget old rate measurements.)
1489 Exim now checks that the per_* options are used with an update mode that
1490 makes sense for the current ACL. For example, when Exim is processing a
1491 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1492 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1493 must specify per_mail/readonly. If you omit the update mode it defaults to
1494 /leaky where that makes sense (as before) or /readonly where required.
1496 The /noupdate option is now undocumented but still supported for
1497 backwards compatibility. It is equivalent to /readonly except that in
1498 ACLs where /readonly is required you may specify /leaky/noupdate or
1499 /strict/noupdate which are treated the same as /readonly.
1501 A useful new feature is the /count= option. This is a generalization
1502 of the per_byte option, so that you can measure the throughput of other
1503 aggregate values. For example, the per_byte option is now equivalent
1504 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1506 The per_rcpt option has been generalized using the /count= mechanism
1507 (though it's more complicated than the per_byte equivalence). When it is
1508 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1509 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1510 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1511 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1512 Note that using per_rcpt with a non-readonly update mode in more than
1513 one ACL will cause the recipients to be double-counted. (The per_mail
1514 and per_byte options don't have this problem.)
1516 The handling of very low rates has changed slightly. If the computed rate
1517 is less than the event's count (usually one) then this event is the first
1518 after a long gap. In this case the rate is set to the same as this event's
1519 count, so that the first message of a spam run is counted properly.
1521 The major new feature is a mechanism for counting the rate of unique
1522 events. The new per_addr option counts the number of different
1523 recipients that someone has sent messages to in the last time period. It
1524 behaves like per_rcpt if all the recipient addresses are different, but
1525 duplicate recipient addresses do not increase the measured rate. Like
1526 the /count= option this is a general mechanism, so the per_addr option
1527 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1528 example, measure the rate that a client uses different sender addresses
1529 with the options per_mail/unique=$sender_address. There are further
1530 details in the main documentation.
1532 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1534 TF/06 Removed a few PCRE remnants.
1536 TF/07 Automatically extract Exim's version number from tags in the git
1537 repository when doing development or release builds.
1539 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1540 Bugzilla 879. Patch from Paul Fisher.
1542 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1543 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1546 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1548 PP/05 Variable $av_failed, true if the AV scanner deferred.
1549 Bugzilla 1078. Patch from John Horne.
1551 PP/06 Stop make process more reliably on build failure.
1552 Bugzilla 1087. Patch from Heiko Schlittermann.
1554 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1555 Bugzilla 1089. Patch from Heiko Schlittermann.
1557 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1558 Bugzilla 1131. Patch from Holger Weiß.
1560 PP/09 Handle IPv6 addresses with SPF.
1561 Bugzilla 860. Patch from Wolfgang Breyha.
1563 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1565 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1568 PP/11 match_* no longer expand right-hand-side by default.
1569 New compile-time build option, EXPAND_LISTMATCH_RHS.
1570 New expansion conditions, "inlist", "inlisti".
1572 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1574 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1576 PP/14 fix log_write() format string regression from TF/03.
1577 Bugzilla 1152. Patch from Dmitry Isaikin.
1583 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1585 PP/02 Harmonised TLS library version reporting; only show if debugging.
1586 Layout now matches that introduced for other libraries in 4.74 PP/03.
1588 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1590 PP/04 New "dns_use_edns0" global option.
1592 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1595 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1596 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1598 TK/01 Updated PolarSSL code to 0.14.2.
1599 Bugzilla 1097. Patch from Andreas Metzler.
1601 PP/07 Catch divide-by-zero in ${eval:...}.
1602 Fixes bugzilla 1102.
1604 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1607 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1608 format-string attack -- SECURITY: remote arbitrary code execution.
1610 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1611 time unintentionally subject to list matching rules, letting the header
1612 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1613 arbitrary string expansion). This allowed for information disclosure.
1615 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1616 INT_MIN/-1 -- value coerced to INT_MAX.
1622 NM/01 Workaround for PCRE version dependency in version reporting
1625 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1626 This fixes portability to compilers other than gcc, notably
1627 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1629 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1630 makefiles for portability to HP-UX and POSIX correctness.
1632 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1633 Also via indented variable definition in the Makefile.
1634 (Debugging by Oliver Heesakkers).
1636 PP/02 Restore caching of spamd results with expanded spamd_address.
1637 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1639 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1640 Improves build reliability. Fix from: Frank Elsner
1642 NM/02 Fix wide character breakage in the rfc2047 coding
1643 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1645 NM/03 Allow underscore in dnslist lookups
1646 Fixes bug 1026. Patch from Graeme Fowler
1648 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1649 Code patches from Adam Ciarcinski of NetBSD.
1651 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1654 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1655 is logged, to avoid truncation. Patch from John Horne.
1657 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1658 Patch from Jakob Hirsch.
1660 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1661 SQL string expansion failure details.
1662 Patch from Andrey Oktyabrski.
1664 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1665 Patch from Simon Arlott.
1667 PP/09 New lookups functionality failed to compile on old gcc which rejects
1668 extern declarations in function scope.
1669 Patch from Oliver Fleischmann
1671 PP/10 Use sig_atomic_t for flags set from signal handlers.
1672 Check getgroups() return and improve debugging.
1673 Fixed developed for diagnosis in bug 927 (which turned out to be
1676 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1677 Patch from Mark Zealey.
1679 PP/12 Bugzilla 1056: Improved spamd server selection.
1680 Patch from Mark Zealey.
1682 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1683 Based on patch from Heiko Schlittermann.
1685 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1686 Patch from Uwe Doering, sign-off by Michael Haardt.
1688 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1689 variable declaration deep within a block. Bug and patch from
1692 PP/15 lookups-Makefile IRIX compatibility coercion.
1694 PP/16 Make DISABLE_DKIM build knob functional.
1696 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1697 Patch by Simon Arlott
1699 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1700 variable argument macros. Our copy now differs from upstream.
1706 TF/01 Failure to get a lock on a hints database can have serious
1707 consequences so log it to the panic log.
1709 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1710 controlled using the smtp_confirmation log selector.
1712 TF/03 Include the error message when we fail to unlink a spool file.
1714 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1715 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1716 for maintaining out-of-tree patches for some time.
1718 PP/01 Bugzilla 139: Documentation and portability issues.
1719 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1720 Handle per-OS dynamic-module compilation flags.
1722 PP/02 Let /dev/null have normal permissions.
1723 The 4.73 fixes were a little too stringent and complained about the
1724 permissions on /dev/null. Exempt it from some checks.
1725 Reported by Andreas M. Kirchwitz.
1727 PP/03 Report version information for many libraries, including
1728 Exim version information for dynamically loaded libraries. Created
1729 version.h, now support a version extension string for distributors
1730 who patch heavily. Dynamic module ABI change.
1732 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1733 privilege escalation vulnerability whereby the Exim run-time user
1734 can cause root to append content of the attacker's choosing to
1737 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1740 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1741 If dropping privileges for untrusted macros, we disabled normal logging
1742 on the basis that it would fail; for the Exim run-time user, this is not
1743 the case, and it resulted in successful deliveries going unlogged.
1744 Fixed. Reported by Andreas Metzler.
1750 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1751 only prepend for the Resent-* case. Fixes regression introduced in
1752 Exim 4.70 by NM/22 for Bugzilla 607.
1754 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1755 increasing numbers of administrators be bitten by this.
1757 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1759 PP/03 Bugzilla 994: added openssl_options main configuration option.
1761 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1763 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1765 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1767 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1768 without a peer certificate, leading to a segfault because of an
1769 assumption that peers always have certificates. Be a little more
1770 paranoid. Problem reported by Martin Tscholak.
1772 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1773 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1774 NB: ClamAV planning to remove STREAM in "middle of 2010".
1775 CL also introduces -bmalware, various -d+acl logging additions and
1776 more caution in buffer sizes.
1778 PP/09 Implemented reverse_ip expansion operator.
1780 PP/10 Bugzilla 937: provide a "debug" ACL control.
1782 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1784 PP/12 Bugzilla 973: Implement --version.
1786 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1788 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1790 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1792 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1793 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1794 ignore trailing whitespace.
1796 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1798 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1801 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1802 Notification from Dr Andrew Aitchison.
1804 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1805 ExtendedDetectionInfo response format.
1806 Notification from John Horne.
1808 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1811 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1812 XSL and documented dependency on system catalogs, with examples of how
1815 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1818 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1819 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1820 configuration file which is writeable by the Exim user or group.
1822 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1823 of configuration files to cover files specified with the -C option if
1824 they are going to be used with root privileges, not just the default
1827 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1828 option (effectively making it always true).
1830 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1831 files to be used while preserving root privileges.
1833 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1834 that rogue child processes cannot use them.
1836 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1837 run-time user, instead of root.
1839 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1840 Exim run-time user without dropping privileges.
1842 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1843 result string, instead of calling string_vformat() twice with the same
1846 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1847 for other users. Others should always drop root privileges if they use
1848 -C on the command line, even for a whitelisted configure file.
1850 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1852 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1858 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1859 $data_path, and $header_path variables; fixed documentation bugs and
1862 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1863 exipick to access non-standard spools, including the "frozen" queue
1866 NM/01 Bugzilla 965: Support mysql stored procedures.
1867 Patch from Alain Williams
1869 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1871 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1872 Patch from Andreas Metzler
1874 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1875 Patch from Kirill Miazine
1877 NM/05 Bugzilla 671: Added umask to procmail example.
1879 JJ/03 installed exipick 20100323.0, fixing doc bug
1881 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1882 directory. Notification and patch from Dan Rosenberg.
1884 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1886 TK/02 Improve log output when DKIM signing operation fails.
1888 MH/01 Treat the transport option dkim_domain as a colon separated
1889 list, not as a single string, and sign the message with each element,
1890 omitting multiple occurences of the same signer.
1892 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1893 Bugzilla 985, 986. Patch by Simon Arlott
1895 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1896 Patch by Simon Arlott
1898 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1899 MBX locking. Notification from Dan Rosenberg.
1905 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1907 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1909 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1911 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1913 NM/04 Bugzilla 915: Flag broken perl installation during build.
1919 TK/01 Added patch by Johannes Berg that expands the main option
1920 "spamd_address" if it starts with a dollar sign.
1922 TK/02 Write list of recipients to X-Envelope-Sender header when building
1923 the mbox-format spool file for content scanning (suggested by Jakob
1926 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1927 (http://www.dcc-servers.net/) support via dccifd. Activated by
1928 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1930 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1931 by Mark Daniel Reidel <mr@df.eu>.
1933 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1934 When building exim an external PCRE library is now needed -
1935 PCRE is a system library on the majority of modern systems.
1936 See entry on PCRE_LIBS in EDITME file.
1938 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1939 conversation. Added nologin parameter to request.
1940 Patch contributed by Kirill Miazine.
1942 TF/01 Do not log submission mode rewrites if they do not change the address.
1944 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1946 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1947 log files in place. Contributed by Roberto Lima.
1949 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1951 TF/03 Bugzilla 615: When checking the local_parts router precondition
1952 after a local_part_suffix or local_part_prefix option, Exim now
1953 does not use the address's named list lookup cache, since this
1954 contains cached lookups for the whole local part.
1956 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1957 Robert Millan. Documentation is in experimental-spec.txt.
1959 TF/04 Bugzilla 668: Fix parallel build (make -j).
1961 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1963 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1964 Patch provided by Jan Srzednicki.
1966 TF/05 Leading white space used to be stripped from $spam_report which
1967 wrecked the formatting. Now it is preserved.
1969 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1970 that they are available at delivery time.
1972 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1974 TF/08 TLS error reporting now respects the incoming_interface and
1975 incoming_port log selectors.
1977 TF/09 Produce a more useful error message if an SMTP transport's hosts
1978 setting expands to an empty string.
1980 NM/06 Bugzilla 744: EXPN did not work under TLS.
1981 Patch provided by Phil Pennock.
1983 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1984 Patch provided by Richard Godbee.
1986 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1987 acl_smtp_notquit, added index entry.
1989 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1990 Patch provided by Eugene Bujak.
1992 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1993 accept(). Patch provided by Maxim Dounin.
1995 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1996 Patch provided by Phil Pennock.
1998 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2000 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2001 Patch provided by Brad "anomie" Jorsch.
2003 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2004 Patch provided by Dean Brooks.
2006 TK/05 Add native DKIM support (does not depend on external libraries).
2008 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2009 Patch provided by Graeme Fowler.
2011 NM/16 Bugzilla 851: Documentation example syntax fix.
2013 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2015 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2018 NM/19 Bugzilla 745: TLS version reporting.
2019 Patch provided by Phil Pennock.
2021 NM/20 Bugzilla 167: bool: condition support.
2022 Patch provided by Phil Pennock.
2024 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2025 clients. Patch provided by Phil Pennock.
2027 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2028 Patch provided by Brad "anomie" Jorsch.
2030 NM/23 Bugzilla 687: Fix misparses in eximstats.
2031 Patch provided by Heiko Schlittermann.
2033 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2034 Patch provided by Heiko Schlittermann.
2036 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2037 plus update to original patch.
2039 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2041 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2042 Patch provided by David Brownlee.
2044 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2046 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2048 NM/30 Bugzilla 888: TLS documentation bugfixes.
2050 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2052 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2053 Unlike the original bugzilla I have changed all shell scripts in src tree.
2055 NM/33 Bugzilla 898: Transport filter timeout fix.
2056 Patch by Todd Rinaldo.
2058 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2059 Patch by Serge Demonchaux.
2061 NM/35 Bugzilla 39: Base64 decode bug fixes.
2062 Patch by Jakob Hirsch.
2064 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2066 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2068 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2070 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2076 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2077 ALT-N's libdkim that I have put here:
2078 http://duncanthrax.net/exim-experimental/
2080 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2081 were called 'true' and it seems that C99 defines that as a reserved
2082 keyword to be used with 'bool' variable types. That means you could
2083 not include C99-style headers which use bools without triggering
2084 build errors in sieve.c.
2086 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2087 as mailq or other aliases. Changed the --help handling significantly
2088 to do whats expected. exim_usage() emits usage/help information.
2090 SC/01 Added the -bylocaldomain option to eximstats.
2092 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2094 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2096 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2102 PH/01 Another patch from the Sieve maintainer.
2104 PH/02 When an IPv6 address is converted to a string for single-key lookup
2105 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2106 dots are used instead of colons so that keys in lsearch files need not
2107 contain colons. This was done some time before quoting was made available
2108 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2109 (notated using the quote facility) so as to distinguish them from IPv4
2110 keys. This meant that lookups for IP addresses in host lists did not work
2111 for iplsearch lookups.
2113 This has been fixed by arranging for IPv6 addresses to be expressed with
2114 colons if the lookup type is iplsearch. This is not incompatible, because
2115 previously such lookups could never work.
2117 The situation is now rather anomalous, since one *can* have colons in
2118 ordinary lsearch keys. However, making the change in all cases is
2119 incompatible and would probably break a number of configurations.
2121 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2124 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2125 conversion specification without a maximum field width, thereby enabling
2126 a rogue spamd server to cause a buffer overflow. While nobody in their
2127 right mind would setup Exim to query an untrusted spamd server, an
2128 attacker that gains access to a server running spamd could potentially
2129 exploit this vulnerability to run arbitrary code as the Exim user.
2131 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2132 $primary_hostname instead of what libspf2 thinks the hosts name is.
2134 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2135 a directory entry by the name of the lookup key. Previously, if a
2136 symlink pointed to a non-existing file or a file in a directory that
2137 Exim lacked permissions to read, a lookup for a key matching that
2138 symlink would fail. Now it is enough that a matching directory entry
2139 exists, symlink or not. (Bugzilla 503.)
2141 PH/03 The body_linecount and body_zerocount variables are now exported in the
2144 PH/04 Added the $dnslist_matched variable.
2146 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2147 This means they are set thereafter only if the connection becomes
2150 PH/06 Added the client_condition to authenticators so that some can be skipped
2151 by clients under certain conditions.
2153 PH/07 The error message for a badly-placed control=no_multiline_responses left
2154 "_responses" off the end of the name.
2156 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2158 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2159 (without spaces) instead of just copying the configuration text.
2161 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2163 PH/11 Added $max_received_linelength.
2165 PH/12 Added +ignore_defer and +include_defer to host lists.
2167 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2168 way in which PCRE > 7.0 is built.
2170 PH/14 Implemented queue_only_load_latch.
2172 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2173 MAIL command. The effect was to mangle the value on 64-bit systems.
2175 PH/16 Another patch from the Sieve maintainer.
2177 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2179 PH/18 If a system quota error occurred while trying to create the file for
2180 a maildir delivery, the message "Mailbox is full" was not appended to the
2181 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2182 only to a quota excession during the actual writing of the file.
2184 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2185 characters?) which causes problems in log lines. The DN values are now
2186 passed through string_printing() before being added to log lines.
2188 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2189 and InterBase are left for another time.)
2191 PH/21 Added message_body_newlines option.
2193 PH/22 Guard against possible overflow in moan_check_errorcopy().
2195 PH/23 POSIX allows open() to be a macro; guard against that.
2197 PH/24 If the recipient of an error message contained an @ in the local part
2198 (suitably quoted, of course), incorrect values were put in $domain and
2199 $local_part during the evaluation of errors_copy.
2205 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2206 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2209 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2210 issue a MAIL command.
2212 PH/02 In an ACL statement such as
2214 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2216 if a client was not listed at all, or was listed with a value other than
2217 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2218 the condition was not true (as it should be), so access was not denied.
2219 The bug was that the ! inversion was incorrectly passed on to the second
2220 item. This has been fixed.
2222 PH/03 Added additional dnslists conditions == and =& which are different from
2223 = and & when the dns lookup returns more than one IP address.
2225 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2226 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2228 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2229 FSYNC, which compiles an option called disable_fsync that allows for
2230 bypassing fsync(). The documentation is heavily laced with warnings.
2232 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2234 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2235 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2236 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2237 including adding "make clean"; (3) Added -fPIC when compiling the test
2238 dynamically loaded module, to get rid of a warning.
2240 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2241 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2242 = 0s. The bug is otherwise harmless.
2244 PH/07 There was a bug in the dovecot authenticator such that the value of
2245 $auth1 could be overwritten, and so not correctly preserved, after a
2246 successful authentication. This usually meant that the value preserved by
2247 the server_setid option was incorrect.
2249 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2251 PH/09 Installed PCRE release 7.0.
2253 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2254 run for batched SMTP input. It is now run at the start of every message
2255 in the batch. While fixing this I discovered that the process information
2256 (output by running exiwhat) was not always getting set for -bs and -bS
2257 input. This is fixed, and it now also says "batched" for BSMTP.
2259 PH/11 Added control=no_pipelining.
2261 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2262 patch, slightly modified), and move the expansion of helo_data till after
2263 the connection is made in the smtp transport (so it can use these
2266 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2268 PH/14 Added log_selector = +pid.
2270 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2272 PH/16 Add ${if forany and ${if forall.
2274 PH/17 Added dsn_from option to vary the From: line in DSNs.
2276 PH/18 Flush SMTP output before performing a callout, unless control =
2277 no_callout_flush is set.
2279 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2280 was true (the default) a successful delivery failed to delete the retry
2281 item, thus causing premature timeout of the address. The bug is now
2284 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2286 PH/21 Long custom messages for fakedefer and fakereject are now split up
2287 into multiline responses in the same way that messages for "deny" and
2288 other ACL rejections are.
2290 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2291 with slight modification.
2293 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2294 draft, changing the syntax and factoring some duplicate code".
2296 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2297 for deliveries of the second and subsequent messages over the same SMTP
2300 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2301 ${reduce, with only minor "tidies".
2303 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2305 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2306 expansion side effects.
2308 PH/27 When a message times out after an over-quota error from an Exim-imposed
2309 quota, the bounce message says "mailbox is full". This message was not
2310 being given when it was a system quota that was exceeded. It now should
2313 MH/03 Made $recipients available in local_scan(). local_scan() already has
2314 better access to the recipient list through recipients_list[], but
2315 $recipients can be useful in postmaster-provided expansion strings.
2317 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2318 in the case of a MAIL command with additional options following the
2319 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2320 were accidentally chopped off.
2322 PH/29 SMTP synchronization checks are implemented when a command is read -
2323 there is a check that no more input is waiting when there shouldn't be
2324 any. However, for some commands, a delay in an ACL can mean that it is
2325 some time before the response is written. In this time, more input might
2326 arrive, invalidly. So now there are extra checks after an ACL has run for
2327 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2328 pipelining has not been advertised.
2330 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2332 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2333 returned in the SMTP response when smtp_return_error_details was false.
2334 This has been fixed.
2336 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2337 instead of the C I/O that was originally supplied, because problems were
2338 reported on Solaris.
2340 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2341 Exim which did not show up earlier: it was assuming that a call to
2342 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2343 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2344 was a macro that became an assignment, so it seemed to work. This has
2345 changed to a proper function call with a void return, hence the compile
2346 error. Exim's code has been fixed.
2348 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2351 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2353 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2355 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2356 because it turns out to be too big or there is a timeout) there is no
2357 "Completed" line in the log. When some messages of this type were
2358 selected by exigrep, they were listed as "not completed". Others were
2359 picked up by some special patterns. I have improved the selection
2360 criteria to be more general.
2362 PH/38 The host_find_failed option in the manualroute router can now be set
2363 to "ignore", to completely ignore a host whose IP address cannot be
2364 found. If all hosts are ignored, the behaviour is controlled by the new
2365 host_all_ignored option.
2367 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2368 homing or because of multiple MX records with /mx) generated more than
2369 one IP address, and the following item turned out to be the local host,
2370 all the secondary addresses of the first item were incorrectly removed
2371 from the list, along with the local host and any following hosts (which
2372 is what is supposed to happen).
2374 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2375 whoever called Exim into the -H file. In the case of the daemon it was
2376 behaving confusingly. When first started, it used values for whoever
2377 started the daemon, but after a SIGHUP it used the Exim user (because it
2378 calls itself on a restart). I have changed the code so that it now always
2381 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2382 message are rejected with the same error (e.g. no authentication or bad
2383 sender address), and a DATA command is nevertheless sent (as can happen
2384 with PIPELINING or a stupid MUA), the error message that was given to the
2385 RCPT commands is included in the rejection of the DATA command. This is
2386 intended to be helpful for MUAs that show only the final error to their
2389 PH/42 Another patch from the Sieve maintainer.
2391 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2392 Eximstats - Fixed some broken HTML links and added missing column headers
2394 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2395 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2397 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2399 PH/43 Yet another patch from the Sieve maintainer.
2401 PH/44 I found a way to check for a TCP/IP connection going away before sending
2402 the response to the final '.' that terminates a message, but only in the
2403 case where the client has not sent further data following the '.'
2404 (unfortunately, this is allowed). However, in many cases there won't be
2405 any further data because there won't be any more messages to send. A call
2406 to select() can be used: if it shows that the input is "ready", there is
2407 either input waiting, or the socket has been closed. An attempt to read
2408 the next input character can distinguish the two cases. Previously, Exim
2409 would have sent an OK response which the client would never have see.
2410 This could lead to message repetition. This fix should cure that, at
2411 least in a lot of common cases.
2413 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2414 advertised in response to EHLO.
2420 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2421 fixed by 4.65/MH/01 (is this a record?) are fixed:
2423 (i) An empty string was always treated as zero by the numeric comparison
2424 operators. This behaviour has been restored.
2426 (ii) It is documented that the numeric comparison operators always treat
2427 their arguments as decimal numbers. This was broken in that numbers
2428 starting with 0 were being interpreted as octal.
2430 While fixing these problems I realized that there was another issue that
2431 hadn't been noticed. Values of message_size_limit (both the global option
2432 and the transport option) were treated as octal if they started with 0.
2433 The documentation was vague. These values are now always treated as
2434 decimal, and I will make that clear in the documentation.
2440 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2441 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2444 MH/01 Don't check that the operands of numeric comparison operators are
2445 integers when their expansion is in "skipping" mode (fixes bug
2446 introduced by 4.64-PH/07).
2448 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2449 child addresses, Exim now panics and dies. Previously, because the count
2450 is held in a short int, deliveries were likely to be lost. As such a
2451 large number of recipients for a single message is ridiculous
2452 (performance will be very, very poor), I have chosen to impose a limit
2453 rather than extend the field.
2459 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2460 leftover -K file (the existence of which was triggered by #402).
2461 While we were at it, introduced process PID as part of the -K
2462 filename. This should rule out race conditions when creating
2465 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2466 processing considerably. Previous code took too long for large mails,
2467 triggering a timeout which in turn triggers #401.
2469 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2470 in the DK code in transports.c. sendfile() is not really portable,
2471 hence the _LINUX specificness.
2473 TF/01 In the add_headers option to the mail command in an Exim filter,
2474 there was a bug that Exim would claim a syntax error in any
2475 header after the first one which had an odd number of characters
2478 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2479 callout verification, Exim cached a "reject" for the entire domain. This
2480 is correct for most verifications, but it is not correct for a recipient
2481 verification with use_sender or use_postmaster set, because in that case
2482 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2483 case of MAIL FROM:<> rejection from other early rejections (e.g.
2484 rejection of HELO). When verifying a recipient using a non-null MAIL
2485 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2486 Whatever the result of the callout, the value of the domain cache is
2487 left unchanged (for any other kind of callout, getting as far as trying
2488 RCPT means that the domain itself is ok).
2490 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2493 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2494 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2495 session, the daemon ignores ECONNECT errors and logs others; it now
2496 ignores EPIPE as well.
2498 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2499 (quoted-printable decoding).
2501 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2502 later the small subsequent patch to fix an introduced bug.
2504 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2506 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2508 PH/08 An error is now given if message_size_limit is specified negative.
2510 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2511 to be given (somewhat) arbitrary names.
2513 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2516 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2517 miscellaneous code fixes
2519 PH/10 Added the log_reject_target ACL modifier to specify where to log
2522 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2523 hostname. This is wrong, because it relates to the incoming message (and
2524 probably the interface on which it is arriving) and not to the outgoing
2525 callout (which could be using a different interface). This has been
2526 changed to use the value of the helo_data option from the smtp transport
2527 instead - this is what is used when a message is actually being sent. If
2528 there is no remote transport (possible with a router that sets up host
2529 addresses), $smtp_active_hostname is used.
2531 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2532 tweaks were necessary in order to get it to work (see also 21 below):
2533 (a) The code assumed that strncpy() returns a negative number on buffer
2534 overflow, which isn't the case. Replaced with Exim's string_format()
2536 (b) There were several signed/unsigned issues. I just did the minimum
2537 hacking in of casts. There is scope for a larger refactoring.
2538 (c) The code used strcasecmp() which is not a standard C function.
2539 Replaced with Exim's strcmpic() function.
2540 (d) The code set only $1; it now sets $auth1 as well.
2541 (e) A simple test gave the error "authentication client didn't specify
2542 service in request". It would seem that Dovecot has changed its
2543 interface. Fortunately there's a specification; I followed it and
2544 changed what the client sends and it appears to be working now.
2546 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2549 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2550 address A is aliased to B and C, where B exists and C does not. Without
2551 -v the output is "A verified" because verification stops after a
2552 successful redirection if more than one address is generated. However,
2553 with -v the child addresses are also verified. Exim was outputting "A
2554 failed to verify" and then showing the successful verification for C,
2555 with its parentage. It now outputs "B failed to verify", showing B's
2556 parentage before showing the successful verification of C.
2558 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2559 look up a TXT record in a specific list after matching in a combined
2562 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2563 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2564 they consult the DNS. I had assumed they would set it the way they
2565 wanted; and indeed my experiments on Linux seem to show that in some
2566 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2567 To be on the safe side, however, I have now made the interface to
2568 host_find_byname() similar to host_find_bydns(), with an argument
2569 containing the DNS resolver options. The host_find_byname() function now
2570 sets these options at its start, just as host_find_bydns() does. The smtp
2571 transport options dns_qualify_single and dns_search_parents are passed to
2572 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2573 of host_find_byname() use the default settings of RES_DEFNAMES
2574 (qualify_single) but not RES_DNSRCH (search_parents).
2576 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2577 spool_read_header() do less string testing, by means of a preliminary
2578 switch on the second character of optional "-foo" lines. (This is
2579 overdue, caused by the large number of possibilities that now exist.
2580 Originally there were few.) While I was there, I also converted the
2581 str(n)cmp tests so they don't re-test the leading "-" and the first
2582 character, in the hope this might squeeze out yet more improvement.
2584 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2585 flag allowing group syntax was set by the header_syntax check but not
2586 turned off, possible causing trouble later; (2) The flag was not being
2587 set at all for the header_verify test, causing "group"-style headers to
2588 be rejected. I have now set it in this case, and also caused header_
2589 verify to ignore an empty address taken from a group. While doing this, I
2590 came across some other cases where the code for allowing group syntax
2591 while scanning a header line wasn't quite right (mostly, not resetting
2592 the flag correctly in the right place). These bugs could have caused
2593 trouble for malformed header lines. I hope it is now all correct.
2595 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2596 with the "reply" argument non-NULL. The code, however (which originally
2597 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2598 but it didn't always do it. This confused somebody who was copying the
2599 code for some other use. I have removed all the tests.
2601 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2602 feature that was used to support insecure browsers during the U.S. crypto
2603 embargo. It requires special client support, and Exim is probably the
2604 only MTA that supported it -- and would never use it because real RSA is
2605 always available. This code has been removed, because it had the bad
2606 effect of slowing Exim down by computing (never used) parameters for the
2607 RSA_EXPORT functionality.
2609 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2610 authenticator to fail if there's a tab character in the incoming data
2611 (there should never be unless someone is messing about, as it's supposed
2612 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2613 if the connection is using TLS or if the remote IP is the same as the
2614 local IP, and the "valid-client-cert option" if a client certificate has
2617 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2618 authenticators. This can be used for authorization after authentication
2619 succeeds. (In the case of plaintext, it servers for both authentication
2622 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2623 if any retry times were supplied.
2625 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2626 connection, where there is no client IP address to check. In this
2627 situation, the verify now always succeeds.
2629 PH/25 Applied John Jetmore's -Mset patch.
2631 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2633 PH/27 In a string expansion for a processed (not raw) header when multiple
2634 headers of the same name were present, leading whitespace was being
2635 removed from all of them, but trailing whitespace was being removed only
2636 from the last one. Now trailing whitespace is removed from each header
2637 before concatenation. Completely empty headers in a concatenation (as
2638 before) are ignored.
2640 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2641 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2643 PH/29 [Removed. This was a change that I later backed out, and forgot to
2644 correct the ChangeLog entry (that I had efficiently created) before
2645 committing the later change.]
2647 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2648 address errors (4xx response to RCPT) over the same connection as other
2649 messages routed to the same hosts. Such deliveries are always "forced",
2650 so retry times are not inspected. This resulted in far too many retries
2651 for the affected addresses. The effect occurred only when there were more
2652 hosts than the hosts_max_try setting in the smtp transport when it had
2653 the 4xx errors. Those hosts that it had tried were not added to the list
2654 of hosts for which the message was waiting, so if all were tried, there
2655 was no problem. Two fixes have been applied:
2657 (i) If there are any address or message errors in an SMTP delivery, none
2658 of the hosts (tried or untried) are now added to the list of hosts
2659 for which the message is waiting, so the message should not be a
2660 candidate for sending over the same connection that was used for a
2661 successful delivery of some other message. This seems entirely
2662 reasonable: after all the message is NOT "waiting for some host".
2663 This is so "obvious" that I'm not sure why it wasn't done
2664 previously. Hope I haven't missed anything, but it can't do any
2665 harm, as the worst effect is to miss an optimization.
2667 (ii) If, despite (i), such a delivery is accidentally attempted, the
2668 routing retry time is respected, so at least it doesn't keep
2669 hammering the server.
2671 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2672 in ${readsocket because some servers need this prod.
2674 PH/32 Added some extra debug output when updating a wait-xxx database.
2676 PH/33 The hint "could be header name not terminated by colon", which has been
2677 given for certain expansion errors for a long time, was not being given
2678 for the ${if def:h_colon_omitted{... case.
2680 PH/34 The spec says: "With one important exception, whenever a domain list is
2681 being scanned, $domain contains the subject domain." There was at least
2682 one case where this was not true.
2684 PH/35 The error "getsockname() failed: connection reset by peer" was being
2685 written to the panic log as well as the main log, but it isn't really
2686 panic-worthy as it just means the connection died rather early on. I have
2687 removed the panic log writing for the ECONNRESET error when getsockname()
2690 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2691 runs only) independently of the message's sender address. This meant
2692 that, if the 4xx error was in fact related to the sender, a different
2693 message to the same recipient with a different sender could confuse
2694 things. In particular, this can happen when sending to a greylisting
2695 server, but other circumstances could also provoke similar problems.
2696 I have changed the default so that the retry time for these errors is now
2697 based a combination of the sender and recipient addresses. This change
2698 can be overridden by setting address_retry_include_sender=false in the
2701 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2702 remote server are returned as part of bounce messages. This was not
2703 happening for LMTP over a pipe (the lmtp transport), but now it is the
2704 same for both kinds of LMTP.
2706 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2707 in header lines that were in fact CNAMEs. This is no longer the case.
2709 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2710 and queue runs started by the daemon processed all messages. This has
2711 been fixed so that -R and -S can now usefully be given with -q<time>.
2713 PH/40 Import PCRE release 6.7 (fixes some bugs).
2715 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2717 PH/42 Give an error if -q is specified more than once.
2719 PH/43 Renamed the variables $interface_address and $interface_port as
2720 $received_ip_address and $received_port, to make it clear that these
2721 values apply to message reception, and not to the outgoing interface when
2722 a message is delivered. (The old names remain recognized, of course.)
2724 PH/44 There was no timeout on the connect() call when using a Unix domain
2725 socket in the ${readsocket expansion. There now is.
2727 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2728 be meaningful with "accept".
2730 SC/01 Eximstats V1.43
2731 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2733 SC/02 Eximstats V1.44
2734 Use a glob alias rather than an array ref in the generated
2735 parser. This improves both readability and performance.
2737 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2738 Collect SpamAssassin and rejection statistics.
2739 Don't display local sender or destination tables unless
2740 there is data to show.
2741 Added average volumes into the top table text output.
2743 SC/04 Eximstats V1.46
2744 Collect data on the number of addresses (recipients)
2745 as well as the number of messages.
2747 SC/05 Eximstats V1.47
2748 Added 'Message too big' to the list of mail rejection
2749 reasons (thanks to Marco Gaiarin).
2751 SC/06 Eximstats V1.48
2752 Mainlog lines which have GMT offsets and are too short to
2753 have a flag are now skipped.
2755 SC/07 Eximstats V1.49 (Alain Williams)
2756 Added the -emptyok flag.
2758 SC/08 Eximstats V1.50
2759 Fixes for obtaining the IP address from reject messages.
2761 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2762 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2763 whitespace changes from 4.64-PH/27
2765 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2768 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2769 are found, allow negative numbers in numeric criteria)
2771 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2773 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2774 to match changes made in 4.64-PH/43
2776 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2778 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2779 "domain" is actually an IP address, and force a failure. This locks out
2780 those revolvers/nameservers that support "A-for-A" lookups, in
2781 contravention of the specifications.
2783 PH/48 When a host name was looked up from an IP address, and the subsequent
2784 forward lookup of the name timed out, the host name was left in
2785 $sender_host_name, contrary to the specification.
2787 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2788 restricted to single-key lookups, Exim was not diagnosing an error if
2789 * or *@ was used with a query-style lookup.
2791 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2793 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2794 long ago, but noone interested enough thought of it. Let's just say that
2795 the "1.1" means that there are some new functions that weren't there at
2796 some point in the past.
2798 PH/51 Error processing for expansion failure of helo_data from an smtp
2799 transport during callout processing was broken.
2801 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2802 tested/used via the -bh/-bhc/-bs options.
2804 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2805 bug, fixed in subsequent PCRE releases).
2807 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2808 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2810 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2816 SC/01 Use a glob alias rather than an array ref in eximstats generated
2817 parser. This improves both readability and performance.
2819 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2820 Don't display local sender or destination tables in eximstats unless
2821 there is data to show.
2822 Added average volumes into the eximstats top table text output.
2824 SC/03 Collect data on the number of addresses (recipients) as well
2825 as the number of messages in eximstats.
2827 TF/01 Correct an error in the documentation for the redirect router. Exim
2828 does (usually) call initgroups() when daemonizing.
2830 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2831 with consistent privilege compared to when running as a daemon.
2833 TF/03 Note in the spec that $authenticated_id is not set for local
2834 submissions from trusted users.
2836 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2837 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2839 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2840 by adding some example configuration directives to the default
2841 configuration file. A little bit of work is required to uncomment the
2842 directives and define how usernames and passwords are checked, but
2843 there is now a framework to start from.
2845 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2846 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2847 without this. I don't know how relevant this is to other LDAP libraries.
2849 PH/02 Add the verb name to the "unknown ACL verb" error.
2851 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2853 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2855 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2856 directory not expanded when it should be if an expanded home directory
2857 was set for the address (which is overridden by the transport).
2859 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2862 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2863 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2864 because it is too late at that time, and has no effect.
2866 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2867 security issue with \' (bugzilla #107). I could not use the
2868 PQescapeStringConn() function, because it needs a PGconn value as one of
2871 PH/08 When testing addresses using -bt, indicate those final addresses that
2872 are duplicates that would not cause an additional delivery. At least one
2873 person was confused, thinking that -bt output corresponded to deliveries.
2874 (Suppressing duplicates isn't a good idea as you lose the information
2875 about possibly different redirections that led to the duplicates.)
2877 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2878 systems where poll() doesn't work, in particular OS X.
2880 PH/10 Added more information to debugging output for retry time not reached.
2882 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2883 operations in malware.c.
2885 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2888 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2889 syslog_duplication set false, log lines that would normally be written
2890 both the the main log and to the reject log were not written to syslog at
2893 PH/14 In the default configuration, change the use of "message" in ACL warn
2894 statements to "add_header".
2896 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2897 not followed by a command (e.g. "seen endif").
2899 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2900 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2903 PH/17 Added extra conditions to the default value of delay_warning_condition
2907 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2908 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2909 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2912 The Auto-Submitted: and various List- headers are standardised, whereas I
2913 don't think Precedence: ever was.
2915 PH/18 Refactored debugging code in route_finduser() to show more information,
2916 in particular, the error code if getpwnam() issues one.
2918 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2919 This is apparently needed in addition to the PH/07 change above to avoid
2920 any possible encoding problems.
2922 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2923 but not after initializing Perl.
2925 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2926 output them only if debugging. By default they are written stderr,
2927 apparently, which is not desirable.
2929 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2932 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2935 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2937 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2938 authenticated or an ident call has been made. Suppress the default
2939 values for $authenticated_id and $authenticated_sender (but permit -oMai
2940 and -oMas) when testing with -bh.
2942 PH/24 Re-jigged the order of the tests in the default configuration so that the
2943 tests for valid domains and recipients precede the DNS black list and CSA
2944 tests, on the grounds that those ones are more expensive.
2946 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2947 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2948 This bug exists in every version of Exim that I still have, right back to
2951 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2952 However, an attempt to turn on case-sensitivity in a regex key by
2953 including (?-i) didn't work because the subject string was already
2954 lowercased, and the effects were non-intuitive. It turns out that a
2955 one-line patch can be used to allow (?-i) to work as expected.
2961 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2962 other effects) broke the use of negated acl sub-conditions.
2964 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2967 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2968 "Deny" causes Exim to reject the incoming connection with a 554 error.
2969 Unfortunately, if there is a major crisis, such as a disk failure,
2970 tcp-wrappers gives "deny", whereas what one would like would be some
2971 kind of temporary error. A kludge has been added to help with this.
2972 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2973 554 error is used if errno is still zero or contains ENOENT (which occurs
2974 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2977 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2979 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2980 errors. Otherwise a message that provokes a temporary error (when other
2981 messages do not) can cause a whole host to time out.
2983 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2984 addresses were routed directly to files or pipes from a redirect router.
2985 File deliveries just didn't batch; pipe deliveries might have suffered
2988 PH/06 A failure to get a lock for a hints database would erroneously always say
2989 "Failed to get write lock", even when it was really a read lock.
2991 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2992 of 0600. This has been changed to use the value of the lockfile_mode
2993 option (which defaults to 0600).
2995 PH/08 Applied small patch from the Sieve maintainer.
2997 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2998 folder from quota calculations, a direct delivery into this folder messed
2999 up the contents of the maildirsize file. This was because the regex was
3000 used only to exclude .Trash (or whatever) when the size of the mailbox
3001 was calculated. There was no check that a delivery was happening into an
3002 excluded directory. This bug has been fixed by ignoring all quota
3003 processing for deliveries into excluded directories.
3005 PH/10 Added the maildirfolder_create_regex option to appendfile.
3011 PH/01 The code for finding all the local interface addresses on a FreeBSD
3012 system running IPv6 was broken. This may well have applied to all BSD
3013 systems, as well as to others that have similar system calls. The broken
3014 code found IPv4 interfaces correctly, but gave incorrect values for the
3015 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3016 that it would not match correctly against @[] and not recognize the IPv6
3019 PH/02 The ipliteral router was not recognizing addresses of the form user@
3020 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3022 PH/03 Added disable_ipv6.
3024 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3025 decoded form, because it is most often used to construct To: headers
3026 lines in autoreplies, and the decoded form may well be syntactically
3027 invalid. However, $reply_address has leading white space removed, and all
3028 newlines turned into spaces so that the autoreply transport does not
3031 PH/05 If group was specified without a user on a router, and no group or user
3032 was specified on a transport, the group from the router was ignored.
3034 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3035 for visible compile-time settings that can be used to change these
3036 numbers, for those that want even more. Backwards compatibility with old
3037 spool files has been maintained. However, going back to a previous Exim
3038 release will lost any variables that are in spool files.
3040 PH/07 Two small changes when running in the test harness: increase delay when
3041 passing a TCP/IP connection to a new process, in case the original
3042 process has to generate a bounce, and remove special handling of
3043 127.0.0.2 (sic), which is no longer necessary.
3045 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3046 be the same on different OS.
3048 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3051 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3052 whether --show-vars was specified or not
3054 JJ/02 exipick: Added support for new ACL variable spool format introduced
3057 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3058 syntactically invalid From: or Reply-to: line, and a filter used this to
3059 generate an autoreply, and therefore failed to obtain an address for the
3060 autoreply, Exim could try to deliver to a non-existent relative file
3061 name, causing unrelated and misleading errors. What now happens is that
3062 it logs this as a hard delivery error, but does not attempt to create a
3065 PH/11 The exinext utility has a -C option for testing purposes, but although
3066 the given file was scanned by exinext itself; it wasn't being passed on
3067 when Exim was called.
3069 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3070 an end-of-file indication when reading a command response.
3072 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3073 compiled. In many other places in Exim, IPv6 addresses are always
3074 recognized, so I have changed this. It also means that IPv4 domain
3075 literals of the form [IPV4:n.n.n.n] are now always recognized.
3077 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3078 used if the router is not running as root, for example, when verifying at
3079 ACL time, or when using -bh. The debugging output from this situation was
3080 non-existent - all you got was a failure to exec. I have made two
3083 (a) Failures to set uid/gid, the current directory, or a process leader
3084 in a subprocess such as that created by queryprogram now generate
3085 suitable debugging output when -d is set.
3087 (b) The queryprogram router detects when it is not running as root,
3088 outputs suitable debugging information if -d is set, and then runs
3089 the subprocess without attempting to change uid/gid.
3091 PH/15 Minor change to Makefile for building test_host (undocumented testing
3094 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3095 additional section of a DNS packet that returns MX or SRV records.
3096 Instead, it always explicitly searches for A/AAAA records. This avoids
3097 major problems that occur when a DNS server includes only records of one
3098 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3099 fixed another bug: if SRV records were looked up and the corresponding
3100 address records were *not* found in the additional section, the port
3101 values from the SRV records were lost.
3103 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3104 using the correct key (the original address) when searching the retry
3105 rules in order to find which one to use for generating the retry hint.
3107 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3108 adding the default one. Similarly, if it contains a Reply-To: header, the
3109 errors_reply_to option, if set, is not used.
3111 PH/19 When calculating a retry time, Exim used to measure the "time since
3112 failure" by looking at the "first failed" field in the retry record. Now
3113 it does not use this if it is later than than the arrival time of the
3114 message. Instead it uses the arrival time. This makes for better
3115 behaviour in cases where some deliveries succeed, thus re-setting the
3116 "first failed" field. An example is a quota failure for a huge message
3117 when small messages continue to be delivered. Without this change, the
3118 "time since failure" will always be short, possible causing more frequent
3119 delivery attempts for the huge message than are intended.
3120 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3122 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3123 $1, $2, $3) because the numerical variables can be reset during some
3124 expansion items (e.g. "match"), thereby losing the authentication data.
3126 PH/21 Make -bV show the size of off_t variables so that the test suite can
3127 decide whether to run tests for quotas > 2G.
3129 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3130 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3131 filecount value is greater than 2G or if a quota value is greater than 2G
3132 on a system where the size of off_t is not greater than 4, a panic error
3135 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3136 never match. The debug and -bh output now contains an explicit error
3137 message indicating a malformed IPv4 address or mask.
3139 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3140 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3141 PH/23 above applies.
3143 PH/25 Do not write to syslog when running in the test harness. The only
3144 occasion when this arises is a failure to open the main or panic logs
3145 (for which there is an explicit test).
3147 PH/26 Added the /no_tell option to "control=freeze".
3149 PH/27 If a host name lookup failed very early in a connection, for example, if
3150 the IP address matched host_lookup and the reverse lookup yielded a name
3151 that did not have a forward lookup, an error message of the form "no IP
3152 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3153 could be logged. Now it outputs the IP address instead of "NULL".
3155 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3156 allows the sender and the authenticated sender to be set when
3157 submitting a message from within Exim. Since child_open_exim() is
3158 documented for local_scan(), the new function should be too.
3160 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3161 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3162 results in an empty string is now treated as unset.
3164 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3166 PH/31 Added sender_verify_fail logging option.
3168 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3169 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3170 tidied the source and removed it altogether.
3172 PH/33 When a queue run was abandoned because the load average was too high, a
3173 log line was always written; now it is written only if the queue_run log
3174 selector is set. In addition, the log line for abandonment now contains
3175 information about the queue run such as the pid. This is always present
3176 in "start" and "stop" lines but was omitted from the "abandon" line.
3178 PH/34 Omit spaces between a header name and the colon in the error message that
3179 is given when verify = headers_syntax fails (if there are lots of them,
3180 the message gets confusing).
3182 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3183 names, as there are now some PTR records that contain slashes. This check
3184 is only to protect against broken name servers that fall over on strange
3185 characters, so the fact that it applies to all lookups doesn't matter.
3187 PH/36 Now that the new test suite is complete, we can remove some of the
3188 special code in Exim that was needed for the old test suite. For example,
3189 sorting DNS records because real resolvers return them in an arbitrary
3190 order. The new test suite's fake resolver always returns records in the
3193 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3194 bounces) except when queue_only is set, to avoid logging races between
3195 the different processes.
3197 PH/38 Panic-die if .include specifies a non-absolute path.
3199 PH/39 A tweak to the "H" retry rule from its user.
3201 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3202 a label. They prevented compilation on older perls.
3204 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3205 a warning to be raised on newish perls.
3207 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3208 on queue. Changes to match documented behaviour of showing count of
3209 messages matching specified criteria.
3211 PH/40 Changed the default ident timeout from 30s to 5s.
3213 PH/41 Added support for the use of login_cap features, on those BSD systems
3214 that have them, for controlling the resources used by pipe deliveries.
3216 PH/42 The content-scanning code uses fopen() to create files in which to put
3217 message data. Previously it was not paying any attention to the mode of
3218 the files. Exim runs with umask(0) because the rest of the code creates
3219 files with open(), and sets the required mode explicitly. Thus, these
3220 files were ending up world-writeable. This was not a big issue, because,
3221 being within the spool directory, they were not world-accessible. I have
3222 created a function called modefopen, which takes an additional mode
3223 argument. It sets umask(777), creates the file, chmods it to the required
3224 mode, then resets the umask. All the relevant calls to fopen() in the
3225 content scanning code have been changed to use this function.
3227 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3228 to 24 hours. This avoids potential overflow problems when processing G
3229 and H retry rules. I suspect nobody ever tinkers with this value.
3231 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3233 PH/45 When the plaintext authenticator is running as a client, the server's
3234 challenges are checked to ensure they are valid base64 strings. By
3235 default, the authentication attempt is cancelled if an invalid string is
3236 received. Setting client_ignore_invalid_base64 true ignores these errors.
3237 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3238 they are received. Thus, the responses can be made to depend on the
3239 challenges. If an invalid string is ignored, an empty string is placed in
3242 PH/46 Messages that are created by the autoreply transport now contains a
3243 References: header, in accordance with RFCs 2822 and 3834.
3245 PH/47 Added authenticated_sender_force to the smtp transport.
3247 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3249 PH/49 Installed latest patch from the Sieve maintainer.
3251 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3252 was also set, the appendfile transport was unnecessarily scanning a
3253 directory of message files (e.g. for maildir delivery) to find the count
3254 of files (along with the size), even though it did not need this
3255 information. It now does the scan only if it needs to find either the
3256 size of the count of files.
3258 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3260 PH/52 Two bugs concerned with error handling when the smtp transport is
3263 (i) Exim was not creating retry information for temporary errors given
3264 for individual recipients after the DATA command when the smtp transport
3265 was used in LMTP mode. This meant that they could be retried too
3266 frequently, and not timed out correctly.
3268 (ii) Exim was setting the flag that allows error details to be returned
3269 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3270 recipients that were returned after the DATA command.
3272 PH/53 This is related to PH/52, but is more general: for any failing address,
3273 when detailed error information was permitted to be returned to the
3274 sender, but the error was temporary, then after the final timeout, only
3275 "retry timeout exceeded" was returned. Now it returns the full error as
3276 well as "retry timeout exceeded".
3278 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3279 do this, and (what is worse) MTAs that accept it.
3281 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3282 will now be deprecated.
3284 PH/56 New os.c-cygwin from the Cygwin maintainer.
3286 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3287 formats (previously only available in exim formats via -bpr, -bpru,
3288 and -bpra. Now also available in native and exiqgrep formats)
3290 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3291 with very large, slow to parse queues
3293 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3295 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3297 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3298 responses to authentication challenges, though it was showing the
3299 challenges; (ii) I've removed the CR characters from the debug output for
3302 PH/58 Allow for the insertion of a newline as well as a space when a string
3303 is turned into more than one encoded-word during RFC 2047 encoding. The
3304 Sieve code now uses this.
3306 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3307 data_4xx, lost_connection, tls_required.
3309 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3310 message was being sent as an SMTP response.
3312 PH/61 Add -l and -k options to exicyclog.
3314 PH/62 When verifying, if an address was redirected to one new address, so that
3315 verification continued, and the new address failed or deferred after
3316 having set something in $address_data, the value of $address_data was not
3317 passed back to the ACL. This was different to the case when no
3318 redirection occurred. The value is now passed back in both cases.
3320 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3321 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3322 use login_cap.h, so on its own it isn't the distinguishing feature. The
3323 new name refers directly to the setclassresources() function.
3325 PH/65 Added configuration files for NetBSD3.
3327 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3329 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3330 is preferred over IPv4.
3332 PH/68 The bounce_return_message and bounce_return_body options were not being
3333 honoured for bounces generated during the reception of non-SMTP messages.
3334 In particular, this applied to messages rejected by the ACL. This bug has
3335 been fixed. However, if bounce_return_message is true and bounce_return_
3336 body is false, the headers that are returned for a non-SMTP message
3337 include only those that have been read before the error was detected.
3338 (In the case of an ACL rejection, they have all been read.)
3340 PH/69 The HTML version of the specification is now built in a directory called
3341 spec_html instead of spec.html, because the latter looks like a path with
3342 a MIME-type, and this confuses some software.
3344 PH/70 Catch two compiler warnings in sieve.c.
3346 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3347 function verify_get_ident() calls ip_connect() to connect a socket, but
3348 if the "connect()" function timed out, ip_connect() used to close the
3349 socket. However, verify_get_ident() also closes the socket later, and in
3350 between Exim writes to the log, which may get opened at this point. When
3351 the socket was closed in ip_connect(), the log could get the same file
3352 descriptor number as the socket. This naturally causes chaos. The fix is
3353 not to close the socket in ip_connect(); the socket should be closed by
3354 the function that creates it. There was only one place in the code where
3355 this was missing, in the iplookup router, which I don't think anybody now
3356 uses, but I've fixed it anyway.
3358 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3359 well as to direct DNS lookups. Otherwise the handling of names in host
3360 lists is inconsistent and therefore confusing.
3366 PH/01 Two changes to the default runtime configuration:
3368 (1) Move the checks for relay_from_hosts and authenticated clients from
3369 after to before the (commented out) DNS black list checks.
3371 (2) Add control=submission to the relay_from_hosts and authenticated
3372 clients checks, on the grounds that messages accepted by these
3373 statements are most likely to be submissions.
3375 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3377 (1) Generate an error if the third argument for the ${prvs expansion is
3380 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3383 (3) Reset the variables that are obtained from the first argument of
3384 ${prvscheck and used in the second argument before leaving the code,
3385 because their memory is reclaimed, so using them afterwards may do
3388 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3389 one (it's much easier than Tom thought :-).
3391 (5) Because of (4), we can now allow for the use of $prvscheck_result
3392 inside the third argument.
3394 PH/03 For some reason, the default setting of PATH when running a command from
3395 a pipe transport was just "/usr/bin". I have changed it to
3398 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3399 anything to be listed in the output from -bV.
3401 PH/05 When a filter generated an autoreply, the entire To: header line was
3402 quoted in the delivery log line, like this:
3404 => >A.N.Other <ano@some.domain> <original@ddress> ...
3406 This has been changed so that it extracts the operative address. There
3407 may be more than one such address. If so, they are comma-separated, like
3410 => >ano@some.domain,ona@other.domain <original@ddress> ...
3412 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3413 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3414 not being looked up in the rDNS to get a host name, Exim was showing the
3415 IP address twice in Received: lines, even though the IP addresses were
3416 identical. For example:
3418 Received: from [1.2.3.4] (helo=[1.2.3.4])
3420 However, if the real host name was known, it was omitting the HELO data
3421 if it matched the actual IP address. This has been tidied up so that it
3422 doesn't show the same IP address twice.
3424 PH/07 When both +timestamp and +memory debugging was on, the value given by
3425 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3426 called by the debug printing, thereby overwriting the timestamp buffer.
3427 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3429 PH/08 When the original message was included in an autoreply transport, it
3430 always said "this is a copy of the message, including all the headers",
3431 even if body_only or headers_only was set. It now gives an appropriate
3434 PH/09 Applied a patch from the Sieve maintainer which:
3436 o fixes some comments
3437 o adds the (disabled) notify extension core
3438 o adds some debug output for the result of if/elsif tests
3439 o points to the current vacation draft in the documentation
3440 and documents the missing references header update
3444 o fixes a bug in processing the envelope test (when testing
3445 multiple envelope elements, the last element determined the
3448 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3449 Electronic Mail") by including:
3451 Auto-submitted: auto-generated
3453 in the messages that it generates (bounce messages and others, such as
3454 warnings). In the case of bounce messages for non-SMTP messages, there was
3455 also a typo: it was using "Auto_submitted" (underscore instead of
3456 hyphen). Since every message generated by Exim is necessarily in response
3457 to another message, thes have all been changed to:
3459 Auto-Submitted: auto-replied
3461 in accordance with these statements in the RFC:
3463 The auto-replied keyword:
3465 - SHOULD be used on messages sent in direct response to another
3466 message by an automatic process,
3468 - MUST NOT be used on manually-generated messages,
3470 - MAY be used on Delivery Status Notifications (DSNs) and Message
3471 Disposition Notifications (MDNs),
3473 - MUST NOT be used on messages generated by automatic or periodic
3474 processes, except for messages which are automatic responses to
3477 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3478 to the default Received: header definition.
3480 PH/12 Added log selector acl_warn_skipped (default on).
3482 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3483 variables because (a) they are in the wrong storage pool and (b) even if
3484 they were copied, it wouldn't work properly because of the caching.
3486 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3487 checking when decoding. Apparently there are clients that generate
3488 overlong encoded strings. Why am I not surprised?
3490 PH/15 If the first argument of "${if match_address" was not empty, but did not
3491 contain an "@" character, Exim crashed. Now it writes a panic log message
3492 and treats the condition as false.
3494 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3496 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3497 extension "envelope-auth". The code is finished and in agreement with
3498 other implementations, but there is no documentation so far and in fact,
3499 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3500 not changing the active code.
3502 Print executed "if" and "elsif" statements when debugging is used. This
3503 helps a great deal to understand what a filter does.
3505 Document more things not specified clearly in RFC3028. I had all this
3506 sorted out, when out of a sudden new issues came to my mind. Oops."
3508 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3511 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3512 canonical form (as documented). However, after a host name lookup from
3513 the IP address, check_host() was doing a simple string comparison with
3514 addresses acquired from the DNS when checking that the found name did
3515 have the original IP as one of its addresses. Since any found IPv6
3516 addresses are likely to be in abbreviated form, the comparison could
3517 fail. Luckily, there already exists a function for doing the comparison
3518 by converting both addresses to binary, so now that is used instead of
3519 the text comparison.
3521 PH/20 There was another similar case to PH/19, when a complete host name was
3522 given in a host list; looking up its IP address could give an abbreviated
3523 form, whereas the current host's name might or might not be abbreviated.
3524 The same fix has been applied.
3530 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3531 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3534 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3535 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3537 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3539 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3540 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3541 Also, for "Auto-Submitted", treat anything other than "no" as
3542 non-personal, in accordance with RFC 3834. (Previously it treated
3543 anything starting "auto-" as non-personal.)
3545 TF/01 The control=submission/name=... option had a problem with syntax
3546 errors if the name included a slash character. The /name= option
3547 now slurps the rest of the string, so it can include any characters
3548 but it must come last in the list of options (after /sender_retain
3551 PH/05 Some modifications to the interface to the fake nameserver for the new
3559 TK/01 Added the "success_on_redirect" address verification option. See
3560 NewStuff for rationale and an example.
3562 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3564 PH/02 Patch to exigrep to allow it to work on syslog lines.
3566 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3567 fread() to skip over the body file's header line, because in Cygwin the
3568 header line is locked and is inaccessible.
3570 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3571 co-exist for some time) to make it clear that it is the Exim ID that is
3572 referenced, not the Message-ID: header line.
3574 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3575 string_format() function, because snprintf() does not exist on all
3578 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3579 ${stat: expansion item.
3581 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3582 protocol synchronization error", to keep the pedants happy.
3584 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3585 well as for IRIX systems, when gcc is being used. See the host.c source
3588 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3590 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3593 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3594 transport and to the smtp transport in LMTP mode.
3596 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3598 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3599 run a filter in a subprocess. This could lead to confusion in subsequent
3600 lookups in the parent process. There should also be a search_tidyup() at
3601 the end of the subprocess.
3603 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3604 only if the host matched helo_try_verify_hosts, which caused the
3605 verification to occur when the EHLO/HELO command was issued. The ACL just
3606 tested the remembered result. Now, if a previous verification attempt has
3607 not happened, "verify = helo" does it there and then.
3609 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3611 TK/03 Fix log output including CR from clamd.
3613 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3614 exist provoked a memory error which could cause a segfault.
3616 PH/15 Installed PCRE 6.2
3618 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3620 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3621 of the problem. Specifically, suggested +O2 rather than +O1 for the
3624 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3626 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3627 record, the port was not correctly logged when the outgoing_port log
3628 selector was set (it logged the transort's default port).
3630 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3631 fallback_hosts, and "hosts" in the smtp transport.
3633 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3634 host errors such as "Connection refused".
3636 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3637 authentication with radiusclient 0.4.9:
3639 - Error returned from rc_read_config was caught wrongly
3640 - Username/password not passed on to radius server due to wrong length.
3642 The presumption is that some radiusclient API changes for 4.51/PH/17
3643 were not taken care of correctly. The code is still untested by me (my
3644 Linux distribution still has 0.3.2 of radiusclient), but it was
3645 contributed by a Radius user.
3647 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3648 expanding the "port" option of the smtp transport.
3650 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3651 while reading a MIME header. Thanks to Tom Hughes for a patch.
3653 PH/24 Include config.h inside local_scan.h so that configuration settings are
3656 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3657 that in an ACL for RCPT (for example), you can examine exactly what was
3660 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3661 commands, but it was not correctly comparing the address with the actual
3662 client host address. Thus, it would show the EHLO address in Received:
3663 header lines when this was not necessary.
3665 PH/27 Added the % operator to ${eval:}.
3667 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3668 it should be ignoring failures (because with -C, for example, it has lost
3669 privilege). It wasn't ignoring creation failures other than "already
3672 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3675 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3676 because an input line was too long, either on its own, or by virtue of
3677 too many continuations, the temporary file was not being removed, and the
3678 return code was incorrect.
3680 PH/31 Missing "BOOL" in function definition in filtertest.c.
3682 PH/32 Applied Sieve patches from the maintainer.
3684 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3686 PH/33 Added "verify = not_blind".
3688 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3689 Local/Makefile (with some defaults set). These are used in built scripts
3690 such as exicyclog, but they have never been used in the exim_install
3691 script (though there are many overriding facilities there). I have
3692 arranged that the exim_install script now takes note of these two
3695 PH/35 Installed configuration files for Dragonfly.
3697 PH/36 When a locally submitted message by a trusted user did not contain a
3698 From: header, and the sender address was obtained from -f or from an SMTP
3699 MAIL command, and the trusted user did not use -F to supply a sender
3700 name, $originator_name was incorrectly used when constructing a From:
3701 header. Furthermore, $originator_name was used for submission mode
3702 messages from external hosts without From: headers in a similar way,
3703 which is clearly wrong.
3705 PH/37 Added control=suppress_local_fixups.
3707 PH/38 When log_selector = +received_sender was set, and the addition of the
3708 sender made the log line's construction buffer exactly full, or one byte
3709 less than full, an overflow happened when the terminating "\n" was
3712 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3713 when the result of a list match is failure because a DNS lookup failed.
3715 PH/40 RM_COMMAND is now used in the building process.
3717 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3718 the "build-* directories that it finds.
3720 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3721 address in a domain literal was a prefix of an interface address.
3723 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3724 when verifying a sender address, unless rewrite_headers is false.
3726 PH/44 Wrote a long comment about why errors_to addresses are verified as
3727 recipients, not senders.
3729 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3730 the ratelimit ACL was added.
3732 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3734 PH/46 Added extra information about PostgreSQL errors to the error string.
3736 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3737 avoiding the need to install special zones in a real server. This is
3738 backwards compatible; if it can't find the fake resolver, it drops back.
3739 Thus, both old and new test suites can be run.
3741 TF/02 Added util/ratelimit.pl
3743 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3744 clock is set back in time.
3746 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3747 Candler <B.Candler@pobox.com>.
3749 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3750 OK for addresses that are the result of redirections.
3752 PH/48 A number of further additions for the benefit of the new test suite,
3753 including a fake gethostbyname() that interfaces to the fake DNS resolver
3756 TF/06 The fix for widen_domains has also been applied to qualify_single and
3757 search_parents which are the other dnslookup options that can cause
3760 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3763 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3765 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3766 service" if it encountered an MX record with an empty target hostname.
3767 The message is now "an MX or SRV record indicated no SMTP service".
3769 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3770 if the DNS of the sending site is misconfigured. This is quite a
3771 common situation. This change restores the behaviour of treating a
3772 helo verification defer as a failure.
3774 PH/51 If self=fail was set on a router, the bounce message did not include the
3775 actual error message.
3781 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3783 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3784 command itself ended in error, the underlying message about the transport
3785 filter timeout was being overwritten with the pipe command error. Now the
3786 underlying error message should be appended to the second error message.
3788 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3790 PH/02 Reduce the amount of output that "make" produces by default. Full output
3791 can still be requested.
3793 PH/03 The warning log line about a condition test deferring for a "warn" verb
3794 was being output only once per connection, rather than after each
3795 occurrence (because it was using the same function as for successful
3796 "warn" verbs). This seems wrong, so I have changed it.
3798 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3799 it should not have, which might have caused a crash in the right
3800 circumstances, but probably never did.
3802 PH/04 Installed a modified version of Tony Finch's patch to make submission
3803 mode fix the return path as well as the Sender: header line, and to
3804 add a /name= option so that you can make the user's friendly name appear
3807 TF/03 Added the control = fakedefer ACL modifier.
3809 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3810 Mark Lowes for thorough testing.
3812 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3814 TK/03 Merged latest SRS patch from Miles Wilton.
3816 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3817 with the definition in sysexits.h (which is #included earlier).
3818 Fortunately, Exim does not actually use EX_OK. The code used to try to
3819 preserve the sysexits.h value, by assuming that macro definitions were
3820 scanned for macro replacements. I have been disabused of this notion,
3821 so now the code just undefines EX_OK before #including unistd.h.
3823 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3824 in the smtp transport. When a block could not be written in a single
3825 write() function, the timeout was being re-applied to each part-write.
3826 This seems wrong - if the receiver was accepting one byte at a time it
3827 would take for ever. The timeout is now adjusted when this happens. It
3828 doesn't have to be particularly precise.
3830 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3831 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3833 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3834 without a domain if the check to <postmaster@domain> fails.
3836 SC/01 Eximstats: added -xls and the ability to specify output files
3837 (patch written by Frank Heydlauf).
3839 SC/02 Eximstats: use FileHandles for outputting results.
3841 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3843 SC/04 Eximstats: fixed display of large numbers with -nvr option
3845 SC/05 Eximstats: fixed merging of reports with empty tables.
3847 SC/06 Eximstats: added the -include_original_destination flag
3849 SC/07 Eximstats: removed tabs and trailing whitespace.
3851 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3853 TK/06 MBOX spool code: Add real "From " MBOX separator line
3854 so the .eml file is really in mbox format (even though
3855 most programs do not really care). Patch from Alex Miller.
3857 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3858 The latter is generated from $received_to and is only set if the
3859 message has one envelope recipient. SA can use these headers,
3860 obviously out-of-the-box. Patch from Alex Miller.
3862 PH/08 The ${def test on a variable was returning false if the variable's
3863 value was "0", contrary to what the specification has always said!
3864 The result should be true unless the variable is empty.
3866 PH/09 The syntax error of a character other than { following "${if
3867 def:variable_name" (after optional whitespace) was not being diagnosed.
3868 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3869 accidental colon was present, for example, could give incorrect results.
3871 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3872 result is used (not including appendfile, where other changes are about
3875 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3876 This involved changing a lot of size variables from int to off_t. It
3877 should work with maildirs and everything.
3879 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3880 spamd dying while we are connected to it.
3882 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3885 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3886 function for BDB 4.3.
3888 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3890 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3891 the sources for PCRE have been re-arranged and more files are now
3894 PH/15 The code I had for printing potentially long long variables in PH/11
3895 above was not the best (it lost precision). The length of off_t variables
3896 is now inspected at build time, and an appropriate printing format (%ld
3897 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3898 to be "long long int" or "long int". This is needed for the internal
3899 formatting function string_vformat().
3901 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3902 the configuration file to be ":syslog", then the script "guesses" where
3903 the logs files are, rather than using the compiled in default. In our
3904 case the guess is not the same as the compiled default, so the script
3905 suddenly stopped working when I started to use syslog. The patch checks
3906 to see if log_file_path is "". If so, it attempts to read it from exim
3907 with no configuration file to get the compiled in version, before it
3908 falls back to the previous guessing code."
3910 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3911 implementing BATV in an Exim configuration. See NewStuff for the gory
3914 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3915 Makefile that are specific to HP-UX.
3917 PH/18 If the "use_postmaster" option was set for a recipient callout together
3918 with the "random" option, the postmaster address was used as the MAIL
3919 FROM address for the random test, but not for the subsequent recipient
3920 test. It is now used for both.
3922 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3923 patch removes a few documentation additions to RFC 3028, because the
3924 latest draft now contains them. It adds the new en;ascii-case comparator
3925 and a new error check for 8bit text in MIME parts. Comparator and
3926 require names are now matched exactly. I enabled the subaddress
3927 extension, but it is not well tested yet (read: it works for me)."
3929 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3930 rework some of the code of TK/09 above to avoid the hardwired use of
3931 "%lld" and "long long". Replaced the call to snprintf() with a call to
3934 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3935 records point to non-existent hosts", "retry timeout exceeded", and
3936 "retry time not reached for any host after a long failure period".
3938 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3939 experimental DomainKeys support:
3941 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3942 (2) On an error such as an illegally used "control", the wrong name for
3943 the control was given.
3945 These problems did NOT occur unless DomainKeys support was compiled.
3947 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3949 PH/24 Added ${if match_ip condition.
3951 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3952 databases so that it will be absolutely obvious if a crash occurs in the
3953 DB library. This is a regular occurrence (often caused by mis-matched
3956 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3957 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3958 picked up on a user's system that detects such things. There doesn't seem
3959 to be a gcc warning option for this - only an attribute that has to be
3960 put on the function's prototype. It seems that in Fedora Core 4 they have
3961 set this on a number of new functions. No doubt there will be more in due
3964 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3965 specify a transport. However, if an address that was verified by such a
3966 router was the subject of a callout, Exim crashed because it tried to
3967 read the rcpt_include_affixes from the non-existent transport. Now it
3968 just assumes that the setting of that option is false. This bug was
3969 introduced by 4.51/PH/31.
3971 PH/28 Changed -d+all to exclude +memory, because that information is very
3972 rarely of interest, but it makes the output a lot bigger. People tend to
3973 do -d+all out of habit.
3975 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3976 code in os-type was giving problems when libc.so lives in lib64, like on
3979 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3980 aren't the modern standard, and it seems that some systems' include files
3981 don't always have them. Exim was already checking for some of the newer
3982 ones like T_AAAA, and defining it itself. I've added checks for all the
3983 record types that Exim uses.
3985 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3986 not automatically generating a new one, as it is supposed to. This
3987 prevented TLS from working. If the file did exist, but contained invalid
3988 data, a new version was generated, as expected. It was only the case of a
3989 non-existent file that was broken.
3991 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3992 with a change in libdomainkeys > 0.64.
3994 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3995 from DNS. If the selector record carries the flag, it now has
3996 precedence over the domain-wide flag.
3998 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4000 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4001 the use of a port name that isn't defined in /etc/services) occurred, the
4002 message was deferred as in a normal delivery, and thus remained on the
4003 spool, instead of being failed because of the mua_wrapper setting. This
4004 is now fixed, and I tidied up some of the mua_wrapper messages at the
4007 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4008 the messages in a hash of arrays rather than using individual hashes.
4009 This is a bit cleaner and results in dramatic memory savings, albeit
4010 at a slight CPU cost.
4012 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4013 as requested by Marc Sherman.
4015 SC/10 Eximstats: added histograms for user specified patterns as requested
4018 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4020 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4021 fopen() in the content-scanning modules that did not already have it.
4027 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4028 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4030 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4032 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4034 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4035 to test Sieve filters that use "vacation".
4037 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4038 that changes the way the GnuTLS parameters are stored in the cache file.
4039 The new format can be generated externally. For backward compatibility,
4040 if the data in the cache doesn't make sense, Exim assumes it has read an
4041 old-format file, and it generates new data and writes a new file. This
4042 means that you can't go back to an older release without removing the
4045 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4046 work if there are any delivery delays because "one_time" forces the
4047 parent to be marked "delivered", so its unseen clone is never tried
4048 again. For this reason, Exim now forbids the simultaneous setting of
4051 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4052 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4053 entry if you want to know the details. The fix, however, neglected to
4054 consider the case when local delivery batching is involved. The test for
4055 "previously delivered" was not happening when checking to see if an
4056 address could be batched with a previous (undelivered) one; under
4057 certain circumstances this could lead to multiple deliveries to the same
4060 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4061 in its include files, and this causes problems building Exim.
4063 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4064 header_syntax) but Exim was just ignoring anything given after a slash.
4065 In particular, this caused confusion with an attempt to use "verify =
4066 reverse_host_lookup/defer_ok". An error is now given when options are
4067 supplied for verify items that do not have them. (Maybe reverse_host_
4068 lookup should have a defer_ok option, but that's a different point.)
4070 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4071 defined by RFC 821) to 2048, because there were problems with some AUTH
4072 commands, and RFC 1869 says the size should be increased for extended
4073 SMTP commands that take arguments.
4075 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4078 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4079 "unknown" error; now it says that the functionality isn't in the binary.
4081 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4082 an address' error message when a string expansion fails (syntax or
4083 whatever). Otherwise the password may appear in the log. Following change
4084 PH/42 below, there is no longer a chance of it appearing in a bounce
4087 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4089 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4090 of its addresses was ever tried. (Bugzilla bug #2.)
4092 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4093 the result incorrectly in the debug output. (It correctly added a newline
4094 to what was transported.)
4096 TF/01 Added $received_time.
4098 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4099 commented out examples of how to interface to a virus scanner and to
4100 SpamAssassin. Also added commented examples of av_scanner and
4101 spamd_address settings.
4103 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4104 and controls are allowed in which ACLs. There were a couple of minor
4105 errors. Some of the entries in the conditions table (which is a table of
4106 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4107 negation of where the condition IS allowed.
4109 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4111 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4112 header file does not have a version number, so I've had to invent a new
4113 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4114 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4115 radiusclient), but it was contributed by a Radius user.
4117 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4118 files or directories, for OpenSSL.
4120 PH/19 When an Exim process that is running as root has to create an Exim log
4121 file, it does so in a subprocess that runs as exim:exim so as to get the
4122 ownership right at creation (otherwise, other Exim processes might see
4123 the file with the wrong ownership). There was no test for failure of this
4124 fork() call, which would lead to the process getting stuck as it waited
4125 for a non-existent subprocess. Forks do occasionally fail when resources
4126 run out. I reviewed all the other calls to fork(); they all seem to check
4129 PH/20 When checking for unexpected SMTP input at connect time (before writing
4130 the banner), Exim was not dealing correctly with a non-positive return
4131 from the read() function. If the client had disconnected by this time,
4132 the result was a log entry for a synchronization error with an empty
4133 string after "input=" when read() returned zero. If read() returned -1
4134 (an event I could not check), uninitialized data bytes were printed.
4135 There were reports of junk text (parts of files, etc) appearing after
4138 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4140 PH/22 Added support for macro redefinition, and (re)definition in between
4141 driver and ACL definitions.
4143 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4144 forgetting to use the resulting value; it was using the unexpanded value.
4146 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4147 hadn't been configured. The fix is from Juergen Kreileder, who
4148 understands it better than I do:
4150 "Here's what I see happening with three configured cyrus_sasl
4151 authenticators configured (plain, login, cram-md5):
4153 On startup auth_cyrus_sasl_init() gets called for each of these.
4154 This means three calls to sasl_listmech() without a specified mech_list.
4155 => SASL tests which mechs of all available mechs actually work
4156 => three warnings about OTP not working
4157 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4159 With the patch, sasl_listmech() also gets called three times. But now
4160 SASL's mech_list option is set to the server_mech specified in the the
4161 authenticator. Or in other words, the answer from sasl_listmech()
4162 gets limited to just the mech you're testing for (which is different
4164 => the return list contains just 'plain' or 'login', 'cram-md5' or
4165 nothing depending on the value of ob->server_mech.
4167 I've just tested the patch: Authentication still works fine,
4168 unavailable mechs specified in the exim configuration are still
4169 caught, and the auth.log warnings about OTP are gone."
4171 PH/25 When debugging is enabled, the contents of the command line are added
4172 to the debugging output, even when log_selector=+arguments is not
4175 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4176 answer is "GNU", and only if the return is "GNU/something" is the answer
4179 PH/27 $acl_verify_message is now set immediately after the failure of a
4180 verification in an ACL, and so is available in subsequent modifiers. In
4181 particular, the message can be preserved by coding like this:
4183 warn !verify = sender
4184 set acl_m0 = $acl_verify_message
4186 Previously, $acl_verify_message was set only while expanding "message"
4187 and "log_message" when a very denied access.
4189 PH/28 Modified OS/os.c-Linux with
4191 -#ifndef OS_LOAD_AVERAGE
4192 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4194 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4195 nomenclature these days.)
4197 PH/29 Installed patch from the Sieve maintainer that adds the options
4198 sieve_useraddress and sieve_subaddress to the redirect router.
4200 PH/30 In these circumstances:
4201 . Two addresses routed to the same list of hosts;
4202 . First host does not offer TLS;
4203 . First host accepts first address;
4204 . First host gives temporary error to second address;
4205 . Second host offers TLS and a TLS session is established;
4206 . Second host accepts second address.
4207 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4208 and peerdn, if requested) that were in fact used only for the second
4211 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4212 attention to any local part prefix or suffix that was matched by the
4213 router that accepted the address. It now behaves in the same way as it
4214 does for delivery: the affixes are removed from the local part unless
4215 rcpt_include_affixes is set on the transport.
4217 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4218 timeout during the DATA phase of an incoming message.
4220 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4221 applied a patch sanctioned by the Sieve maintainer.
4223 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4224 the uid or gid is negative. A case of a negative gid caused this to be
4225 noticed. The fix allows for either to be negative.
4227 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4228 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4231 PH/36 The change PH/12 above was broken. Fixed it.
4233 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4234 the grounds that routing the same address twice would always produce the
4235 same answer. This might have been true once, but it is certainly no
4236 longer true now. Routing a child address may depend on the previous
4237 routing that produced that child. Some complicated redirection strategies
4238 went wrong when messages had multiple recipients, and made Exim's
4239 behaviour dependent on the order in which the addresses were given.
4241 I have moved the duplicate checking until after the routing is complete.
4242 Exim scans the addresses that are assigned to local and remote
4243 transports, and removes any duplicates. This means that more work will be
4244 done, as duplicates will always all be routed, but duplicates are
4245 presumably rare, so I don't expect this is of any significance.
4247 For deliveries to pipes, files, and autoreplies, the duplicate checking
4248 still happens during the routing process, since they are not going to be
4251 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4252 It corrects a timeout issue with spamd. This is Ian's comment: "The
4253 background is that sometimes spamd either never reads data from a
4254 connection it has accepted, or it never writes response data. The exiscan
4255 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4256 blindly assumes that writes won't block so it may never time out."
4258 PH/39 Allow G after quota size as well as K and M.
4260 PH/40 The value set for $authenticated_id in an authenticator may not contain
4261 binary zeroes or newlines because the value is written to log lines and
4262 to spool files. There was no check on this. Now the value is run through
4263 the string_printing() function so that such characters are converted to
4264 printable escape sequences.
4266 PH/41 $message_linecount is a new variable that contains the total number of
4267 lines in the message. Compare $body_linecount, which is the count for the
4270 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4271 bounce and delay warning messages, except in certain special cases, which
4274 (a) An SMTP error message from a remote host;
4275 (b) A message specified in a :fail: redirection;
4276 (c) A message specified in a "fail" command in a system filter;
4277 (d) A message specified in a FAIL return from the queryprogram router;
4278 (e) A message specified by the cannot_route_message router option.
4280 In these cases only, Exim does include the error details in bounce and
4281 warning messages. There are also a few cases where bland messages such
4282 as "unrouteable address" or "local delivery error" are given.
4284 PH/43 $value is now also set for the "else" part of a ${run expansion.
4286 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4287 being worked on, but at least Exim now implements the latest version to
4290 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4291 process to complete was treated as a delivery failure, a timeout while
4292 writing the message to the pipe was logged, but erroneously treated as a
4293 successful delivery. Such timeouts include transport filter timeouts. For
4294 consistency with the overall process timeout, these timeouts are now
4295 treated as errors, giving rise to delivery failures by default. However,
4296 there is now a new Boolean option for the pipe transport called
4297 timeout_defer, which, if set TRUE, converts the failures into defers for
4298 both kinds of timeout. A transport filter timeout is now identified in
4301 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4302 systems where "make" and "gmake" are different, calling "gmake" at top
4303 level broke things. I've arranged for the value of $(MAKE) to be passed
4304 from the Makefile to this script so that it can call the same version of
4308 A note about Exim versions 4.44 and 4.50
4309 ----------------------------------------
4311 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4312 changes of various kinds. As a consequence, a big documentation update was
4313 needed. This delayed the release for rather longer than seemed good, especially
4314 in the light of a couple of (minor) security issues. Therefore, the changes
4315 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4316 release. So 4.44 and 4.50 are in effect two different branches that both start
4319 I have left the 4.50 change log unchanged; it contains all the changes since
4320 4.43. The change log for 4.44 is below; many of its items are identical to
4321 those for 4.50. This seems to be the most sensible way to preserve the
4322 historical information.
4328 1. Minor wording change to the doc/README.SIEVE file.
4330 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4331 computation of the current number of files was incorrect.
4333 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4334 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4337 4. Give more explanation in the error message when the command for a transport
4338 filter fails to execute.
4340 5. There are several places where Exim runs a non-Exim command in a
4341 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4342 was being done only for the command run by the queryprogram router. It is
4343 now done for all such subprocesses. The other cases are: ${run, transport
4344 filters, and the commands run by the lmtp and pipe transports.
4346 6. Added CONFIGURE_GROUP build-time option.
4348 7. Some older OS have a limit of 256 on the maximum number of file
4349 descriptors. Exim was using setrlimit() to set 1000 as a large value
4350 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4351 systems. I've change it so that if it can't get 1000, it tries for 256.
4353 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4354 was an oversight, and furthermore, ever since the addition of extra
4355 controls (e.g. 4.43/32), the checks on when to allow different forms of
4356 "control" were broken. There should now be diagnostics for all cases when a
4357 control that does not make sense is encountered.
4359 9. Added the /retain_sender option to "control=submission".
4361 10. $recipients is now available in the predata ACL (oversight).
4363 11. Tidy the search cache before the fork to do a delivery from a message
4364 received from the command line. Otherwise the child will trigger a lookup
4365 failure and thereby defer the delivery if it tries to use (for example) a
4366 cached ldap connection that the parent has called unbind on.
4368 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4369 of $address_data from the recipient verification was clobbered by the
4370 sender verification.
4372 13. The value of address_data from a sender verification is now available in
4373 $sender_address_data in subsequent conditions in the ACL statement.
4375 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4377 15. Added a new option "connect=<time>" to callout options, to set a different
4380 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4381 was its contents. (It was OK if the option was not defined at all.)
4383 17. A "Completed" log line is now written for messages that are removed from
4384 the spool by the -Mrm option.
4386 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4387 information about exactly what failed.
4389 19. Added -dd to debug only the daemon process.
4391 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4392 handles timeouts, both on the server side and network timeouts. Renamed the
4393 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4395 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4396 It is now set to "smtps".
4398 22. $host_address is now set to the target address during the checking of
4399 ignore_target_hosts.
4401 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4402 being passed; this would have caused $sender_host_name to have been used if
4403 matching the list had actually called for a host name (not very likely,
4404 since this list is usually IP addresses). A host name is now passed as
4407 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4408 code that specifies a non-restarting handler (typically sigaction() in
4409 modern systems) in an attempt to fix a rare and obscure crash bug.
4411 25. Narrowed the window for a race in the daemon that could cause it to ignore
4412 SIGCHLD signals. This is not a major problem, because they are used only to
4413 wake it up if nothing else does.
4415 26. A malformed maildirsize file could cause Exim to calculate negative values
4416 for the mailbox size or file count. Odd effects could occur as a result.
4417 The maildirsize information is now recalculated if the size or filecount
4420 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4421 support for a long time. Removed HAVE_SYS_VFS_H.
4423 28. Installed the latest version of exipick from John Jetmore.
4425 29. In an address list, if the pattern was not a regular expression, an empty
4426 subject address (from a bounce message) matched only if the pattern was an
4427 empty string. Non-empty patterns were not even tested. This was the wrong
4428 because it is perfectly reasonable to use an empty address as part of a
4429 database query. An empty address is now tested by patterns that are
4430 lookups. However, all the other forms of pattern expect the subject to
4431 contain a local part and a domain, and therefore, for them, an empty
4432 address still always fails if the pattern is not itself empty.
4434 30. Exim went into a mad DNS loop when attempting to do a callout where the
4435 host was specified on an smtp transport, and looking it up yielded more
4436 than one IP address.
4438 31. Re-factored the code for checking spool and log partition space into a
4439 function that finds that data and another that does the check. The former
4440 is then used to implement four new variables: $spool_space, $log_space,
4441 $spool_inodes, and $log_inodes.
4443 32. The RFC2047 encoding function was originally intended for short strings
4444 such as real names; it was not keeping to the 75-character limit for
4445 encoded words that the RFC imposes. It now respects the limit, and
4446 generates multiple encoded words if necessary. To be on the safe side, I
4447 have increased the buffer size for the ${rfc2047: expansion operator from
4450 33. It is now permitted to omit both strings after an "if" condition; if the
4451 condition is true, the result is "true". As before, when the second string
4452 is omitted, a false condition yields an empty string. This makes it less
4453 cumbersome to write custom ACL and router conditions.
4455 34. Failure to deliver a bounce message always caused it to be frozen, even if
4456 there was an errors_to setting on the router. The errors_to setting is now
4459 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4460 canonical form (fully expanded) before being placed in
4461 $sender_host_address.
4463 36. The table in the code that translates DNS record types into text (T_A to
4464 "A" for instance) was missing entries for NS and CNAME. It is just possible
4465 that this could have caused confusion if both these types were looked up
4466 for the same domain, because the text type is used as part of Exim's
4467 per-process caching. But the chance of anyone hitting this buglet seems
4470 37. The dnsdb lookup has been extended in a number of ways.
4472 (1) There is a new type, "zns", which walks up the domain tree until it
4473 finds some nameserver records. It should be used with care.
4475 (2) There is a new type, "mxh", which is like "mx" except that it returns
4476 just the host names, not the priorities.
4478 (3) It is now possible to give a list of domains (or IP addresses) to be
4479 looked up. The behaviour when one of the lookups defers can be
4480 controlled by a keyword.
4482 (4) It is now possible to specify the separator character for use when
4483 multiple records are returned.
4485 38. The dnslists ACL condition has been extended: it is now possible to supply
4486 a list of IP addresses and/or domains to be looked up in a particular DNS
4489 39. Added log_selector=+queue_time_overall.
4491 40. When running the queue in the test harness, wait just a tad after forking a
4492 delivery process, to get repeatability of debugging output.
4494 41. Include certificate and key file names in error message when GnuTLS fails
4495 to set them up, because the GnuTLS error message doesn't include the name
4496 of the failing file when there is a problem reading it.
4498 42. Allow both -bf and -bF in the same test run.
4500 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4502 44. The "Exiscan patch" is now merged into the mainline Exim source.
4504 45. Sometimes the final signoff response after QUIT could fail to get
4505 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4506 < 0 before doing a fflush(). This bug looks as though it goes back to the
4507 introduction of TLS in release 3.20, but "sometimes" must have been rare
4508 because the tests only now provoked it.
4510 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4511 (this can affect the format of dates).
4513 47. exim_tidydb, when checking for the continued existence of a message for
4514 which it has found a message-specific retry record, was not finding
4515 messages that were in split spool directories. Consequently, it was
4516 deleting retry records that should have stayed in existence.
4518 48. Steve fixed some bugs in eximstats.
4520 49. The SPA authentication driver was not abandoning authentication and moving
4521 on to the next authenticator when an expansion was forced to fail,
4522 contradicting the general specification for all authenticators. Instead it
4523 was generating a temporary error. It now behaves as specified.
4525 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4526 (the order specifies the preference for clients). The order is now AES256,
4527 AES128, 3DES, ARCFOUR128.
4529 51. Small patch to Sieve code - explicitly set From: when generating an
4532 52. Exim crashed if a remote delivery caused a very long error message to be
4533 recorded - for instance if somebody sent an entire SpamAssassin report back
4534 as a large number of 550 error lines. This bug was coincidentally fixed by
4535 increasing the size of one of Exim's internal buffers (big_buffer) that
4536 happened as part of the Exiscan merge. However, to be on the safe side, I
4537 have made the code more robust (and fixed the comments that describe what
4540 53. Now that there can be additional text after "Completed" in log lines (if
4541 the queue_time_overall log selector is set), a one-byte patch to exigrep
4542 was needed to allow it to recognize "Completed" as not the last thing in
4545 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4546 patch that reportedly fixes this has been added. I am not expert enough to
4547 create a test for it. This is what the patch creator wrote:
4549 "I found a little strange behaviour of ldap code when working with
4550 Windows 2003 AD Domain, where users was placed in more than one
4551 Organization Units. When I tried to give exim partial DN, the exit code
4552 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4553 But simultaneously result of request was absolutely normal ldap result,
4554 so I produce this patch..."
4556 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4557 REFERENCE, so I have modified the code to exclude the patch when that macro
4560 55. Some experimental protocols are using DNS PTR records for new purposes. The
4561 keys for these records are domain names, not reversed IP addresses. The
4562 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4563 leaves it alone. Component reversal etc. now happens only for IP addresses.
4566 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4568 57. Double the size of the debug message buffer (to 2048) so that more of very
4569 long debug lines gets shown.
4571 58. The exicyclog utility now does better if the number of log files to keep
4572 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4574 59. Two changes related to the smtp_active_hostname option:
4576 (1) $smtp_active_hostname is now available as a variable.
4577 (2) The default for smtp_banner uses $smtp_active_hostname instead
4578 of $primary_hostname.
4580 60. The host_aton() function is supposed to be passed a string that is known
4581 to be a valid IP address. However, in the case of IPv6 addresses, it was
4582 not checking this. This is a hostage to fortune. Exim now panics and dies
4583 if the condition is not met. A case was found where this could be provoked
4584 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4585 components; fortuitously, this particular loophole had already been fixed
4586 by change 4.50/55 above.
4588 If there are any other similar loopholes, the new check in host_aton()
4589 itself should stop them being exploited. The report I received stated that
4590 data on the command line could provoke the exploit when Exim was running as
4591 exim, but did not say which command line option was involved. All I could
4592 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4593 running as the user.
4596 61. There was a buffer overflow vulnerability in the SPA authentication code
4597 (which came originally from the Samba project). I have added a test to the
4598 spa_base64_to_bits() function which I hope fixes it.
4601 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4602 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4604 63. The daemon start-up calls getloadavg() while still root for those OS that
4605 need the first call to be done as root, but it missed one case: when
4606 deliver_queue_load_max is set with deliver_drop_privilege. This is
4607 necessary for the benefit of the queue runner, because there is no re-exec
4608 when deliver_drop_privilege is set.
4610 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4611 This has been fixed.
4613 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4614 was in use, was not putting the data itself into the right store pool;
4615 consequently, it could be overwritten for a subsequent message in the same
4616 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4619 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4621 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4622 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4623 the function were treating the return as a boolean value, which happened to
4624 work because 0=false and not-0=true, but is not correct code.
4626 68. The host_aton() function was not handling scoped IPv6 addresses (those
4627 with, for example, "%eth0" on the end) correctly.
4629 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4630 negated items (that is, ~something) in unsigned ints. Some compilers
4631 apparently mutter when there is no cast.
4633 70. If an address verification called from an ACL failed, and did not produce a
4634 user-specific message (i.e. there was only a "system" message), nothing was
4635 put in $acl_verify_message. In this situation, it now puts the system
4638 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4639 session; change 4.31/43 added the unwanted input to the log line - except
4640 that it did not do this in the start of session case. It now does.
4642 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4643 This is wrong and can cause the other end to generate a synchronization
4644 error if it is another Exim or anything else that does the synchronization
4645 check. A QUIT command is no longer sent after a timeout.
4647 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4648 during host lookups.
4650 74. The defer_ok option of callout verification was not working if it was used
4651 when verifying addresses in header lines, that is, for this case:
4653 verify = header_sender/callout=defer_ok
4655 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4656 those file descriptors could be used for SMTP connections. If anything
4657 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4658 could be sent to the SMTP client, causing chaos. The daemon now opens
4659 stdin, stdout, and stderr to /dev/null when it puts itself into the
4662 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4663 log by default. The user can override this with suitable Perl magic.
4665 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4666 the log message when discard triggers, was not working for the DATA ACL or
4667 for the non-SMTP ACL.
4669 78. Error message wording change in sieve.c.
4671 79. If smtp_accept_max_per_host was set, the number of connections could be
4672 restricted to fewer than expected, because the daemon was trying to set up
4673 a new connection before checking whether the processes handling previous
4674 connections had finished. The check for completed processes is now done
4675 earlier. On busy systems, this bug wouldn't be noticed because something
4676 else would have woken the daemon, and it would have reaped the completed
4679 80. If a message was submitted locally by a user whose login name contained one
4680 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4681 It caused a spool format error. I have fixed the spool reading code. A
4682 related problem was that the "from" clause in the Received: line became
4683 illegal because of the space(s). It is now covered by ${quote_local_part.
4685 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4688 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4690 83. Patch from Sieve maintainer to fix unterminated string problem in
4691 "vacation" handling.
4693 84. Some minor changes to the Linux configuration files to help with other
4694 OS variants using glibc.
4696 85. One more patch for Sieve to update vacation handling to latest spec.
4699 ----------------------------------------------------
4700 See the note above about the 4.44 and 4.50 releases.
4701 ----------------------------------------------------
4707 1. Change 4.43/35 introduced a bug that caused file counts to be
4708 incorrectly computed when quota_filecount was set in an appendfile
4711 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4712 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4715 3. Give more explanation in the error message when the command for a transport
4716 filter fails to execute.
4718 4. There are several places where Exim runs a non-Exim command in a
4719 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4720 was being done only for the command run by the queryprogram router. It is
4721 now done for all such subprocesses. The other cases are: ${run, transport
4722 filters, and the commands run by the lmtp and pipe transports.
4724 5. Some older OS have a limit of 256 on the maximum number of file
4725 descriptors. Exim was using setrlimit() to set 1000 as a large value
4726 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4727 systems. I've change it so that if it can't get 1000, it tries for 256.
4729 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4730 was an oversight, and furthermore, ever since the addition of extra
4731 controls (e.g. 4.43/32), the checks on when to allow different forms of
4732 "control" were broken. There should now be diagnostics for all cases when a
4733 control that does not make sense is encountered.
4735 7. $recipients is now available in the predata ACL (oversight).
4737 8. Tidy the search cache before the fork to do a delivery from a message
4738 received from the command line. Otherwise the child will trigger a lookup
4739 failure and thereby defer the delivery if it tries to use (for example) a
4740 cached ldap connection that the parent has called unbind on.
4742 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4743 of $address_data from the recipient verification was clobbered by the
4744 sender verification.
4746 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4747 was its contents. (It was OK if the option was not defined at all.)
4749 11. A "Completed" log line is now written for messages that are removed from
4750 the spool by the -Mrm option.
4752 12. $host_address is now set to the target address during the checking of
4753 ignore_target_hosts.
4755 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4756 being passed; this would have caused $sender_host_name to have been used if
4757 matching the list had actually called for a host name (not very likely,
4758 since this list is usually IP addresses). A host name is now passed as
4761 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4762 code that specifies a non-restarting handler (typically sigaction() in
4763 modern systems) in an attempt to fix a rare and obscure crash bug.
4765 15. Narrowed the window for a race in the daemon that could cause it to ignore
4766 SIGCHLD signals. This is not a major problem, because they are used only to
4767 wake it up if nothing else does.
4769 16. A malformed maildirsize file could cause Exim to calculate negative values
4770 for the mailbox size or file count. Odd effects could occur as a result.
4771 The maildirsize information is now recalculated if the size or filecount
4774 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4775 support for a long time. Removed HAVE_SYS_VFS_H.
4777 18. Updated exipick to current release from John Jetmore.
4779 19. Allow an empty sender to be matched against a lookup in an address list.
4780 Previously the only cases considered were a regular expression, or an
4783 20. Exim went into a mad DNS lookup loop when doing a callout where the
4784 host was specified on the transport, if the DNS lookup yielded more than
4787 21. The RFC2047 encoding function was originally intended for short strings
4788 such as real names; it was not keeping to the 75-character limit for
4789 encoded words that the RFC imposes. It now respects the limit, and
4790 generates multiple encoded words if necessary. To be on the safe side, I
4791 have increased the buffer size for the ${rfc2047: expansion operator from
4794 22. Failure to deliver a bounce message always caused it to be frozen, even if
4795 there was an errors_to setting on the router. The errors_to setting is now
4798 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4799 canonical form (fully expanded) before being placed in
4800 $sender_host_address.
4802 24. Updated eximstats to version 1.33
4804 25. Include certificate and key file names in error message when GnuTLS fails
4805 to set them up, because the GnuTLS error message doesn't include the name
4806 of the failing file when there is a problem reading it.
4808 26. Expand error message when OpenSSL has problems setting up cert/key files.
4811 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4812 (this can affect the format of dates).
4814 28. exim_tidydb, when checking for the continued existence of a message for
4815 which it has found a message-specific retry record, was not finding
4816 messages that were in split spool directories. Consequently, it was
4817 deleting retry records that should have stayed in existence.
4819 29. eximstats updated to version 1.35
4820 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4821 1.35 - bugfix such that pie charts by volume are generated correctly
4823 30. The SPA authentication driver was not abandoning authentication and moving
4824 on to the next authenticator when an expansion was forced to fail,
4825 contradicting the general specification for all authenticators. Instead it
4826 was generating a temporary error. It now behaves as specified.
4828 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4829 (the order specifies the preference for clients). The order is now AES256,
4830 AES128, 3DES, ARCFOUR128.
4832 31. Small patch to Sieve code - explicitly set From: when generating an
4835 32. Exim crashed if a remote delivery caused a very long error message to be
4836 recorded - for instance if somebody sent an entire SpamAssassin report back
4837 as a large number of 550 error lines. This bug was coincidentally fixed by
4838 increasing the size of one of Exim's internal buffers (big_buffer) that
4839 happened as part of the Exiscan merge. However, to be on the safe side, I
4840 have made the code more robust (and fixed the comments that describe what
4843 33. Some experimental protocols are using DNS PTR records for new purposes. The
4844 keys for these records are domain names, not reversed IP addresses. The
4845 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4846 leaves it alone. Component reversal etc. now happens only for IP addresses.
4849 34. The host_aton() function is supposed to be passed a string that is known
4850 to be a valid IP address. However, in the case of IPv6 addresses, it was
4851 not checking this. This is a hostage to fortune. Exim now panics and dies
4852 if the condition is not met. A case was found where this could be provoked
4853 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4854 components; fortuitously, this particular loophole had already been fixed
4855 by change 4.50/55 or 4.44/33 above.
4857 If there are any other similar loopholes, the new check in host_aton()
4858 itself should stop them being exploited. The report I received stated that
4859 data on the command line could provoke the exploit when Exim was running as
4860 exim, but did not say which command line option was involved. All I could
4861 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4862 running as the user.
4865 35. There was a buffer overflow vulnerability in the SPA authentication code
4866 (which came originally from the Samba project). I have added a test to the
4867 spa_base64_to_bits() function which I hope fixes it.
4870 36. The daemon start-up calls getloadavg() while still root for those OS that
4871 need the first call to be done as root, but it missed one case: when
4872 deliver_queue_load_max is set with deliver_drop_privilege. This is
4873 necessary for the benefit of the queue runner, because there is no re-exec
4874 when deliver_drop_privilege is set.
4876 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4877 was in use, was not putting the data itself into the right store pool;
4878 consequently, it could be overwritten for a subsequent message in the same
4879 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4882 38. Sometimes the final signoff response after QUIT could fail to get
4883 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4884 < 0 before doing a fflush(). This bug looks as though it goes back to the
4885 introduction of TLS in release 3.20, but "sometimes" must have been rare
4886 because the tests only now provoked it.
4892 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4893 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4894 Now it returns an integer. A number of calls were still expecting a T/F
4895 return. Fortuitously, in all cases, the tests worked in OK situations,
4896 which is the norm. However, things would have gone wrong on any write
4897 failures on the smtp file descriptor. This function is used when sending
4898 messages over SMTP and also when doing verify callouts.
4900 2. When Exim is called to do synchronous delivery of a locally submitted
4901 message (the -odf or -odi options), it no longer closes stderr before doing
4904 3. Implemented the mua_wrapper option.
4906 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4908 5. Implemented the functions header_remove(), header_testname(),
4909 header_add_at_position(), and receive_remove_recipient(), and exported them
4912 6. If an ACL "warn" statement specified the addition of headers, Exim already
4913 inserted X-ACL-Warn: at the start if there was no header name. However, it
4914 was not making this test for the second and subsequent header lines if
4915 there were newlines in the string. This meant that an invalid header could
4916 be inserted if Exim was badly configured.
4918 7. Allow an ACL "warn" statement to add header lines at the start or after all
4919 the Received: headers, as well as at the end.
4921 8. Added the rcpt_4xx retry error code.
4923 9. Added postmaster_mailfrom=xxx to callout verification option.
4925 10. Added mailfrom=xxxx to the callout verification option, for verify=
4928 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4929 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4931 12. Inserted some casts to stop certain compilers warning when using pointer
4932 differences as field lengths or precisions in printf-type calls (mostly
4933 affecting debugging statements).
4935 13. Added optional readline() support for -be (dynamically loaded).
4937 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4938 same clock tick as a message's arrival, so that its received time was the
4939 same as the "first fail" time on the retry record, and that message
4940 remained on the queue past the ultimate address timeout, every queue runner
4941 would try a delivery (because it was past the ultimate address timeout) but
4942 after another failure, the ultimate address timeout, which should have then
4943 bounced the address, did not kick in. This was a "< instead of <=" error;
4944 in most cases the first failure would have been in the next clock tick
4945 after the received time, and all would be well.
4947 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4948 being recognized when the domain list was tested by the match_domain
4949 condition in an expansion string.
4951 16. Added the ${str2b64: operator.
4953 17. Exim was always calling setrlimit() to set a large limit for the number of
4954 processes, without checking whether the existing limit was already
4955 adequate. (It did check for the limit on file descriptors.) Furthermore,
4956 errors from getrlimit() and setrlimit() were being ignored. Now they are
4957 logged to the main and panic logs, but Exim does carry on, to try to do its
4958 job under whatever limits there are.
4960 18. Imported PCRE 5.0.
4962 19. Trivial typo in log message " temporarily refused connection" (the leading
4965 20. If the log selector return_path_on_delivery was set and an address was
4966 redirected to /dev/null, the delivery process crashed because it assumed
4967 that a return path would always be set for a "successful" delivery. In this
4968 case, the whole delivery is bypassed as an optimization, and therefore no
4971 21. Internal re-arrangement: the function for sending a challenge and reading
4972 a response while authentication was assuming a zero-terminated challenge
4973 string. It's now changed to take a pointer and a length, to allow for
4974 binary data in such strings.
4976 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4978 23. Exim was not respecting finduser_retries when seeking the login of the
4979 uid under which it was called; it was always trying 10 times. (The default
4980 setting of finduser_retries is zero.) Also, it was sleeping after the final
4981 failure, which is pointless.
4983 24. Implemented tls_on_connect_ports.
4985 25. Implemented acl_smtp_predata.
4987 26. If the domain in control=submission is set empty, Exim assumes that the
4988 authenticated id is a complete email address when it generates From: or
4989 Sender: header lines.
4991 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4992 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4993 chown and chgrp in /bin and hostname in /usr/bin.
4995 28. Exim was keeping the "process log" file open after each use, just as it
4996 does for the main log. This opens the possibility of it remaining open for
4997 long periods when the USR1 signal hits a daemon. Occasional processlog
4998 errors were reported, that could have been caused by this. Anyway, it seems
4999 much more sensible not to leave this file open at all, so that is what now
5002 29. The long-running daemon process does not normally write to the log once it
5003 has entered its main loop, and it closes the log before doing so. This is
5004 so that log files can straightforwardly be renamed and moved. However,
5005 there are a couple of unusual error situations where the daemon does write
5006 log entries, and I had neglected to close the log afterwards.
5008 30. The text of an SMTP error response that was received during a remote
5009 delivery was being truncated at 512 bytes. This is too short for some of
5010 the long messages that one sometimes sees. I've increased the limit to
5013 31. It is now possible to make retry rules that apply only when a message has a
5014 specific sender, in particular, an empty sender.
5016 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5017 it possible to be selective about when SMTP synchronization is enforced.
5019 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5021 32. Implemented hosts_connection_nolog.
5023 33. Added an ACL for QUIT.
5025 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5028 35. Added mailbox_size and mailbox_filecount to appendfile.
5030 36. Added control = no_multiline_responses to ACLs.
5032 37. There was a bug in the logic of the code that waits for the clock to tick
5033 in the case where the clock went backwards by a substantial amount such
5034 that the microsecond fraction of "now" was more than the microsecond
5035 fraction of "then" (but the whole seconds number was less).
5037 38. Added support for the libradius Radius client library this is found on
5038 FreeBSD (previously only the radiusclient library was supported).
5044 1. When certain lookups returned multiple values in the form name=value, the
5045 quoting of the values was not always being done properly. Specifically:
5046 (a) If the value started with a double quote, but contained no whitespace,
5048 (b) If the value contained whitespace other than a space character (i.e.
5049 tabs or newlines or carriage returns) it was not quoted.
5050 This fix has been applied to the mysql and pgsql lookups by writing a
5051 separate quoting function and calling it from the lookup code. The fix
5052 should probably also be applied to nisplus, ibase and oracle lookups, but
5053 since I cannot test any of those, I have not disturbed their existing code.
5055 2. A hit in the callout cache for a specific address caused a log line with no
5056 reason for rejecting RCPT. Now it says "Previous (cached) callout
5057 verification failure".
5059 3. There was an off-by-one bug in the queryprogram router. An over-long
5060 return line was truncated at 256 instead of 255 characters, thereby
5061 overflowing its buffer with the terminating zero. As well as fixing this, I
5062 have increased the buffer size to 1024 (and made a note to document this).
5064 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5065 when Exim is waiting for an SMTP response from a remote server, Exim
5066 restarts its select() call on the socket, thereby resetting its timeout.
5067 This is not a problem when such interrupts are rare. Somebody set up a cron
5068 job to run exiwhat every 2 minutes, which is less than the normal select()
5069 timeout (5 or 10 minutes). This meant that the select() timeout never
5070 kicked in because it was always reset. I have fixed this by comparing the
5071 time when an interrupt arrives with the time at the start of the first call
5072 to select(). If more time than the timeout has elapsed, the interrupt is
5073 treated as a timeout.
5075 5. Some internal re-factoring in preparation for the addition of Sieve
5076 extensions (by MH). In particular, the "personal" test is moved to a
5077 separate function, and given an option for scanning Cc: and Bcc: (which is
5078 not set for Exim filters).
5080 6. When Exim created an email address using the login of the caller as the
5081 local part (e.g. when creating a From: or Sender: header line), it was not
5082 quoting the local part when it contained special characters such as @.
5084 7. Installed new OpenBSD configuration files.
5086 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5087 try to make them clearer.
5089 9. Callout options, other than the timeout value, were being ignored when
5090 verifying sender addresses in header lines. For example, when using
5092 verify = header_sender/callout=no_cache
5094 the cache was (incorrectly) being used.
5096 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5097 only the Cygwin environment.
5099 11. When return_path_on_delivery was set as a log selector, if different remote
5100 addresses in the same message used different return paths and parallel
5101 remote delivery occurred, the wrong values would sometimes be logged.
5102 (Whenever a remote delivery process finished, the return path value from
5103 the most recently started remote delivery process was logged.)
5105 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5106 lines when AUTH and/or TLS are in use. This is the "received protocol"
5107 field. Exim used to use "asmtp" for authenticated SMTP, without any
5108 indication (in the protocol name) for TLS use. Now it follows the RFC and
5109 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5110 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5111 names appear in log lines as well as in Received: header lines.
5113 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5114 extensions, and comparison tests, and to fix some bugs.
5116 14. Changes to the "personal" filter test:
5118 (1) The test was buggy in that it was just doing the equivalent of
5119 "contains" tests on header lines. For example, if a user's address was
5120 anne@some.where, the "personal" test would incorrectly be true for
5122 To: susanne@some.where
5124 This test is now done by extracting each address from the header in turn,
5125 and checking the entire address. Other tests that are part of "personal"
5126 are now done using regular expressions (for example, to check local parts
5127 of addresses in From: header lines).
5129 (2) The list of non-personal local parts in From: addresses has been
5130 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5131 taken from the Sieve specification recommendations.
5133 (3) If the message contains any header line starting with "List-" it is
5134 treated as non-personal.
5136 (4) The test for "circular" in the Subject: header line has been removed
5137 because it now seems ill-conceived.
5139 15. Minor typos in src/EDITME comments corrected.
5141 16. Installed latest exipick from John Jetmore.
5143 17. If headers_add on a router specified a text string that was too long for
5144 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5145 of string_sprintf() is now avoided.
5147 18. $message_body_size was not set (it was always zero) when running the DATA
5148 ACL and the local_scan() function.
5150 19. For the "mail" command in an Exim filter, no default was being set for
5151 the once_repeat time, causing a random time value to be used if "once" was
5152 specified. (If the value happened to be <= 0, no repeat happened.) The
5153 default is now 0s, meaning "never repeat". The "vacation" command was OK
5154 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5155 (I found it when inspecting the code).
5157 20. There is now an overall timeout for performing a callout verification. It
5158 defaults to 4 times the callout timeout, which applies to individual SMTP
5159 commands during the callout. The overall timeout applies when there is more
5160 than one host that can be tried. The timeout is checked before trying the
5161 next host. This prevents very long delays if there are a large number of
5162 hosts and all are timing out (e.g. when the network connections are timing
5163 out). The value of the overall timeout can be changed by specifying an
5164 additional sub-option for "callout", called "maxwait". For example:
5166 verify = sender/callout=5s,maxwait=20s
5168 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5169 to the end before writing, but this should make it even safer).
5171 22. Exim was forgetting that it had advertised PIPELINING for the second and
5172 subsequent messages on an SMTP connection. It was also not resetting its
5173 memory on STARTTLS and an internal HELO.
5175 23. When Exim logs an SMTP synchronization error within a session, it now
5176 records whether PIPELINING has been advertised or not.
5178 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5179 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5180 rather than long int.
5182 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5184 26. Added the never_mail option to autoreply.
5190 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5191 crash if the getsockname() call failed; this can happen if a connection is
5192 closed very soon after it is established. The problem was simply in the
5193 order in which certain operations were done, causing Exim to try to write
5194 to the SMTP stream before it had set up the file descriptor. The bug has
5195 been fixed by making things happen in the correct order.
5201 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5202 before the connection was closed, thus losing the rejection response.
5204 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5205 some early Solaris releases, but causes trouble in current releases where
5206 socklen_t is defined.
5208 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5211 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5214 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5215 that releasing the top of it at the end releases what was used for sub-
5216 expansions (unless the block got too big). However, discard this block if
5217 the first thing is a variable or header, so that we can use its block when
5218 it is dynamic (useful for very large $message_headers, for example).
5220 6. Lookups now cache *every* query, not just the most recent. A new, separate
5221 store pool is used for this. It can be recovered when all lookup caches are
5222 flushed. Lookups now release memory at the end of their result strings.
5223 This has involved some general refactoring of the lookup sources.
5225 7. Some code has been added to the store_xxx() functions to reduce the amount
5226 of flapping under certain conditions.
5228 8. log_incoming_interface used to affect only the <= reception log lines. Now
5229 it causes the local interface and port to be added to several more SMTP log
5230 lines, for example "SMTP connection from", and rejection lines.
5232 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5234 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5236 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5237 could be overwritten at the end of the current message (or the start of a
5238 new message if it was set in a HELO ACL). The value is now preserved for
5239 the duration of the SMTP connection.
5241 12. If a transport had a headers_rewrite setting, and a matching header line
5242 contained an unqualified address, that address was qualified, even if it
5243 did not match any rewriting rules. The underlying bug was that the values
5244 of the flags that permit the existence of unqualified sender and recipient
5245 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5246 non-local messages, and by -bnq for local messages) were not being
5247 preserved with the message after it was received.
5249 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5250 "next input=" as part of the text comprising the host identity instead of
5251 the correct text. The code was using the same buffer for two different
5252 strings. However, depending on which order the printing function evaluated
5253 its arguments, the bug did not always show up. Under Linux, for example, my
5254 test suite worked just fine.
5256 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5257 doesn't work with some older versions of Perl. It has been changed to "my",
5258 which in any case is probably the better facility to use.
5260 15. A really picky compiler found some instances of statements for creating
5261 error messages that either had too many or two few arguments for the format
5264 16. The size of the buffer for calls to the DNS resolver has been increased
5265 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5266 for addresses that have a lot of PTR records. This alleviates a problem; it
5267 does not fully solve it.
5269 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5270 buffer now truncates the list and logs the incident, which is the same
5271 action as happens when Exim is looking up a host name and its aliases.
5272 Previously in this situation something unpredictable would happen;
5273 sometimes it was "internal error: store_reset failed".
5275 18. If a server dropped the connection unexpectedly when an Exim client was
5276 using GnuTLS and trying to read a response, the client delivery process
5277 crashed while trying to generate an error log message.
5279 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5280 string, for example:
5282 warn message = H1: something\nH2: something
5284 the text was added as a single header line from Exim's point of view
5285 though it ended up OK in the delivered message. However, searching for the
5286 second and subsequent header lines using $h_h2: did not work. This has been
5287 fixed. Similarly, if a system filter added multiple headers in this way,
5288 the routers could not see them.
5290 20. Expanded the error message when iplsearch is called with an invalid key to
5291 suggest using net-iplsearch in a host list.
5293 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5294 ACLs are no longer actually imposed (and a message to that effect is
5297 22. If a "gecos" field in a passwd entry contained escaped characters, in
5298 particular, if it contained a \" sequence, Exim got it wrong when building
5299 a From: or a Sender: header from that name. A second bug also caused
5300 incorrect handling when an unquoted " was present following a character
5301 that needed quoting.
5303 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5304 was not being matched caselessly.
5306 24. Arranged for all hyphens in the exim.8 source to be escaped with
5309 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5310 properly. Recipient callout cache records were still being keyed to include
5311 the sender, even when use_sender was set false. This led to far more
5312 callouts that were necessary. The sender is no longer included in the key
5313 when use_sender is false.
5315 26. Added "control = submission" modifier to ACLs.
5317 27. Added the ${base62d: operator to decode base 62 numbers.
5319 28. dnsdb lookups can now access SRV records.
5321 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5322 the configuration file.
5324 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5325 (-v) mode. This makes the output for a verbose queue run more intelligible.
5327 31. Added a use_postmaster feature to recipient callouts.
5329 32. Added the $body_zerocount variable, containing the number of binary zero
5330 bytes in the message body.
5332 33. The time of last modification of the "new" subdirectory is now used as the
5333 "mailbox time last read" when there is a quota error for a maildir
5336 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5338 35. Added +ignore_unknown as a special item in host lists.
5340 36. Code for decoding IPv6 addresses in host lists is now included, even if
5341 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5342 address was recognized as an IP address, but was then not correctly decoded
5343 into binary, causing unexpected and incorrect effects when compared with
5350 1. Very minor rewording of debugging text in manualroute to say "list of
5351 hosts" instead of "hostlist".
5353 2. If verify=header_syntax was set, and a header line with an unqualified
5354 address (no domain) and a large number of spaces between the end of the
5355 name and the colon was received, the reception process suffered a buffer
5356 overflow, and (when I tested it) crashed. This was caused by some obsolete
5357 code that should have been removed. The fix is to remove it!
5359 3. When running in the test harness, delay a bit after writing a bounce
5360 message to get a bit more predictability in the log output.
5362 4. Added a call to search_tidyup() just before forking a reception process. In
5363 theory, someone could use a lookup in the expansion of smtp_accept_max_
5364 per_host which, without the tidyup, could leave open a database connection.
5366 5. Added the variables $recipient_data and $sender_data which get set from a
5367 lookup success in an ACL "recipients" or "senders" condition, or a router
5368 "senders" option, similar to $domain_data and $local_part_data.
5370 6. Moved the writing of debug_print from before to after the "senders" test
5373 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5374 problems for message scanning, either using a data ACL, or using
5375 local_scan() because the Received: header was not generated till after they
5376 were called (in order to set the time as the time of reception completion).
5377 I have revised the way this works. The header is now generated after the
5378 body is received, but before the ACL or local_scan() are called. After they
5379 are run, the timestamp in the header is updated.
5385 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5386 before starting a queue runner without re-exec. This happened only when
5387 deliver_drop_privilege was set or when the Exim user was set to root. The
5388 effect of the bug was that timeouts during subsequent deliveries caused
5389 crashes instead of being properly handled. The handler is now left at its
5390 default (and expected) setting.
5392 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5393 message, again when deliver_drop_privilege is set or Exim is run as root.
5394 The bug described in (1) was not present in this case, but the tidying up
5395 of the other signals was missing. I have made the two cases consistent.
5397 3. The ignore_target_hosts setting on a manualroute router was being ignored
5398 for hosts that were looked up using the /MX notation.
5400 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5403 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5404 operated on the sender address. After changing the $sender_address to <>
5405 for the sender address verify, Exim was re-instated it as the original
5406 (before rewriting) address, but remembering that it had rewritten it, so it
5407 wasn't rewriting it again. This bug also had the effect of breaking the
5408 sender address verification caching when the sender address was rewritten.
5410 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5411 This has been changed so that if the ip literal address matches
5412 ignore_target_hosts, the router declines.
5414 7. Added expansion conditions match_domain, match_address, and match_local_
5415 part (NOT match_host).
5417 8. The placeholder for the Received: header didn't have a length field set.
5419 9. Added code to Exim itself and to exim_lock to test for a specific race
5420 condition that could lead to file corruption when using MBX delivery. The
5421 issue is with the lockfile that is created in /tmp. If this file is removed
5422 after a process has opened it but before that process has acquired a lock,
5423 there is the potential for a second process to recreate the file and also
5424 acquire a lock. This could lead to two Exim processes writing to the file
5425 at the same time. The added code performs the same test as UW imapd; it
5426 checks after acquiring the lock that its file descriptor still refers to
5427 the same named file.
5429 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5430 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5433 11. Added the smtp_active_hostname option. If used, this will typically be made
5434 to depend on the incoming interface address. Because $interface_address is
5435 not set up until the daemon has forked a reception process, error responses
5436 that can happen earlier (such as "too many connections") no longer contain
5439 12. If an expansion in a condition on a "warn" statement fails because a lookup
5440 defers, the "warn" statement is abandoned, and the next ACL statement is
5441 processed. Previously this caused the whole ACL to be aborted.
5443 13. Added the iplsearch lookup type.
5445 14. Added ident_timeout as a log selector.
5447 15. Added tls_certificate_verified as a log selector.
5449 16. Added a global option tls_require_ciphers (compare the smtp transport
5450 option of the same name). This controls incoming TLS connections.
5452 17. I finally figured out how to make tls_require_ciphers do a similar thing
5453 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5454 before starting the TLS session.
5456 18. Tabs are now shown as \t in -bP output.
5458 19. If the log selector return_path_on_delivery was set, Exim crashed when
5459 bouncing a message because it had too many Received: header lines.
5461 20. If two routers both had headers_remove settings, and the first one included
5462 a superfluous trailing colon, the final name in the first list and the
5463 first name in the second list were incorrectly joined into one item (with a
5464 colon in the middle).
5470 1. Added -C and -D options to the exinext utility, mainly to make it easier
5471 to include in the automated testing, but these could be helpful when
5472 multiple configurations are in use.
5474 2. The exinext utility was not formatting the output nicely when there was
5475 an alternate port involved in the retry record key, nor when there was a
5476 message id as well (for retries that were specific to a specific message
5477 and a specific host). It was also confused by IPv6 addresses, because of
5478 the additional colons they contain. I have fixed the IPv4 problem, and
5479 patched it up to do a reasonable job for IPv6.
5481 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5482 delivery, the log line now contains "pipelined" if PIPELINING was used.
5484 4. An SMTP transport process used to panic and die if the bind() call to set
5485 an explicit outgoing interface failed. This has been changed; it is now
5486 treated in the same way as a connect() failure.
5488 5. A reference to $sender_host_name in the part of a conditional expansion
5489 that was being skipped was still causing a DNS lookup. This no longer
5492 6. The def: expansion condition was not recognizing references to header lines
5493 that used bh_ and bheader_.
5495 7. Added the _cache feature to named lists.
5497 8. The code for checking quota_filecount in the appendfile transport was
5498 allowing one more file than it should have been.
5500 9. For compatibility with Sendmail, the command line option
5508 and sets the incoming protocol and host name (for trusted callers). The
5509 host name and its colon can be omitted when only the protocol is to be set.
5510 Note the Exim already has two private options, -pd and -ps, that refer to
5511 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5512 "s", but I don't think that's a major issue.
5514 10. A number of refactoring changes to the code, none of which should affect
5517 (a) The number of logging options was getting close to filling up the
5518 32-bit word that was used as a bit map. I have split them into two classes:
5519 those that are passed in the argument to log_write(), and those that are
5520 only ever tested independently outside of that function. These are now in
5521 separate 32-bit words, so there is plenty of room for expansion again.
5522 There is no change in the user interface or the logging behaviour.
5524 (b) When building, for example, log lines, the code previously used a
5525 macro that called string_cat() twice, in order to add two strings. This is
5526 not really sufficiently general. Furthermore, there was one instance where
5527 it was actually wrong because one of the argument was used twice, and in
5528 one call a function was used. (As it happened, calling the function twice
5529 did not affect the overall behaviour.) The macro has been replaced by a
5530 function that can join an arbitrary number of extra strings onto a growing
5533 (c) The code for expansion conditions now uses a table and a binary chop
5534 instead of a serial search (which was left over from when there were very
5535 few conditions). Also, it now recognizes conditions like "pam" even when
5536 the relevant support is not compiled in: a suitably worded error message is
5537 given if an attempt is made to use such a condition.
5539 11. Added ${time_interval:xxxxx}.
5541 12. A bug was causing one of the ddress fields not to be passed back correctly
5542 from remote delivery subprocesses. The field in question was not being
5543 subsequently used, so this caused to problems in practice.
5545 13. Added new log selectors queue_time and deliver_time.
5547 14. Might have fixed a bug in maildirsizefile handling that threw up
5548 "unexpected character" debug warnings, and recalculated the data
5549 unnecessarily. In any case, I expanded the warning message to give more
5552 15. Added the message "Restricted characters in address" to the statements in
5553 the default ACL that block characters like @ and % in local parts.
5555 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5556 Three changes have been made:
5558 (a) There was a serious bug; a negative response to MAIL caused the whole
5559 recipient domain to be cached as invalid, thereby blocking all messages
5560 to all local parts at the same domain, from all senders. This bug has
5561 been fixed. The domain is no longer cached after a negative response to
5562 MAIL if the sender used is not empty.
5564 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5567 (c) A new callout option, "use_sender" has been added for people who want
5568 the modified behaviour.
5574 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5577 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5578 indeed breaks things for older releases.
5580 3. Added additional logging to the case where there is a problem reading data
5581 from a filter that is running in a subprocess using a pipe, in order to
5582 try to track down a specific problem.
5584 4. Testing facility fudge: when running in the test harness and attempting
5585 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5586 getting "No route to host". Convert this to a timeout.
5588 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5591 6. Some OS don't have socklen_t but use size_t instead. This affects the
5592 fifth argument of getsockopt() amongst other things. This is now
5593 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5594 can be set for individual OS. I have set it for SunOS5, OSF1, and
5595 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5596 some earlier ones do not.
5598 7. Change 4.30/15 was not doing the test caselessly.
5600 8. The standard form for an IPv6 address literal was being rejected by address
5601 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5602 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5603 this, as well as the form without the "IPv6" on the front (but only when
5604 address literals are enabled, of course).
5606 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5608 10. Exim crashed if a message with an empty sender address specified by -f
5609 encountered a router with an errors_to setting. This could be provoked only
5610 by a command such as
5614 where an empty string was supplied; "<>" did not hit this bug.
5616 11. Installed PCRE release 4.5.
5618 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5619 remained set. It is now erased.
5621 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5622 times from message ids (which are base 36 rather than the normal 62).
5624 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5625 were being counted as actual protocol errors, and logged if the log
5626 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5627 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5628 and DATA following a set of rejected RCPTs do not count as protocol errors.
5629 In other words, Exim assumes they were pipelined, though this may not
5630 actually be the case. Of course, in all cases the client gets an
5631 appropriate error code.
5633 15. If a lookup fails in an ACL condition, a message about the failure may
5634 be available; it is used if testing the ACL cannot continue, because most
5635 such messages specify what the cause of the deferral is. However, some
5636 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5637 that caused an old message to be retained and used if a later statement
5638 caused a defer, replacing the real cause of the deferral.
5640 16. If an IP address had so many PTR records that the DNS lookup buffer
5641 was not large enough to hold them, Exim could crash while trying to process
5642 the truncated data. It now detects and logs this case.
5644 17. Further to 4.21/58, another change has been made: if (and only if) the
5645 first line of a message (the first header line) ends with CRLF, a bare LF
5646 in a subsequent header line has a space inserted after it, so as not to
5647 terminate the header.
5649 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5650 unnecessarily, used atoi() instead of strtol(), and didn't check the
5651 termination when getting file sizes from file names by regex.
5653 19. Completely re-implemented the support for maildirsize files, in the light
5654 of a number of problems with the previous contributed implementation
5655 (4.30/29). In particular:
5657 . If the quota is zero, the maildirsize file is maintained, but no quota is
5660 . If the maildir directory does not exist, it is created before any attempt
5661 to write a maildirsize file.
5663 . The quota value in the file is just a cache; if the quota is changed in
5664 the transport, the new value overrides.
5666 . A regular expression is available for excluding directories from the
5669 20. The autoreply transport checks the characters in options that define the
5670 message's headers; it allows continued headers, but it was checking with
5671 isspace() after an embedded newline instead of explicitly looking for a
5674 21. If all the "regular" hosts to which an address was routed had passed their
5675 expiry times, and had not reached their retry times, the address was
5676 bounced, even if fallback hosts were defined. Now Exim should go on to try
5679 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5680 equivalent code in the SMTP transport. Some hosts send humungous responses
5681 to HELO/EHLO, more than 1024 it seems.
5683 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5684 gives compiler warnings in some environments. I've now done it "properly",
5687 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5688 (because of problems with the built-in one) was declared to return uschar *
5689 instead of char *, causing compiler failure.
5691 25. Fixed a file descriptor leak when processing alias/forward files.
5693 26. Fixed a minor format string issue in dbfn.c.
5695 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5697 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5698 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5701 29. When a sender address is verified, it is cached, to save repeating the test
5702 when there is more than one recipient in a message. However, when the
5703 verification involves a callout, it is possible for different callout
5704 options to be set for different recipients. It is too complicated to keep
5705 track of this in the cache, so now Exim always runs a verification when a
5706 callout is required, relying on the callout cache for the optimization.
5707 The overhead is duplication of the address routing, but this should not be
5710 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5711 to be verified with callout=postmaster, and the main callout worked but the
5712 postmaster check failed, the verification correctly failed. However, if a
5713 subsequent RCPT command asked for sender verification *without* the
5714 postmaster check, incorrect caching caused this verification also to fail,
5717 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5718 it was not caching the DNS options (qualify_single, search_parents) that
5719 were used when the lookup failed. A subsequent lookup with different
5720 options therefore always gave the same answer, though there were cases
5721 where it should not have. (Example: a "domains = !$mx_any" option on a
5722 dnslookup router: the "domains" option is always processed without any
5723 widening, but the router might have qualify_single set.) Now Exim uses the
5724 cached value only when the same options are set.
5726 32. Added John Jetmore's "exipick" utility to the distribution.
5728 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5729 than a timeout (e.g. a certificate is required, and is not provided), an
5730 Exim server now closes the connection immediately. Previously it waited for
5731 the client to close - but if the client is SSL, it seems that they each
5732 wait for each other, leading to a delay before one of them times out.
5734 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5735 maintained 0.8.x compatibility because I don't think many are using it, and
5736 it is clearly obsolete.
5738 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5741 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5742 client certificate was expired. A simple patch fixes this, though I don't
5743 understand the full logic of why the verify callback is called multiple
5746 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5747 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5748 which causes problems with some clients (such as the Certicom SSL Plus
5749 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5750 disables the coutermeasure allowing Eudora to connect."
5752 38. Exim was not checking that a write() to a log file succeeded. This could
5753 lead to Bad Things if a log got too big, in particular if it hit a file
5754 size limit. Exim now panics and dies if it cannot write to a log file, just
5755 as it does if it cannot open a log file.
5757 39. Modified OS/Makefile-Linux so that it now contains
5759 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5761 The two -D definitions ensure that Exim is compiled with large file
5762 support, which makes it possible to handle log files that are bigger than
5765 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5766 instance) a domain was checked against a named list that involved a lookup,
5767 causing $domain_data to be set, then another domain was checked against the
5768 same list, then the first domain was re-checked, the value of $domain_data
5769 after the final check could be wrong. In particular, if the second check
5770 failed, it could be set empty. This bug probably also applied to
5773 41. The strip_trailing_dot option was not being applied to the address given
5774 with the -f command-line option.
5776 42. The code for reading a message's header from the spool was incrementing
5777 $received_count, but never initializing it. This meant that the value was
5778 incorrect (doubled) while delivering a message in the same process in which
5779 it was received. In the most common configuration of Exim, this never
5780 happens - a fresh exec is done - but it can happen when
5781 deliver_drop_privilege is set.
5783 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5784 it now includes up to 150 characters of the unexpected data in the log
5787 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5788 and building data strings. The size of both of these buffers was 10 000
5789 bytes - far larger than anybody would *ever* want, thought I. Needless to
5790 say, somebody hit the limit. I have increased the maximum line length to
5791 20 000 and the maximum data length of concatenated lines to 100 000. I have
5792 also fixed two bugs, because there was no checking on these buffers. Tsk,
5793 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5794 buffer is too small.
5796 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5797 lsearch lookups. Now it does.
5799 46. When parsing a route_list item in a manualroute router, a fixed-length
5800 buffer was used for the list of hosts. I made this 1024 bytes long,
5801 thinking that nobody would ever have a list of hosts that long. Wrong.
5802 Somebody had a whole pile of complicated expansion conditions, and the
5803 string was silently truncated, leading to an expansion error. It turns out
5804 that it is easier to change to an unlimited length (owing to other changes
5805 that have happened since this code was originally written) than to build
5806 structure for giving a limitation error. The length of the item that
5807 expands into the list of hosts is now unlimited.
5809 47. The lsearch lookup could not handle data where the length of text line was
5810 more than 4095 characters. Such lines were truncated, leading to shortened
5811 data being returned. It should now handle lines of any length.
5813 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5814 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5817 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5818 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5819 no longer happens in comment lines. A list of replacements is now placed
5820 at the head of all of the source files, except those whose only change is
5821 to replace PERL_COMMAND in the very first #! line.
5823 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5824 messages on the queue, with a bottom-up merge sort, using code contributed
5825 by Michael Haardt. This should make operations like -bp somewhat faster on
5826 large queues. It won't affect queue runners, except when queue_run_in_order
5829 51. Installed eximstats 1.31 in the distribution.
5831 52. Added support for SRV lookups to the dnslookup router.
5833 53. If an ACL referred to $message_body or $message_body_end, the value was not
5834 reset for any messages that followed in the same SMTP session.
5836 54. The store-handling optimization for building very long strings was not
5837 differentiating between the different store pools. I don't think this
5838 actually made any difference in practice, but I've tidied it.
5840 55. While running the routers to verify a sender address, $sender_address
5841 was still set to the sender address. This is wrong, because when routing to
5842 send a bounce to the sender, it would be empty. Therefore, I have changed
5843 it so that, while verifying a sender address, $sender_address is set to <>.
5844 (There is no change to what happens when verifying a recipient address.)
5846 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5847 target A or AAAA records (if not already returned) without resetting the
5848 qualify_single or search_parents options of the DNS resolver. These are
5849 inappropriate in this case because the targets of MX and SRV records must
5850 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5851 target that, when qualified, matched something in the local domain. These
5852 two options are now turned off when doing these lookups.
5854 57. It seems that at least some releases of Reiserfs (which does not have the
5855 concept of a fixed number of inodes) returns zero and not -1 for the
5856 number of available inodes. This interacted badly with check_spool_inodes,
5857 which assumed that -1 was the "no such thing" setting. What I have done is
5858 to check that the total number of inodes is greater than zero before doing
5859 the test of how many are available.
5861 58. When a "warn" ACL statement has a log_message modifier, the message is
5862 remembered, and not repeated. This is to avoid a lot of repetition when a
5863 message has many recipients that cause the same warning to be written.
5864 However, Exim was preserving the list of already written lines for an
5865 entire SMTP session, which doesn't seem right. The memory is now reset if a
5866 new message is started.
5868 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5869 debugging output unless log_rewrite was also set.
5871 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5872 of (int)(handle) when we know that handle contains (void *)(-1).
5874 61. The Exim daemon panic-logs an error return when it closes the incoming
5875 connection. However "connection reset by peer" seems to be common, and
5876 isn't really an error worthy of noting specially, so that particular error
5879 62. When Exim is trying to find all the local interfaces, it used to panic and
5880 die if the ioctl to get the interface flags failed. However, it seems that
5881 on at least one OS (Solaris 9) it is possible to have an interface that is
5882 included in the list of interfaces, but for which you get a failure error
5883 for this call. This happens when the interface is not "plumbed" into a
5884 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5885 failure of the "get flags" call assumes that the interface is down.
5887 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5888 makes life easier for people who are doing arithmetic on fields extracted
5889 from dates, where you often get leading zeros that should not be
5890 interpreted as octal.
5892 64. Added qualify_domain to the redirect router, to override the global
5895 65. If a pathologically long header line contained very many addresses (the
5896 report of this problem mentioned 10 000) and each of them was rewritten,
5897 Exim could use up a very large amount of memory. (It kept on making new
5898 copies of the header line as it rewrote, and never released the old ones.)
5899 At the expense of a bit more processing, the header rewriting function has
5900 been changed so that it no longer eats memory in this way.
5902 66. The generation of the Received: header has been moved from the time that a
5903 message starts to be received, to the time that it finishes. The timestamp
5904 in the Received: header should now be very close to that of the <= log
5905 line. There are two side-effects of this change:
5907 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5908 logged header lines no longer include the local Received: line, because
5909 it has not yet been created. The same applies to a copy of the message
5910 that is returned to a non-SMTP sender when a message is rejected.
5912 (b) When a filter file is tested using -bf, no additional Received: header
5913 is added to the test message. After some thought, I decided that this
5916 This change does not affect the value of $received_for. It is still set
5917 after address rewriting, but before local_scan() is called.
5919 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5921 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5922 gave an unhelpful panic error message, and a defer error. I have managed to
5923 change this behaviour so that it now rejects any supplied certificate,
5924 which seems right, as the list of acceptable certificates is empty.
5926 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5927 gave an unhelpful defer error. I have not managed to make this reject any
5928 supplied certificates, but the error message it gives is "no certificate
5929 supplied", which is not helpful.
5931 70. exigrep's output now also includes lines that are not associated with any
5932 message, but which match the given pattern. Implemented by a patch from
5933 Martin Sluka, which also tidied up the Perl a bit.
5935 71. Recipient callout verification, like sender verification, was using <> in
5936 the MAIL FROM command. This isn't really the right thing, since the actual
5937 sender may affect whether the remote host accepts the recipient or not. I
5938 have changed it to use the actual sender in the callout; this means that
5939 the cache record is now keyed on a recipient/sender pair, not just the
5940 recipient address. There doesn't seem to be a real danger of callout loops,
5941 since a callout by the remote host to check the sender would use <>.
5942 [SEE ABOVE: changed after hitting problems.]
5944 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5945 temporary errors. However, in the case of such a code being given after
5946 the end of a data transmission (i.e. after ".") Exim was failing to write
5947 a retry record for the message. (Yes, there was some broken host that was
5948 actually sending 8xx at this point.)
5950 73. An unknown lookup type in a host list could cause Exim to panic-die when
5951 the list was checked. (An example that provoked this was putting <; in the
5952 middle of a list instead of at the start.) If this happened during a DATA
5953 ACL check, a -D file could be left lying around. This kind of configuration
5954 error no longer causes Exim to die; instead it causes a defer error. The
5955 incident is still logged to the main and panic logs.
5957 74. Buglet left over from Exim 3 conversion. The message "too many messages
5958 in one connection" was written to the rejectlog but not the mainlog, except
5959 when address rewriting (yes!) was being logged.
5961 75. Added write_rejectlog option.
5963 76. When a system filter was run not as root (that is, when system_filter_user
5964 was set), the values of the $n variables were not being returned to the
5965 main process; thus, they were not subsequently available in the $sn
5968 77. Added +return_path_on_delivery log selector.
5970 78. A connection timeout was being treated differently from recipients deferred
5971 when testing hosts_max_try with a message that was older than the host's
5972 retry timeout. (The host should not be counted, thus allowing all hosts to
5973 be tried at least once before bouncing.) This may have been the cause of an
5974 occasionally reported bug whereby a message would remain on the queue
5975 longer than the retry timeout, but would be bounced if a delivery was
5976 forced. I say "may" because I never totally pinned down the problem;
5977 setting up timeout/retry tests is difficult. See also the next item.
5979 79. The ultimate address timeout was not being applied to errors that involved
5980 a combination of host plus message (for example, a timeout on a MAIL
5981 command). When an address resolved to a number of possible hosts, and they
5982 were not all tried for each delivery (e.g. because of hosts_max_try), a
5983 message could remain on the queue longer than the retry timeout.
5985 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5988 81. Fixed an obscure SMTP outgoing bug which required at least the following
5989 conditions: (a) there was another message waiting for the same server;
5990 (b) the server returned 5xx to all RCPT commands in the first message so
5991 that the message was not completed; (c) the server dropped the connection
5992 or gave a negative response to the RSET that Exim sends to abort the
5993 transaction. The observed case was a dropped connection after DATA that had
5994 been sent in pipelining mode. That is, the server had advertised PIPELINING
5995 but was not implementing it correctly. The effect of the bug was incorrect
5996 behaviour, such as trying another host, and this could lead to a crash.
6002 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6003 and daemon.c were passed as pointers to ints; they should have been
6004 pointers to socklen_t variables (which are typically unsigned ints).
6006 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6009 3. Fixed a really odd bug that affected only the testing scheme; patching a
6010 certain fixed string in the binary changed the value of another string that
6011 happened to be identical to the end of the original first string.
6013 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6014 name", it returns that address as the IP address. On some operating
6015 systems (e.g. Solaris), it also passes back the IP address string as the
6016 "host name". However, on others (e.g. Linux), it passes back an empty
6017 string. Exim wasn't checking for this, and was changing the host name to an
6018 empty string, assuming it had been canonicalized.
6020 5. Although rare, it is permitted to have more than one PTR record for a given
6021 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6022 all the names associated with an address, because they do in Solaris.
6023 However, it seems that they do not in Linux for data that comes from the
6024 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6025 I found this out when I moved to a new Linux workstation and tried to run
6026 the Exim test suite.
6028 To get round this problem I have changed the code so that it now does its
6029 own call to the DNS to look up PTR records when searching for a host name.
6030 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6031 addresses that are only in /etc/hosts are still found.
6033 This behaviour is, however, controlled by an option called host_lookup_
6034 order, which defaults to "bydns:byaddr". If people want to use the other
6035 order, or indeed, just use one or the other means of lookup, they can
6036 specify it in this variable.
6038 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6039 some operating systems, this comes back from gethostbyaddr() as an empty
6040 string, and this is what Exim used to test for. However, it seems that in
6041 other systems, "." is yielded. Exim now tests for this case too.
6043 7. The values of check_spool_space and check_log_space are now held internally
6044 as a number of kilobytes instead of an absolute number of bytes. If a
6045 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6046 kilobyte. This means that much larger values can be stored.
6048 8. Exim monitor: an attempt to get the action menu when not actually pointing
6049 at a message produces an empty menu entitled "No message selected". This
6050 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6051 no entries in it ("Shell widget menu has zero width and/or height"). So I
6052 have added a single, blank menu entry in this case.
6054 9. Added ${quote_local_part.
6056 10. MIME decoding is now applied to the contents of Subject: header lines when
6059 11. Now that a reference to $sender_host_address automatically causes a reverse
6060 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6061 host lookup before query-style lookups in lists that might use this
6062 variable. This has therefore been abolished, and the "net-" prefix is no
6063 longer necessary for query-style lookups.
6065 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6066 have been a typo for LFLAGS, so it has been changed.
6068 13. The install script calls Exim with "-C /dev/null" in order to find the
6069 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6070 to be output. However, since Exim outputs its version number before the
6071 error, it didn't break the script. It just looked ugly. I fixed this by
6072 always allowing "-C /dev/null" if the caller is root.
6074 14. Ignore overlarge ACL variable number when reading spool file - insurance
6075 against a later release with more variables having written the file.
6077 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6078 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6079 this, as well as the form without the "IPv6" on the front.
6081 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6082 OS/Makefile-Darwin file.
6084 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6085 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6087 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6088 variable. However, some OS use ints for inodes. I've added cast to long int
6089 to get rid of the compiler warning.
6091 19. I had forgotten to lock out "/../" in configuration file names when
6092 ALT_CONFIG_PREFIX was set.
6094 20. Routers used for verification do not need to specify transports. However,
6095 if such a router generated a host list, and callout was configured, Exim
6096 crashed, because it could not find a port number from the (non-existent)
6097 transport. It now assumes port 25 in this circumstance.
6099 21. Added the -t option to exigrep.
6101 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6102 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6103 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6104 them is set, LOOKUP_LSEARCH is forced.
6106 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6108 24. Added sender and host information to the "rejected by local_scan()" log
6109 line; previously there was no indication of these.
6111 25. Added .include_if_exists.
6113 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6114 Linux. It turns out that not all file systems support this. Apparently some
6115 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6116 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6117 not supported on the file descriptor, is now ignored when Exim is trying to
6118 sync a directory. This applies only to Linux.
6120 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6122 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6123 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6126 29. Added support for maildirsize files from supplied patch (modified a bit).
6128 30. The use of :fail: followed by an empty string could lead Exim to respond to
6129 sender verification failures with (e.g.):
6131 550 Verification failed for <xxx>
6132 550 Sender verify failed
6134 where the first response line was missing the '-' that indicates it is not
6135 the final line of the response.
6137 31. The loop for finding the name of the user that called Exim had a hardwired
6138 limit of 10; it now uses the value of finduser_retries, which is used for
6139 all other user lookups.
6141 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6144 33. Exim was neglecting to zero errno before one call of strtol() when
6145 expanding a string and expecting an integer value. On some systems this
6146 resulted in spurious "integer overflow" errors. Also, it was casting the
6147 result into an int without checking.
6149 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6150 did not work. The code looks as if it has *never* worked, though it appears
6151 to have been documented since at least release 1.62. I have made it work.
6153 35. The "timeout_DNS" error in retry rules, also documented since at least
6154 1.62, also never worked. As it isn't clear exactly what this means, and
6155 clearly it isn't a major issue, I have abolished the feature by treating it
6156 as "timeout", and writing a warning to the main and panic logs.
6158 36. The display of retry rules for -brt wasn't always showing the error code
6161 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6162 timeout_connect_A, timeout_connect_MX.
6164 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6165 to the empty sender.
6167 39. The daemon was not analysing the content of -oX till after it had closed
6168 stderr and disconnected from the controlling terminal. This meant that any
6169 syntax errors were only noted on the panic log, and the return code from
6170 the command was 0. By re-arranging the code a little, I've made the
6171 decoding happen first, so such errors now appear on stderr, and the return
6172 code is 1. However, the actual setting up of the sockets still happens in
6173 the disconnected process, so errors there are still only recorded on the
6176 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6177 connections (as happens on some IP stacks) was logged at start up time as
6178 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6179 it from "IPv6 and IPv4", which means that two separate sockets are being
6182 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6183 says whether AF_INET or AF_INET6 was passed as an argument.
6185 42. Exiwhat output was messed up when time zones were included in log
6188 43. Exiwhat now gives more information about the daemon's listening ports,
6189 and whether -tls-on-connect was used.
6191 44. The "port" option of the smtp transport is now expanded.
6193 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6194 silently ignored. Now an error message is written to the main and panic
6197 46. There's a new ACL modifier called "logwrite" which writes to a log file
6198 as soon as it is encountered.
6200 47. Added $local_user_uid and $local_user_gid at routing time.
6202 48. Exim crashed when trying to verify a sender address that was being
6205 49. Exim was recognizing only a space character after ".include". It now also
6206 recognizes a tab character.
6208 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6209 extracting the relevant information from the specification. The man page no
6210 longer contains scrambled data for the -d option, and I've added a section
6211 at the front about calling Exim under different names.
6213 51. Added "extra_headers" argument to the "mail" command in filter files.
6215 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6218 53. Installed eximstats 1.29.
6220 54. Added transport_filter_timeout as a generic transport option.
6222 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6223 Cc: header lines. This was required by RFC 822, but it not required by RFC
6226 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6227 incoming messages that did not have them. Now it does so only if the
6228 message originates locally, that is, if there is no associated remote host
6229 address. When Resent- header lines are present, this applies to the Resent-
6230 lines rather than the non-Resent- lines.
6232 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6233 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6235 58. Messages for configuration errors now include the name of the main
6236 configuration file - useful now that there may be more than one file in a
6237 list (.included file names were always shown).
6239 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6240 for those rare installations that do not start the daemon as root or run it
6241 setuid root. I've cut out the call to initgroups() if the daemon is not
6244 60. The Exim user and group can now be bound into the binary as text strings
6245 that are looked up at the start of Exim's processing.
6247 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6249 62. Added $mailstore_basename variable.
6251 63. Installed patch to sieve.c from Michael Haardt.
6253 64. When Exim failed to open the panic log after failing to open the main log,
6254 the original message it was trying to log was written to stderr and debug
6255 output, but if they were not available (the usual case in production), it
6256 was lost. Now it is written to syslog before the two lines that record the
6257 failures to open the logs.
6259 65. Users' Exim filters run in subprocesses under the user's uid. It is
6260 possible for a "deliver" command or an alias in a "personal" command to
6261 provoke an address rewrite. If logging of address rewriting is configured,
6262 this fails because the process is not running as root or exim. There may be
6263 a better way of dealing with this, but for the moment (because 4.30 needs
6264 to be released), I have disabled address rewrite logging when running a
6265 filter in a non-root, non-exim process.
6271 1. The buildconfig auxiliary program wasn't quoting the value set for
6272 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6273 not defined. This bug was masked in 4.22 by the effect that was fixed in
6276 2. Some messages that were rejected after a message id was allocated were
6277 shown as "incomplete" by exigrep. It no longer does this for messages that
6278 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6280 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6281 have allow_domain_literals set, the ID did not get logged in the <= line.
6282 Domain literals are now always recognized in Message-ID: header lines.
6284 4. The first argument for a ${extract expansion item is the key name or field
6285 number. Leading and trailing spaces in this item were not being ignored,
6286 causing some misleading effects.
6288 5. When deliver_drop_privilege was set, single queue runner processes started
6289 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6290 same command in the process it spins off) were not dropping privilege.
6292 6. When the daemon running as "exim" started a queue runner, it always
6293 re-executed Exim in the spun-off process. This is a waste of effort when
6294 deliver_drop_privilege is set. The new process now just calls the
6295 queue-runner function directly.
6301 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6304 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6305 ignored. Though the use of -oP was forcing the writing of a pid file, it
6306 was always written to the default place.
6308 3. If the message "no IP address found for host xxxx" is generated during
6309 incoming verification, it is now followed by identification of the incoming
6310 connection (so you can more easily find what provoked it).
6312 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6314 5. Added some features to "harden" Exim a bit more against certain attacks:
6316 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6317 be put in Local/Makefile. This is like the never_users runtime option,
6318 but it cannot be overridden. The default setting is "root".
6320 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6321 prefix string with which any file named in a -C command line option
6324 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6325 is retained for -C and -D only if the caller of Exim is root. Without
6326 it, the exim user may also use -C and -D and retain privilege.
6328 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6329 command line option is disabled.
6331 6. Macro names set by the -D option must start with an upper case letter, just
6332 like macro names defined in the configuration file.
6334 7. Added "dereference=" facility to LDAP.
6336 8. Two instances of the typo "uknown" in the source files are fixed.
6338 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6339 the Configure-Makefile script screwed up while processing it.
6341 10. Incorporated PCRE 4.4.
6343 11. The SMTP synchronization check was not operating right at the start of an
6344 SMTP session. For example, it could not catch a HELO sent before the client
6345 waited for the greeting. There is now a check for outstanding input at the
6346 point when the greeting is written. Because of the duplex, asynchronous
6347 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6348 way, but not yet received, when the check is performed.
6350 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6351 on TCP/IP sockets, because this apparently causes some broken clients to
6354 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6355 unchanged) from the Cygwin maintainer.
6357 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6358 was supported instead of testing for mbx. Effectively a typo.
6360 15. The spa authenticator server code was not checking that the input it
6361 received was valid base64.
6363 16. The debug output line for the "set" modifier in ACLs was not showing the
6364 name of the variable that was being set.
6366 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6368 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6369 lookup on its own. Something else was needed to trigger the lookup. For
6370 example, a match in host_lookup or the need for a host name in a host list.
6371 Now, if $sender_host_name is referenced and the host name has not yet been
6372 looked up, a lookup is performed. If the lookup fails, the variable remains
6373 empty, and $host_lookup_failed is set to "1".
6375 19. Added "eqi" as a case-independent comparison operator.
6377 20. The saslauthd authentication condition could segfault if neither service
6378 nor realm was specified.
6380 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6381 error message that was logged was misleading, and incoming SMTP
6382 connections were dropped. The message is now more accurate, and temporary
6383 errors are given to SMTP connections.
6385 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6386 (except RSET) with a 421 error, until QUIT is received. However, it was
6387 failing to send a response to QUIT.
6389 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6390 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6391 ACL is now run after the helo_try_verify_hosts code.
6393 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6394 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6395 case-independent, so other case variants are also recognized). Apparently
6396 some systems use these upper case variants.
6398 25. If more than two messages were waiting for the same host, and a transport
6399 filter was specified for the transport, Exim sent two messages over the
6400 same TCP/IP connection, and then failed with "socket operation on non-
6401 socket" when it tried to send the third.
6403 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6405 27. The extern definition of crypt16() in expand.c was not being excluded when
6406 the OS had its own crypt16() function.
6408 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6409 as a preferred synonym for return_size_limit, both as an option and as an
6412 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6414 30. Changed the default configuration ACL to relax the local part checking rule
6415 for addresses that are not in any local domains. For these addresses,
6416 slashes and pipe symbols are allowed within local parts, but the sequence
6417 /../ is explicitly forbidden.
6419 31. SPA server authentication was not clearing the challenge buffer before
6422 32. log_message in a "warn" ACL statement was writing to the reject log as
6423 well as to the main log, which contradicts the documentation and doesn't
6424 seem right (because no rejection is happening). So I have stopped it.
6426 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6427 However, I am unable to do any testing of this.
6429 34. Fixed an infelicity in the appendfile transport. When checking directories
6430 for a mailbox, to see if any needed to be created, it was accidentally
6431 using path names with one or more superfluous leading slashes; tracing
6432 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6434 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6435 contents are added to the log line that is written for every discarded
6436 recipient. (Previously a log_message setting was ignored.)
6438 36. The ${quote: operator now quotes the string if it is empty.
6440 37. The install script runs exim in order to find its version number. If for
6441 some reason other than non-existence or emptiness, which it checks, it
6442 could not run './exim', it was installing it with an empty version number,
6443 i.e. as "exim-". This error state is now caught, and the installation is
6446 38. An argument was missing from the function that creates an error message
6447 when Exim fails to connect to the socket for saslauthd authentication.
6448 This could cause Exim to crash, or give a corrupted message.
6450 39. Added isip, isip4, and isip6 to ${if conditions.
6452 40. The ACL variables $acl_xx are now saved with the message, and can be
6453 accessed later in routers, transports, and filters.
6455 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6456 strings in the file are not string-expanded.
6458 42. If a MAIL command specified a SIZE value that was too large to fit into an
6459 int variable, the check against message_size_limit failed. Such values are
6460 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6461 day this will have to be increased, but I don't think I want to be around
6462 when emails are that large.
6469 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6470 iconv() is not standard in FreeBSD.
6472 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6473 IPv6 enabled. The observed symptom was a segmentation fault on return from
6474 the function os_common_find_running_interfaces() in src/os.c.
6476 3. In the check_special_case() function in daemon.c I had used "errno" as an
6477 argument name, which causes warnings on some systems. This was basically a
6478 typo, since it was named "eno" in the comments!
6480 4. The code that waits for the clock to tick (at a resolution of some fraction
6481 of a second) so as to ensure message-id uniqueness was always waiting for
6482 at least one whole tick, when it could have waited for less. [This is
6483 almost certainly not relevant at current processor speeds, where it is
6484 unlikely to ever wait at all. But we try to future-proof.]
6486 5. The function that sleeps for a time interval that includes fractions of a
6487 second contained a race. It did not block SIGALRM between setting the
6488 timer, and suspending (a couple of lines later). If the interval was short
6489 and the sigsuspend() was delayed until after it had expired, the suspension
6490 never ended. On busy systems this could lead to processes getting stuck for
6493 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6494 process, before it forks any delivery processes. The open lookup caching
6495 mechanism meant that the open file or database connection was passed into
6496 the delivery process. The problem was that delivery processes always tidy
6497 up cached lookup data. This could cause a problem for the next delivery
6498 process started by the queue runner, because the external queue runner
6499 process does not know about the closure. So the next delivery process
6500 still has data in the lookup cache. In the case of a file lookup, there was
6501 no problem because closing a file descriptor in a subprocess doesn't affect
6502 the parent. However, if the lookup was caching a connection to a database,
6503 the connection was closed, and the second delivery process was likely to
6504 see errors such as "PGSQL: query failed: server closed the connection
6505 unexpectedly". The problem has been fixed by closing all cached lookups
6506 in a queue runner before running a delivery process.
6508 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6509 seem to have the "const" qualifier which it has on other OS. I've
6512 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6513 *of the same type* (client or server) with the same public name that an
6514 error should be diagnosed.
6516 9. When Exim looked up a host name for an IP address, but failed to find the
6517 original IP address when looking up the host name (a safety check), it
6518 output the message "<ip address> does not match any IP for NULL", which was
6519 confusing, to say the least. The bug was that the host name should have
6520 appeared instead of "NULL".
6522 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6523 user that is built into the binary, and the -C or -D options is used, root
6524 privilege is dropped before the configuration file is read. In addition,
6525 logging is switched to stderr instead of the normal log files. If the
6526 configuration then re-defines the Exim user, the unprivileged environment
6527 is probably not what is expected, so Exim logs a panic warning message (but
6530 However, if deliver_drop_privilege is set, the unprivileged state may well
6531 be exactly what is intended, so the warning has been cut out in that case,
6532 and Exim is allowed to try to write to its normal log files.
6538 1. smtp_return_error_details was not giving details for temporary sender
6539 or receiver verification errors.
6541 2. Diagnose a configuration error if two authenticators have the same public
6544 3. Exim used not to create the message log file for a message until the first
6545 delivery attempt. This could be confusing when incoming messages were held
6546 for policy or load reasons. The message log file is now created at the time
6547 the message is received, and an initial "Received" line is written to it.
6549 4. The automatically generated man page for command line options had a minor
6550 bug that caused no ill effects; however, a more serious problem was that
6551 the procedure for building the man page automatically didn't always
6552 operate. Consequently, release 4.20 contains an out-of-date version. This
6553 shouldn't happen again.
6555 5. When building Exim with embedded Perl support, the script that builds the
6556 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6557 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6559 6. The freeze_tell option was not being used for messages that were frozen on
6560 arrival, either by an ACL or by local_scan().
6562 7. Added the smtp_incomplete_transaction log selector.
6564 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6565 was accepting AUTH without a new EHLO.
6567 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6568 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6571 10. Logging of TCP/IP connections (when configured) now happens in the main
6572 daemon process instead of the child process, so that the TCP/IP connection
6573 count is more accurate (but it can never be perfect).
6575 11. The use of "drop" in a nested ACL was not being handled correctly in the
6576 outer ACL. Now, if condition failure induced by the nested "drop" causes
6577 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6578 or "require"), the connection is dropped.
6580 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6581 that yield "discard" can now be used with an "accept" or a "discard" verb,
6582 but an error is generated for any others (because I can't see a useful way
6583 to define what should happen).
6585 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6586 are now processed in the same way as lines in the Exim configuration file.
6587 In particular, continuation lines are supported.
6589 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6591 15. Added -ti meaning -t -i.
6593 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6594 domains, and warn by logging if others are found.
6596 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6597 returned from GIFCONF when Exim is trying to find the list of interfaces on
6598 a host. The code in os.c has been modified to copy each ifreq to an aligned
6599 structure in all cases.
6601 Also, in some cases, the returned ifreq's were being copied to a 'struct
6602 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6603 means the last couple of bytes of an IPv6 address could be chopped if the
6604 ifreq contained only a normal sockaddr (14 bytes storage).
6606 18. Named domain lists were not supported in the hosts_treat_as_local option.
6607 An entry such as +xxxx was not recognized, and was treated as a literal
6610 19. Ensure that header lines added by a DATA ACL are included in the reject log
6611 if the ACL subsequently rejects the message.
6613 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6614 MD5 (which is deprecated).
6616 21. When testing a filter file using -bf, Exim was writing a message when it
6617 took the sender from a "From " line in the message, but it was not doing so
6618 when it took $return_path from a Return-Path: header line. It now does.
6620 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6621 with a valid header line field name (a series of printing characters
6622 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6624 23. Changed "disc" in the source to "disk" to conform to the documentation and
6625 the book and for uniformity.
6627 24. Ignore Sendmail's -Ooption=value command line item.
6629 25. When execve() failed while trying to run a command in a pipe transport,
6630 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6631 could be confused with a return value of 69 from the command itself. This
6632 has been changed to 127, the value the shell returns if it is asked to run
6633 a non-existent command. The wording for the related log line suggests a
6634 non-existent command as the problem.
6636 26. If received_header_text expands to an empty string, do not add a Received:
6637 header line to the message. (Well, it adds a token one on the spool, but
6638 marks it "old" so that it doesn't get used or transmitted.)
6640 27. Installed eximstats 1.28 (addition of -nt option).
6642 28. There was no check for failure on the call to getsockname() in the daemon
6643 code. This can fail if there is a shortage of resources on the system, with
6644 ENOMEM, for example. A temporary error is now given on failure.
6646 29. Contrary to the C standard, it seems that in some environments, the
6647 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6648 program. Exim now does this explicitly; it affects the formatting of
6649 timestamps using strftime().
6651 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6652 complaints. I've now initialized all the variables, to avoid this.
6654 32. Header lines added by a system filter were not being "seen" during
6655 transport-time rewrites.
6657 33. The info_callback() function passed to OpenSSL is set up with type void
6658 (*)(SSL *, int, int), as described somewhere. However, when calling the
6659 function (actually a macro) that sets it up, the type void(*)() is
6660 expected. I've put in a cast to prevent warnings from picky compilers.
6662 34. If a DNS black list lookup found a CNAME record, but there were no A
6663 records associated with the domain it pointed at, Exim crashed.
6665 35. If a DNS black list lookup returned more than one A record, Exim ignored
6666 all but the first. It now scans all returned addresses if a particular IP
6667 value is being sought. In this situation, the contents of the
6668 $dnslist_value variable are a list of all the addresses, separated by a
6671 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6672 to accept a host name and all its aliases if the forward lookup for any of
6673 them yielded the IP address of the incoming connection. Now it accepts only
6674 those names whose forward lookup yields the correct IP address. Any other
6675 names are discarded. This closes a loophole whereby a rogue DNS
6676 administrator could create reverse DNS records to break through a
6677 wildcarded host restriction in an ACL.
6679 37. If a user filter or a system filter that ran in a subprocess used any of
6680 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6681 the wrong values were passed to the pipe command ($thisaddress had the
6682 value of $0, $0 had the value of $1, etc). This bug was introduced by
6683 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6685 38. Improved the line breaking for long SMTP error messages from ACLs.
6686 Previously, if there was no break point between 40 and 75 characters, Exim
6687 left the rest of the message alone. Two changes have been made: (a) I've
6688 reduced the minimum length to 35 characters; (b) if it can't find a break
6689 point between 35 and 75 characters, it looks ahead and uses the first one
6690 that it finds. This may give the occasional overlong line, but at least the
6691 remaining text gets split now.
6693 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6694 file descriptors might be low, and that setting 1000 would always raise it.
6695 It turns out that in some environments, the limit is already over 1000 and
6696 that lowering it causes trouble. So now Exim takes care not to decrease it.
6698 40. When delivering a message, the value of $return_path is set to $sender_
6699 address at the start of routing (routers may change the value). By an
6700 oversight, this default was not being set up when an address was tested by
6701 -bt or -bv, which affected the outcome if any router or filter referred to
6704 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6705 log only when "message" or "log_message" are set. However, if one of the
6706 conditions was an address verification, or a call to a nested ACL, the
6707 messages generated by the underlying test were being passed through. This
6708 no longer happens. The underlying message is available in $acl_verify_
6709 message for both "message" and "log_message" expansions, so it can be
6710 passed through if needed.
6712 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6713 new expansion $bh_ to give the encoded byte string without charset
6714 translation. Translation happens only if iconv() is available; HAVE_ICONV
6715 indicates this at build time. HEADERS_CHARSET gives the charset to
6716 translate to; headers_charset can change it in the configuration, and
6717 "headers charset" can change it in an individual filter file.
6719 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6720 that creates RFC 2047 encoded "words" labels them as that charset instead
6721 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6722 expansion operator; (ii) when Exim creates a From: line for a local
6723 message; (iii) when a header line is rewritten to include a "phrase" part.
6725 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6726 buggy, causing it to skip the first lines of messages whose message ID
6727 ended in 'D'. This would not have bitten before Exim release 4.14, because
6728 message IDs were unlikely to end in 'D' before then. The effect was to have
6729 incorrect size information for certain domains.
6731 45. #include "config.h" was missing at the start of the crypt16.c module. This
6732 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6735 46. If there was a timeout during a "random" callout check, Exim treated it as
6736 a failure of the random address, and carried on sending RSET and the real
6737 address. If the delay was just some slowness somewhere, the response to the
6738 original RCPT would be taken as a response to RSET and so on, causing
6739 mayhem of various kinds.
6741 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6742 when I implemented it. It didn't allow for the fact that some option values
6743 may legitimately be negative (e.g. size_addition), and it didn't even do
6744 the right test for positive values.
6746 48. Domain names in DNS records are case-independent. Exim always looks them up
6747 in lower case. Some resolvers return domain names in exactly the case they
6748 appear in the zone file, that is, they may contain uppercase letters. Not
6749 all resolvers do this - some return always lower case. Exim was treating a
6750 change of case by a resolver as a change of domain, similar to a widening
6751 of a domain abbreviation. This triggered its re-routing code and so it was
6752 trying to route what was effectively the same domain again. This normally
6753 caused routing to fail (because the router wouldn't handle the domain
6754 twice). Now Exim checks for this case specially, and just changes the
6755 casing of the domain that it ultimately uses when it transmits the message
6758 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6761 50. If a filter generated a file delivery with a non-absolute name (possible if
6762 no home directory exists for the router), the forbid_file option was not
6765 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6766 the existing equality matching.
6768 52. Exim was using ints instead of ino_t variables in some places where it was
6769 dealing with inode numbers.
6771 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6772 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6773 called TMPDIR, and if it finds it is different, it changes its value.
6775 54. The smtp_printf() function is now made available to local_scan() so
6776 additional output lines can be written before returning. There is also an
6777 smtp_fflush() function to enable the detection of a dropped connection.
6778 The variables smtp_input and smtp_batched_input are exported to
6781 55. Changed the default runtime configuration: the message "Unknown user"
6782 has been removed from the ACL, and instead placed on the localuser router,
6783 using the cannot_route_message feature. This means that any verification
6784 failures that generate their own messages won't get overridden. Similarly,
6785 the "Unrouteable address" message that was in the ACL for unverifiable
6786 relay addresses has also been removed.
6788 56. Added hosts_avoid_esmtp to the smtp transport.
6790 57. The exicyclog script was not checking for the esoteric option
6791 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6792 will work only if exicyclog is run under the appropriate euid.
6794 58. Following a discussion on the list, the rules by which Exim recognises line
6795 endings on incoming messages have been changed. The -dropcr and drop_cr
6796 options are now no-ops, retained only for backwards compatibility. The
6797 following line terminators are recognized: LF CRLF CR. However, special
6798 processing applies to CR:
6800 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6801 nor a local message in the state where . is a terminator.
6803 (ii) If a bare CR is encountered in a header line, an extra space is added
6804 after the line terminator so as not to end the header. The reasoning
6805 behind this is that bare CRs in header lines are most likely either
6806 to be mistakes, or people trying to play silly games.
6808 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6809 was being incorrectly given as 18 bytes larger than it should have been.
6810 This is a VOB (very old bug).
6812 60. This may never have affected anything current, but just in case it has:
6813 When the local host is found other than at the start of a list of hosts,
6814 the local host, those with the same MX, and any that follow, are discarded.
6815 When the list in question was part of a longer list of hosts, the following
6816 hosts (not currently being processed) were also being discarded. This no
6817 longer happens. I'm not sure if this situation could ever has previously
6820 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6823 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6824 header. This is something that is recommended in a new Internet Draft, and
6825 is something that is documented as being done by Sendmail. There are two
6826 possible values. For messages generated by the autoreply transport, Exim
6829 Auto-Submitted: auto-replied
6831 whereas for all other generated messages (e.g. bounces) it adds
6833 Auto-Submitted: auto-generated
6835 63. The "personal" condition in filters now includes a test for the
6836 Auto-Submitted: header. If it contains the string "auto-" the message it
6837 not considered personal.
6839 64. Added rcpt_include_affixes as a generic transport option.
6841 65. Added queue_only_override (default true).
6843 66. Added the syslog_duplication option.
6845 67. If what should have been the first header line of a message consisted of
6846 a space followed by a colon, Exim was mis-interpreting it as a header line.
6847 It isn't of course - it is syntactically invalid and should therefore be
6848 treated as the start of the message body. The misbehaviour could have
6849 caused a number of strange effects, including loss of data in subsequent
6850 header lines, and spool format errors.
6852 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6853 client host had authenticated. This control can now be exercised by an ACL
6854 for more flexibility.
6856 69. By default, callouts do not happen when testing with -bh. There is now a
6857 variant, -bhc, which does actually run the callout code, including
6858 consulting and updating the callout cache.
6860 70. Added support for saslauthd authentication, courtesy of Alexander
6863 71. If statvfs() failed on the spool or log directories while checking their
6864 size for availability, Exim confusingly gave the error "space shortage".
6865 Furthermore, in debugging mode it crashed with a floating point exception.
6866 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6867 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6868 serious problem, Exim now writes to the main and panic logs when this
6869 happens, with details of the failure. It then refuses to accept the
6870 incoming message, giving the message "spool directory problem" or "log
6871 directory problem" with a 421 code for SMTP messages.
6873 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6874 0, 1, and 2 exist, because some OS complain for execs without them (see
6875 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6876 descriptors. However, the code omitted to check that the open succeeded,
6877 causing mysterious errors if for some reason the permissions on /dev/null
6878 got screwed. Now Exim writes a message to the main and panic logs, and
6879 bombs out if it can't open /dev/null.
6881 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6882 interact so that it is all more flexible. It is supposed to remain
6883 backwards compatible. Also added extra_local_interfaces.
6885 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6886 to bomb out with an assertion failure - to the client this appears as a
6887 connection drop. This problem occurs in the part of the code that was taken
6888 from the Samba project. Fortunately, the assertion is in a very simple
6889 function, so I have fixed this by reproducing the function inline in the
6890 one place where it is called, and arranging for authentication to fail
6891 instead of killing the process with assert().
6893 75. The SPA client code was not working when the server requested OEM rather
6894 than Unicode encoding.
6896 76. Added code to make require_files with a specific uid setting more usable in
6897 the case where statting the file as root fails - usually a non-root-mounted
6898 NFS file system. When this happens and the failure is EACCES, Exim now
6899 forks a subprocess and does the per-uid checking as the relevant uid.
6901 77. Added process_log_path.
6903 78. If log_file_path was not explicitly set, a setting of check_log_space or
6904 check_log_inodes was ignored.
6906 79. If a space check for the spool or log partitions fails, the incident is now
6907 logged. Of course, in the latter case the data may get lost...
6909 80. Added the %p formatting code to string_format() so that it can be used to
6910 print addresses in debug_print(). Adjusted all the address printing in the
6911 debugging in store.c to use %p rather than %d.
6913 81. There was a concern that a line of code in smtp_in.c could overflow a
6914 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6915 initially expressed, the concern was not well-founded, because trailing
6916 spaces are removed early. However, if the trailing spaces were followed by
6917 a NULL, they did not get removed, so the overflow was possible. Two fixes
6920 (a) I re-wrote the offending code in a cleaner fashion.
6921 (b) If an incoming SMTP command contains a NULL character, it is rejected
6924 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6925 runs initgroups(), so that if the Exim user is in any additional groups,
6926 they will be used during message reception.
6932 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob