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
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 (partial):
136 Avoid using SIZE on the MAIL for a callout verify, on any but
137 the main verify for receipient in uncached-mode.
143 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
144 than -2003 did; needs libidn2 in addition to libidn.
146 JH/02 The path option on a pipe transport is now expanded before use.
148 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
149 Patch provided by "Björn", documentation fix added too.
151 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
152 missing a wire-to-host endian conversion.
154 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
155 close after a BDAT command line could be taken as a following command,
156 giving a synch failure. Fix by only checking for synch immediately
157 before acknowledging the chunk.
159 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
160 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
161 macro. Patches provided by Josh Soref.
163 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
164 Previously we did not; the RFC seems ambiguous and VRFY is not listed
165 by IANA as a service extension. However, John Klensin suggests that we
168 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
169 the dkim code may be unix-mode line endings rather than smtp wire-format
170 CRLF, so prepend a CR to any bare LF.
172 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
173 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
175 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
176 the first were themselves being wrongly included in the feed into dkim
177 processing; with most chunk sizes in use this resulted in an incorrect
178 body hash calculated value.
180 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
181 DKIM signature block, for verification. Although advised against by
182 standards it is specifically not ruled illegal.
184 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
186 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
187 missing a body hash (the bh= tag).
189 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
190 It seems that HAProxy sends the Proxy Protocol information in clear and
191 only then does a TLS startup, so do the same.
193 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
194 TCP connections (such as for Spamd) unless the daemon successfully set
195 Fast Open mode on its listening sockets. This fixes breakage seen on
196 too-old kernels or those not configured for Fast Open, at the cost of
197 requiring both directions being enabled for TFO, and TFO never being used
198 by non-daemon-related Exim processes.
200 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
201 endings, at least on the first header line. Try to canonify any that get
202 past that check, despite the cost.
204 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
205 now limited to an arbitrary five deep, while parsing addresses with the
206 strip_excess_angle_brackets option enabled.
208 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
209 instead leave the unprompted TLS handshake in socket buffer for the
210 TLS library to consume.
212 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
214 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
216 JH/16 Drop variables when they go out of scope. Memory management drops a whole
217 region in one operation, for speed, and this leaves assigned pointers
218 dangling. Add checks run only under the testsuite which checks all
219 variables at a store-reset and panics on a dangling pointer; add code
220 explicitly nulling out all the variables discovered. Fixes one known
221 bug: a transport crash, where a dangling pointer for $sending_ip_address
222 originally assigned in a verify callout, is re-used.
224 PP/06 Drop '.' from @INC in various Perl scripts.
226 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
228 PP/08 Reduce a number of compilation warnings under clang; building with
229 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
230 should be warning-free.
232 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
234 HS/01 Fix portability problems introduced by PP/08 for platforms where
235 realloc(NULL) is not equivalent to malloc() [SunOS et al].
237 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
238 chunk. This allows us to accept broken chunked messages. We need a more
239 general solution here.
241 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
242 already-broken messages in the queue.
244 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
246 JH/19 Fix reference counting bug in routing-generated-address tracking.
252 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
253 supports it and a size is available (ie. the sending peer gave us one).
255 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
256 years of being deprecated). The replacements are the ACLs
257 acl_smtp_mime and acl_not_smtp_mime.
259 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
260 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
261 or even an in-clear connection were permitted. Now, if the host lookup
262 was dnssec and dane was requested then the host is only used if the
263 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
264 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
265 if one fails this test.
266 This means that a poorly-configured remote DNS will make it incommunicado;
267 but it protects against a DNS-interception attack on it.
269 JH/04 Bug 1810: make continued-use of an open smtp transport connection
270 non-noisy when a race steals the message being considered.
272 JH/05 If main configuration option tls_certificate is unset, generate a
273 self-signed certificate for inbound TLS connections.
275 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
276 in rewrites and routers.
278 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
279 and logged a warning sing 4.83; now they are a configuration file error.
281 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
282 (lacking @domain). Apply the same qualification processing as RCPT.
284 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
286 JH/10 Support ${sha256:} applied to a string (as well as the previous
289 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
290 a cutthrough deliver is pending, as we always want to make a connection.
291 This also avoids re-routing the message when later placing the cutthrough
292 connection after a verify cache hit.
293 Do not update it with the verify result either.
295 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
296 when routing results in more than one destination address.
298 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
299 signing (which inhibits the cutthrough capability). Previously only
300 the presence of an option was tested; now an expansion evaluating as
301 empty is permissible (obviously it should depend only on data available
302 when the cutthrough connection is made).
304 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
305 the relevant preceding SMTP command did not note the pipelining mode.
307 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
308 Previously they were not counted.
310 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
311 as one having no matching records. Previously we deferred the message
312 that needed the lookup.
314 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
315 distinguished as "(=".
317 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
318 for missing MX records. Previously it only worked for missing A records.
320 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
322 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
323 after the data-go-ahead and data-ack. Patch from Jason Betts.
325 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
326 even for a "none" policy. Patch from Tony Meyer.
328 JH/22 Fix continued use of a connection for further deliveries. If a port was
329 specified by a router, it must also match for the delivery to be
332 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
333 When one of the recipients of a message was unsuitable for the connection
334 (has no matching addresses), we lost track of needing to mark it
335 deferred. As a result mail would be lost.
337 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
339 JH/25 Decoding ACL controls is now done using a binary search; the source code
340 takes up less space and should be simpler to maintain. Merge the ACL
341 condition decode tables also, with similar effect.
343 JH/26 Fix problem with one_time used on a redirect router which returned the
344 parent address unchanged. A retry would see the parent address marked as
345 delivered, so not attempt the (identical) child. As a result mail would
348 JH/27 Fix a possible security hole, wherein a process operating with the Exim
349 UID can gain a root shell. Credit to http://www.halfdog.net/ for
350 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
353 JH/28 Enable {spool,log} filesystem space and inode checks as default.
354 Main config options check_{log,spool}_{inodes,space} are now
355 100 inodes, 10MB unless set otherwise in the configuration.
357 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
358 Previously it only applied to the main-section connection policy
361 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
363 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
364 by me. Added RFC7919 DH primes as an alternative.
366 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
367 are not in the system include path.
369 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
370 GnuTLS, when a session startup failed (eg because the client disconnected)
371 Exim did stdio operations after fclose. This was exposed by a recent
372 change which nulled out the file handle after the fclose.
374 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
375 signed directly by the cert-signing cert, rather than an intermediate
376 OCSP-signing cert. This is the model used by LetsEncrypt.
378 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
380 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
381 an incoming connection.
383 HS/02 Bug 1802: Do not half-close the connection after sending a request
386 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
387 fallback to "prime256v1".
389 JH/34 SECURITY: Use proper copy of DATA command in error message.
390 Could leak key material. Remotely exploitable. CVE-2016-9963.
396 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
397 and 3.4.4 - once the server is enabled to respond to an OCSP request
398 it does even when not requested, resulting in a stapling non-aware
399 client dropping the TLS connection.
401 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
402 support variable-length bit vectors. No functional change.
404 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
405 The I= interface field on outgoing lines is now after the H= remote
406 host field, same as incoming lines. There is a separate
407 outgoing_interface log selector which allows you to disable the
410 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
411 If not running log_selector +smtp_connection the mainlog would be held
412 open indefinitely after a "too many connections" event, including to a
413 deleted file after a log rotate. Leave the per net connection logging
414 leaving it open for efficiency as that will be quickly detected by the
415 check on the next write.
417 HS/01 Bug 1671: Fix post transport crash.
418 Processing the wait-<transport> messages could crash the delivery
419 process if the message IDs didn't exist for some reason. When
420 using 'split_spool_directory=yes' the construction of the spool
421 file name failed already, exposing the same netto behaviour.
423 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
424 mime_regex ACL conditions.
426 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
427 to DSN fail messages (bounces): remote IP, remote greeting, remote response
428 to HELO, local diagnostic string.
430 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
431 log line to debug. Even when configured with a tls authenticator many
432 client connections are expected to not authenticate in this way, so
433 an authenticate fail is not an error.
435 HS/02 Add the Exim version string to the process info. This way exiwhat
436 gives some more detail about the running daemon.
438 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
439 matter for fast-change records such as DNSBLs.
441 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
442 retry record, even if constant. There may be multiple transports with
443 different interface settings and the retry behaviour needs to be kept
446 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
448 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
450 JH/10 Bug 840: fix log_defer_output option of pipe transport
452 JH/11 Bug 830: use same host for all RCPTS of a message, even under
453 hosts_randomize. This matters a lot when combined with mua_wrapper.
455 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
456 ${quote_pgsql:<string>} operator.
458 JH/13 Bug 1708: avoid misaligned access in cached lookup.
460 JH/14 Change header file name for freeradius-client. Relevant if compiling
461 with Radius support; from the Gentoo tree and checked under Fedora.
463 JH/15 Bug 1712: Introduce $prdr_requested flag variable
465 JH/16 Bug 1714: Permit an empty string as expansion result for transport
466 option transport_filter, meaning no filtering.
468 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
470 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
471 defaults to "*" (all hosts). The variable is now available when not built
472 with TLS, default unset, mainly to enable keeping the testsuite sane.
473 If a server certificate is not supplied (via tls_certificate) an error is
474 logged, and clients will find TLS connections fail on startup. Presumably
475 they will retry in-clear.
476 Packagers of Exim are strongly encouraged to create a server certificate
477 at installation time.
479 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
480 with the $config_file variable.
482 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
483 in transport context, after the attempt, and per-recipient. The latter type
484 is per host attempted. The event data is the error message, and the errno
485 information encodes the lookup type (A vs. MX) used for the (first) host,
486 and the trailing two digits of the smtp 4xx response.
488 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
489 to write to mainlog (or rejectlog, paniclog) in the window between file
490 creation and permissions/ownership being changed. Particularly affects
491 installations where exicyclog is run as root, rather than exim user;
492 result is that the running daemon panics and dies.
494 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
496 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
497 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
498 "pri" and "weight". Note that the previous implicit priority given by the
499 list order is no longer honoured.
501 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
504 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
505 by defining SUPPORT_SOCKS.
507 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
508 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
509 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
510 variables are renamed to proxy_{local,external}_{address,port}.
512 JH/27 Move Internationalisation support from Experimental to mainline, enabled
513 for a build by defining SUPPORT_I18N
515 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
516 of the query string, and make ${quote_redis:} do that quoting.
518 JH/29 Move Events support from Experimental to mainline, enabled by default
519 and removable for a build by defining DISABLE_EVENT.
521 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
523 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
524 cached by the daemon.
526 JH/32 Move Redis support from Experimental to mainline, enabled for a build
527 by defining LOOKUP_REDIS. The libhiredis library is required.
529 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
530 keys are given for lookup.
532 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
533 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
534 only supported when built with TLS support. The PolarSSL SHA routines
535 are still used when the TLS library is too old for convenient support.
537 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
538 openssl_options), for security. OpenSSL forces this from version 1.1.0
539 server-side so match that on older versions.
541 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
542 allocation for $value could be released as the expansion processing
543 concluded, but leaving the global pointer active for it.
545 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
546 and to use the domains and local_parts ACL conditions.
548 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
549 incorrectly not doubled on cutthrough transmission, hence seen as a
550 body-termination at the receiving system - resulting in truncated mails.
551 Commonly the sender saw a TCP-level error, and retransmitted the message
552 via the normal store-and-forward channel. This could result in duplicates
553 received - but deduplicating mailstores were liable to retain only the
554 initial truncated version.
556 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
558 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
560 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
561 we're in there, support oversigning also; bug 1309.
563 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
565 HS/04 Add support for keep_environment and add_environment options.
567 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
568 either intentional arithmetic overflow during PRNG, or testing config-
571 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
572 delivery resulted in actual delivery. Cancel cutthrough before DATA
575 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
576 rejects a recipient: pass the reject to the originator.
578 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
579 Many were false-positives and ignorable, but it's worth fixing the
582 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
583 for the new environment-manipulation done at startup. Move the routines
584 from being local to tls.c to being global via the os.c file.
586 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
587 an extract embedded as result-arg for a map, the first arg for extract
588 is unavailable so we cannot tell if this is a numbered or keyed
589 extraction. Accept either.
595 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
598 JH/02 The smtp transport option "multi_domain" is now expanded.
600 JH/03 The smtp transport now requests PRDR by default, if the server offers
603 JH/04 Certificate name checking on server certificates, when exim is a client,
604 is now done by default. The transport option tls_verify_cert_hostnames
605 can be used to disable this per-host. The build option
606 EXPERIMENTAL_CERTNAMES is withdrawn.
608 JH/05 The value of the tls_verify_certificates smtp transport and main options
609 default to the word "system" to access the system default CA bundle.
610 For GnuTLS, only version 3.0.20 or later.
612 JH/06 Verification of the server certificate for a TLS connection is now tried
613 (but not required) by default. The verification status is now logged by
614 default, for both outbound TLS and client-certificate supplying inbound
617 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
620 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
621 Status Notification (bounce) messages are now MIME format per RFC 3464.
622 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
623 under the control of the dsn_advertise_hosts option, and routers may
624 have a dsn_lasthop option.
626 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
627 default, modifiable by a malware= option. The list separator for
628 the options can now be changed in the usual way. Bug 68.
630 JH/10 The smtp_receive_timeout main option is now expanded before use.
632 JH/11 The incoming_interface log option now also enables logging of the
633 local interface on delivery outgoing connections.
635 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
636 if the interface and destination host and port all match.
638 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
641 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
642 Patch from Andrew Lewis.
644 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
645 now supports optional time-restrictions, weighting, and priority
646 modifiers per server. Patch originally by <rommer@active.by>.
648 JH/16 The spamd_address main option now supports a mixed list of local
649 and remote servers. Remote servers can be IPv6 addresses, and
650 specify a port-range.
652 JH/17 Bug 68: The spamd_address main option now supports an optional
653 timeout value per server.
655 JH/18 Bug 1581: Router and transport options headers_add/remove can
656 now have the list separator specified.
658 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
661 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
664 JH/21 Support for the A6 type of dns record is withdrawn.
666 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
667 rather than the verbs used.
669 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
670 from 255 to 1024 chars.
672 JH/24 Verification callouts now attempt to use TLS by default.
674 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
675 are generic router options now. The defaults didn't change.
677 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
678 Original patch from Alexander Shikoff, worked over by JH.
680 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
681 files if ZCAT_COMMAND is not executable.
683 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
685 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
687 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
688 Normally benign, it bites when the pair was led to by a CNAME;
689 modern usage is to not canonicalize the domain to a CNAME target
690 (and we were inconsistent anyway for A-only vs AAAA+A).
692 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
694 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
695 when evaluating $sender_host_dnssec.
697 JH/31 Check the HELO verification lookup for DNSSEC, adding new
698 $sender_helo_dnssec variable.
700 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
702 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
704 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
706 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
707 documented as working, but never had. Support all but $spam_report.
709 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
710 added for tls authenticator.
712 HS/03 Add perl_taintmode main config option
718 TL/01 When running the test suite, the README says that variables such as
719 no_msglog_check are global and can be placed anywhere in a specific
720 test's script, however it was observed that placement needed to be near
721 the beginning for it to behave that way. Changed the runtest perl
722 script to read through the entire script once to detect and set these
723 variables, reset to the beginning of the script, and then run through
724 the script parsing/test process like normal.
726 TL/02 The BSD's have an arc4random API. One of the functions to induce
727 adding randomness was arc4random_stir(), but it has been removed in
728 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
729 function when detected.
731 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
732 cause callback expansion.
734 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
735 syntax errors in an expansion can be treated as a string instead of
736 logging or causing an error, due to the internal use of bool_lax
737 instead of bool when processing it.
739 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
740 server certificates when making smtp deliveries.
742 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
744 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
746 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
748 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
749 Merged patch from Sebastian Wiedenroth.
751 JH/05 Fix results-pipe from transport process. Several recipients, combined
752 with certificate use, exposed issues where response data items split
753 over buffer boundaries were not parsed properly. This eventually
754 resulted in duplicates being sent. This issue only became common enough
755 to notice due to the introduction of connection certificate information,
756 the item size being so much larger. Found and fixed by Wolfgang Breyha.
758 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
759 size buffer was used, resulting in syntax errors when an expansion
762 JH/07 Add support for directories of certificates when compiled with a GnuTLS
763 version 3.3.6 or later.
765 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
766 is EXPERIMENTAL_EVENT, the main-configuration and transport options
767 both become "event_action", the variables become $event_name, $event_data
768 and $event_defer_errno. There is a new variable $verify_mode, usable in
769 routers, transports and related events. The tls:cert event is now also
770 raised for inbound connections, if the main configuration event_action
773 TL/06 In test suite, disable OCSP for old versions of openssl which contained
774 early OCSP support, but no stapling (appears to be less than 1.0.0).
776 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
777 server certificate names available under the smtp transport option
778 "tls_verify_cert_hostname" now do not permit multi-component wildcard
781 JH/10 Time-related extraction expansions from certificates now use the main
782 option "timezone" setting for output formatting, and are consistent
783 between OpenSSL and GnuTLS compilations. Bug 1541.
785 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
786 encoded parameter in the incoming message. Bug 1558.
788 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
789 include certificate info, eximon was claiming there were spoolfile
792 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
794 JH/14 Log delivery-related information more consistently, using the sequence
795 "H=<name> [<ip>]" wherever possible.
797 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
798 are problematic for Debian distribution, omit them from the release
801 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
803 JH/16 Fix string representation of time values on 64bit time_t architectures.
806 JH/17 Fix a null-indirection in certextract expansions when a nondefault
807 output list separator was used.
812 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
813 checkers that were complaining about end of non-void function with no
816 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
817 This was a regression introduced in 4.83 by another bugfix.
819 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
821 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
822 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
828 TF/01 Correctly close the server side of TLS when forking for delivery.
830 When a message was received over SMTP with TLS, Exim failed to clear up
831 the incoming connection properly after forking off the child process to
832 deliver the message. In some situations the subsequent outgoing
833 delivery connection happened to have the same fd number as the incoming
834 connection previously had. Exim would try to use TLS and fail, logging
835 a "Bad file descriptor" error.
837 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
838 utilities have not been installed.
840 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
841 temporary space as the ACL may create new global variables.
843 TL/01 LDAP support uses per connection or global context settings, depending
844 upon the detected version of the libraries at build time.
846 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
847 to extract and use the src ip:port in logging and expansions as if it
848 were a direct connection from the outside internet. PPv2 support was
849 updated based on HAProxy spec change in May 2014.
851 JH/02 Add ${listextract {number}{list}{success}{fail}}.
853 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
854 Properly escape header and check for NULL return.
856 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
859 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
861 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
862 characters in header names, implemented as a verify condition.
863 Contributed by Michael Fischer v. Mollard.
865 TL/05 Rename SPF condition results err_perm and err_temp to standardized
866 results permerror and temperror. Previous values are deprecated but
867 still accepted. In a future release, err_perm and err_temp will be
868 completely removed, which will be a backward incompatibility if the
869 ACL tests for either of these two old results. Patch contributed by
870 user bes-internal on the mailing list.
872 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
874 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
875 selectors, in both main and reject logs.
877 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
880 JH/07 Add malware type "sock" for talking to simple daemon.
882 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
884 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
885 routers/transports under cutthrough routing.
887 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
888 numbers. Touch up "bool" conditional to keep the same definition.
890 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
892 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
894 JH/12 Expand items in router/transport headers_add or headers_remove lists
895 individually rather than the list as a whole. Bug 1452.
897 Required for reasonable handling of multiple headers_ options when
898 they may be empty; requires that headers_remove items with embedded
899 colons must have them doubled (or the list-separator changed).
901 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
902 view the policy declared in the DMARC record. Currently, $dmarc_status
903 is a combined value of both the record presence and the result of the
906 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
908 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
909 dnslookup router and the smtp transport (applying to the forward
912 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
913 of ldap servers used for a specific lookup. Patch provided by Heiko
916 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
917 New variable $lookup_dnssec_authenticated for observability.
919 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
920 Patch submitted by Lars Timman.
922 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
924 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
925 Requires trusted mode and valid format message id, aborts otherwise.
926 Patch contributed by Heiko Schlichting.
928 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
929 certextract with support for various fields. Bug 1358.
931 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
932 is requested by default, modifiable by smtp transport option
935 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
936 operate on certificate variables to give certificate fingerprints
937 Also new ${sha256:cert_variable}.
939 JH/23 The PRDR feature is moved from being Experimental into the mainline.
941 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
942 Christian Aistleitner.
944 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
946 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
947 file. Patch from Wolfgang Breyha.
949 JH/25 Expand the coverage of the delivery $host and $host_address to
950 client authenticators run in verify callout. Bug 1476.
952 JH/26 Port service names are now accepted for tls_on_connect_ports, to
953 align with daemon_smtp_ports. Bug 72.
955 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
956 support and error reporting did not work properly.
958 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
959 and is readable. Patch from Andrew Colin Kissa.
961 TL/14 Enhance documentation of ${run expansion and how it parses the
962 commandline after expansion, particularly in the case when an
963 unquoted variable expansion results in an empty value.
965 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
967 PP/02 Fix internal collision of T_APL on systems which support RFC3123
968 by renaming away from it. Addresses GH issue 15, reported by
971 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
973 TL/15 SECURITY: prevent double expansion in math comparison functions
974 (can expand unsanitized data). Not remotely exploitable.
981 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
983 PP/02 Make -n do something, by making it not do something.
984 When combined with -bP, the name of an option is not output.
986 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
989 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
990 $sender_host_name and config options to manage this, and basic check
993 PP/05 DSCP support for outbound connections and control modifier for inbound.
995 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
996 (Only plugin which currently uses this is kerberos4, which nobody should
997 be using, but we should make it available and other future plugins might
998 conceivably use it, even though it would break NAT; stuff *should* be
999 using channel bindings instead).
1001 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1002 name; added for Sendmail compatibility; requires admin caller.
1003 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1004 just ignore it); requires trusted caller.
1005 Also parse but ignore: -Ac -Am -X<logfile>
1008 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1010 TL/02 Add +smtp_confirmation as a default logging option.
1012 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1013 Patch by Magnus Holmgren from 2007-02-20.
1015 TL/04 Bugzilla 1281 - Spec typo.
1016 Bugzilla 1283 - Spec typo.
1017 Bugzilla 1290 - Spec grammar fixes.
1019 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1021 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1023 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1024 mailing list by by Dmitry Isaikin.
1026 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1028 JH/02 Support "G" suffix to numbers in ${if comparisons.
1030 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1032 NM/01 Bugzilla 1197 - Spec typo
1033 Bugzilla 1196 - Spec examples corrections
1035 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1037 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1038 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1041 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1042 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1044 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1045 "acl {{name}{arg}...}", and optional args on acl condition
1048 JH/05 Permit multiple router/transport headers_add/remove lines.
1050 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1052 JH/07 Avoid using a waiting database for a single-message-only transport.
1053 Performance patch from Paul Fisher. Bugzilla 1262.
1055 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1058 JH/09 Add $headers_added variable, with content from use of ACL modifier
1059 add_header (but not yet added to the message). Bugzilla 199.
1061 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1062 Pulled from Bugzilla 817 by Wolfgang Breyha.
1064 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1066 (nb: this is the same fix as in Exim 4.80.1)
1068 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1071 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1074 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1076 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1077 advertise SMTP AUTH mechanism to us, instead of a generic
1078 protocol violation error. Also, make Exim more robust to bad
1079 data from the Dovecot auth socket.
1081 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1083 When a queue runner is handling a message, Exim first routes the
1084 recipient addresses, during which it prunes them based on the retry
1085 hints database. After that it attempts to deliver the message to
1086 any remaining recipients. It then updates the hints database using
1089 So if a recipient address works intermittently, it can get repeatedly
1090 deferred at routing time. The retry hints record remains fresh so the
1091 address never reaches the final cutoff time.
1093 This is a fairly common occurrence when a user is bumping up against
1094 their storage quota. Exim had some logic in its local delivery code
1095 to deal with this. However it did not apply to per-recipient defers
1096 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1098 This change adds a proper retry rule check during routing so that the
1099 final cutoff time is checked against the message's age. We only do
1100 this check if there is an address retry record and there is not a
1101 domain retry record; this implies that previous attempts to handle
1102 the address had the retry_use_local_parts option turned on. We use
1103 this as an approximation for the destination being like a local
1104 delivery, as in LMTP.
1106 I suspect this new check makes the old local delivery cutoff check
1107 redundant, but I have not verified this so I left the code in place.
1109 TF/02 Correct gecos expansion when From: is a prefix of the username.
1111 Test 0254 submits a message to Exim with the header
1115 When I ran the test suite under the user fanf2, Exim expanded
1116 the header to contain my full name, whereas it should have added
1117 a Resent-Sender: header. It erroneously treats any prefix of the
1118 username as equal to the username.
1120 This change corrects that bug.
1122 GF/01 DCC debug and logging tidyup
1123 Error conditions log to paniclog rather than rejectlog.
1124 Debug lines prefixed by "DCC: " to remove any ambiguity.
1126 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1128 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1129 Bug spotted by Jeremy Harris; was flawed since initial commit.
1130 Would have resulted in OCSP responses post-SNI triggering an Exim
1131 NULL dereference and crash.
1133 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1135 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1136 Bug detection, analysis and fix by Samuel Thibault.
1137 Bugzilla 1331, Debian bug #698092.
1139 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1141 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1142 Server implementation by Todd Lyons, client by JH.
1143 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1144 config variable "prdr_enable" controls whether the server
1145 advertises the facility. If the client requests PRDR a new
1146 acl_data_smtp_prdr ACL is called once for each recipient, after
1147 the body content is received and before the acl_smtp_data ACL.
1148 The client is controlled by both of: a hosts_try_prdr option
1149 on the smtp transport, and the server advertisement.
1150 Default client logging of deliveries and rejections involving
1151 PRDR are flagged with the string "PRDR".
1153 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1154 fclose(). Diagnosis by Todd Lyons.
1156 PP/17 Update configure.default to handle IPv6 localhost better.
1157 Patch by Alain Williams (plus minor tweaks).
1160 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1161 This is now consistent with GnuTLS, and is now documented: the
1162 previous undocumented portable approach to treating the option as
1163 unset was to force an expansion failure. That still works, and
1164 an empty string is now equivalent.
1166 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1167 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1168 not performing validation itself.
1170 PP/20 Added force_command boolean option to pipe transport.
1171 Patch from Nick Koston, of cPanel Inc.
1173 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1176 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1178 PP/21 Fix eximon continuous updating with timestamped log-files.
1179 Broken in a format-string cleanup in 4.80, missed when I repaired the
1180 other false fix of the same issue.
1181 Report and fix from Heiko Schlichting.
1184 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1185 Report from Prashanth Katuri.
1187 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1188 It's SecureTransport, so affects any MacOS clients which use the
1189 system-integrated TLS libraries, including email clients.
1191 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1192 using a MIME ACL for non-SMTP local injection.
1193 Report and assistance in diagnosis by Warren Baker.
1195 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1197 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1199 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1200 last id that failed so it may be referenced in subsequent ACL's.
1202 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1205 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1206 ldap library initialization, allowing self-signed CA's to be
1207 used. Also properly sets require_cert option later in code by
1208 using NULL (global ldap config) instead of ldap handle (per
1209 session). Bug diagnosis and testing by alxgomz.
1211 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1212 the src/util/ subdirectory.
1214 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1215 renamed to Transport Post Delivery Action by Jeremy Harris, as
1218 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1219 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1220 redis_servers = needs to be configured which will be used by the redis
1221 lookup. Patch from Warren Baker, of The Packet Hub.
1223 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1225 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1226 hostname or reverse DNS when processing a host list. Used suggestions
1227 from multiple comments on this bug.
1229 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1231 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1232 Missed a few lines, added it to make the runtest require no keyboard
1235 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1236 contains upper case chars. Make router use caseful_local_part.
1238 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1239 support when GnuTLS has been built with p11-kit.
1245 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1247 This, or similar/improved, will also be change PP/11 of 4.82.
1253 PP/01 Handle short writes when writing local log-files.
1254 In practice, only affects FreeBSD (8 onwards).
1255 Bugzilla 1053, with thanks to Dmitry Isaikin.
1257 NM/01 Bugzilla 949 - Documentation tweak
1259 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1262 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1264 PP/02 Implemented gsasl authenticator.
1266 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1268 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1269 `pkg-config foo` for cflags/libs.
1271 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1272 with rest of GSASL and with heimdal_gssapi.
1274 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1275 `pkg-config foo` for cflags/libs for the TLS implementation.
1277 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1278 properties get this fed in as external SSF. A number of robustness
1279 and debugging improvements to the cyrus_sasl authenticator.
1281 PP/08 cyrus_sasl server now expands the server_realm option.
1283 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1284 Patch by Jeremy Harris.
1286 PP/10 Added dbmjz lookup type.
1288 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1290 PP/12 MAIL args handles TAB as well as SP, for better interop with
1291 non-compliant senders.
1292 Analysis and variant patch by Todd Lyons.
1294 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1295 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1296 Patch from Dirk Mueller <dmueller@suse.com>
1298 PP/13 tls_peerdn now print-escaped for spool files.
1299 Observed some $tls_peerdn in wild which contained \n, which resulted
1300 in spool file corruption.
1302 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1303 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1304 or write after TLS renegotiation, which otherwise led to messages
1307 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1308 as a tracking header (ie: a signed header comes before the signature).
1309 Patch from Wolfgang Breyha.
1311 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1312 comma-sep list; embedded commas doubled.
1314 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1316 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1318 Report and patch from Dmitry Banschikov.
1320 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1321 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1322 failures. We appear to now support TLS1.1+ with Exim.
1324 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1325 lets Exim select keys and certificates based upon TLS SNI from client.
1326 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1327 before an outbound SMTP session. New log_selector, +tls_sni.
1329 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1330 NULL dereference. Report and patch from Alun Jones.
1332 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1333 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1334 Not seeing resolver debug output on NetBSD, but suspect this is a
1335 resolver implementation change.
1337 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1338 Left warnings. Added "eximon gdb" invocation mode.
1340 PP/21 Defaulting "accept_8bitmime" to true, not false.
1342 PP/22 Added -bw for inetd wait mode support.
1344 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1345 locate the relevant includes and libraries. Made this the default.
1347 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1348 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1350 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1351 This may cause build issues on older platforms.
1353 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1354 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1355 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1356 Added SNI support via GnuTLS too.
1357 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1359 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1361 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1362 Applied second patch from Janne, implementing suggestion to default
1363 multiple-strings-in-record handling to match SPF spec.
1365 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1367 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1368 read-only, out of scope).
1369 Patch from Wolfgang Breyha, report from Stuart Northfield.
1371 PP/29 Fix three issues highlighted by clang analyser static analysis.
1372 Only crash-plausible issue would require the Cambridge-specific
1373 iplookup router and a misconfiguration.
1374 Report from Marcin Mirosław.
1376 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1378 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1379 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1380 As part of this, removing so much warning spew let me fix some minor
1381 real issues in debug logging.
1383 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1384 assignment on my part. Fixed.
1386 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1387 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1388 Janne Snabb (who went above and beyond: thank you).
1390 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1391 string otherwise requires a connection and a bunch more work and it's
1392 relatively easy to get wrong. Should also expose TLS library linkage
1395 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1396 64-bit ${eval} (JH/03).
1398 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1399 GNU libc to support some of the 64-bit stuff, should not lead to
1400 conflicts. Defined before os.h is pulled in, so if a given platform
1401 needs to override this, it can.
1403 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1404 protection layer was required, which is not implemented.
1405 Bugzilla 1254, patch from Wolfgang Breyha.
1407 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1408 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1409 tls_dhparam take prime identifiers. Also unbreak combination of
1410 OpenSSL+DH_params+TLSSNI.
1412 PP/39 Disable SSLv2 by default in OpenSSL support.
1418 PP/01 Solaris build fix for Oracle's LDAP libraries.
1419 Bugzilla 1109, patch from Stephen Usher.
1421 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1423 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1426 TF/02 Fix a couple more cases where we did not log the error message
1427 when unlink() failed. See also change 4.74-TF/03.
1429 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1430 lock up or crash if it happened to be inside a call to libc when it
1431 got a SIGUSR1 from exiwhat.
1433 The SIGUSR1 handler appends the current process status to the process
1434 log which is later printed by exiwhat. It used to use the general
1435 purpose logging code to do this, but several functions it calls are
1436 not safe for signals.
1438 The new output code in the SIGUSR1 handler is specific to the process
1439 log, and simple enough that it's easy to inspect for signal safety.
1440 Removing some special cases also simplifies the general logging code.
1441 Removing the spurious timestamps from the process log simplifies
1444 TF/04 Improved ratelimit ACL condition.
1446 The /noupdate option has been deprecated in favour of /readonly which
1447 has clearer semantics. The /leaky, /strict, and /readonly update modes
1448 are mutually exclusive. The update mode is no longer included in the
1449 database key; it just determines when the database is updated. (This
1450 means that when you upgrade Exim will forget old rate measurements.)
1452 Exim now checks that the per_* options are used with an update mode that
1453 makes sense for the current ACL. For example, when Exim is processing a
1454 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1455 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1456 must specify per_mail/readonly. If you omit the update mode it defaults to
1457 /leaky where that makes sense (as before) or /readonly where required.
1459 The /noupdate option is now undocumented but still supported for
1460 backwards compatibility. It is equivalent to /readonly except that in
1461 ACLs where /readonly is required you may specify /leaky/noupdate or
1462 /strict/noupdate which are treated the same as /readonly.
1464 A useful new feature is the /count= option. This is a generalization
1465 of the per_byte option, so that you can measure the throughput of other
1466 aggregate values. For example, the per_byte option is now equivalent
1467 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1469 The per_rcpt option has been generalized using the /count= mechanism
1470 (though it's more complicated than the per_byte equivalence). When it is
1471 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1472 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1473 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1474 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1475 Note that using per_rcpt with a non-readonly update mode in more than
1476 one ACL will cause the recipients to be double-counted. (The per_mail
1477 and per_byte options don't have this problem.)
1479 The handling of very low rates has changed slightly. If the computed rate
1480 is less than the event's count (usually one) then this event is the first
1481 after a long gap. In this case the rate is set to the same as this event's
1482 count, so that the first message of a spam run is counted properly.
1484 The major new feature is a mechanism for counting the rate of unique
1485 events. The new per_addr option counts the number of different
1486 recipients that someone has sent messages to in the last time period. It
1487 behaves like per_rcpt if all the recipient addresses are different, but
1488 duplicate recipient addresses do not increase the measured rate. Like
1489 the /count= option this is a general mechanism, so the per_addr option
1490 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1491 example, measure the rate that a client uses different sender addresses
1492 with the options per_mail/unique=$sender_address. There are further
1493 details in the main documentation.
1495 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1497 TF/06 Removed a few PCRE remnants.
1499 TF/07 Automatically extract Exim's version number from tags in the git
1500 repository when doing development or release builds.
1502 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1503 Bugzilla 879. Patch from Paul Fisher.
1505 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1506 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1509 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1511 PP/05 Variable $av_failed, true if the AV scanner deferred.
1512 Bugzilla 1078. Patch from John Horne.
1514 PP/06 Stop make process more reliably on build failure.
1515 Bugzilla 1087. Patch from Heiko Schlittermann.
1517 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1518 Bugzilla 1089. Patch from Heiko Schlittermann.
1520 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1521 Bugzilla 1131. Patch from Holger Weiß.
1523 PP/09 Handle IPv6 addresses with SPF.
1524 Bugzilla 860. Patch from Wolfgang Breyha.
1526 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1528 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1531 PP/11 match_* no longer expand right-hand-side by default.
1532 New compile-time build option, EXPAND_LISTMATCH_RHS.
1533 New expansion conditions, "inlist", "inlisti".
1535 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1537 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1539 PP/14 fix log_write() format string regression from TF/03.
1540 Bugzilla 1152. Patch from Dmitry Isaikin.
1546 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1548 PP/02 Harmonised TLS library version reporting; only show if debugging.
1549 Layout now matches that introduced for other libraries in 4.74 PP/03.
1551 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1553 PP/04 New "dns_use_edns0" global option.
1555 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1558 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1559 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1561 TK/01 Updated PolarSSL code to 0.14.2.
1562 Bugzilla 1097. Patch from Andreas Metzler.
1564 PP/07 Catch divide-by-zero in ${eval:...}.
1565 Fixes bugzilla 1102.
1567 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1570 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1571 format-string attack -- SECURITY: remote arbitrary code execution.
1573 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1574 time unintentionally subject to list matching rules, letting the header
1575 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1576 arbitrary string expansion). This allowed for information disclosure.
1578 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1579 INT_MIN/-1 -- value coerced to INT_MAX.
1585 NM/01 Workaround for PCRE version dependency in version reporting
1588 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1589 This fixes portability to compilers other than gcc, notably
1590 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1592 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1593 makefiles for portability to HP-UX and POSIX correctness.
1595 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1596 Also via indented variable definition in the Makefile.
1597 (Debugging by Oliver Heesakkers).
1599 PP/02 Restore caching of spamd results with expanded spamd_address.
1600 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1602 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1603 Improves build reliability. Fix from: Frank Elsner
1605 NM/02 Fix wide character breakage in the rfc2047 coding
1606 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1608 NM/03 Allow underscore in dnslist lookups
1609 Fixes bug 1026. Patch from Graeme Fowler
1611 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1612 Code patches from Adam Ciarcinski of NetBSD.
1614 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1617 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1618 is logged, to avoid truncation. Patch from John Horne.
1620 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1621 Patch from Jakob Hirsch.
1623 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1624 SQL string expansion failure details.
1625 Patch from Andrey Oktyabrski.
1627 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1628 Patch from Simon Arlott.
1630 PP/09 New lookups functionality failed to compile on old gcc which rejects
1631 extern declarations in function scope.
1632 Patch from Oliver Fleischmann
1634 PP/10 Use sig_atomic_t for flags set from signal handlers.
1635 Check getgroups() return and improve debugging.
1636 Fixed developed for diagnosis in bug 927 (which turned out to be
1639 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1640 Patch from Mark Zealey.
1642 PP/12 Bugzilla 1056: Improved spamd server selection.
1643 Patch from Mark Zealey.
1645 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1646 Based on patch from Heiko Schlittermann.
1648 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1649 Patch from Uwe Doering, sign-off by Michael Haardt.
1651 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1652 variable declaration deep within a block. Bug and patch from
1655 PP/15 lookups-Makefile IRIX compatibility coercion.
1657 PP/16 Make DISABLE_DKIM build knob functional.
1659 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1660 Patch by Simon Arlott
1662 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1663 variable argument macros. Our copy now differs from upstream.
1669 TF/01 Failure to get a lock on a hints database can have serious
1670 consequences so log it to the panic log.
1672 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1673 controlled using the smtp_confirmation log selector.
1675 TF/03 Include the error message when we fail to unlink a spool file.
1677 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1678 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1679 for maintaining out-of-tree patches for some time.
1681 PP/01 Bugzilla 139: Documentation and portability issues.
1682 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1683 Handle per-OS dynamic-module compilation flags.
1685 PP/02 Let /dev/null have normal permissions.
1686 The 4.73 fixes were a little too stringent and complained about the
1687 permissions on /dev/null. Exempt it from some checks.
1688 Reported by Andreas M. Kirchwitz.
1690 PP/03 Report version information for many libraries, including
1691 Exim version information for dynamically loaded libraries. Created
1692 version.h, now support a version extension string for distributors
1693 who patch heavily. Dynamic module ABI change.
1695 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1696 privilege escalation vulnerability whereby the Exim run-time user
1697 can cause root to append content of the attacker's choosing to
1700 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1703 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1704 If dropping privileges for untrusted macros, we disabled normal logging
1705 on the basis that it would fail; for the Exim run-time user, this is not
1706 the case, and it resulted in successful deliveries going unlogged.
1707 Fixed. Reported by Andreas Metzler.
1713 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1714 only prepend for the Resent-* case. Fixes regression introduced in
1715 Exim 4.70 by NM/22 for Bugzilla 607.
1717 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1718 increasing numbers of administrators be bitten by this.
1720 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1722 PP/03 Bugzilla 994: added openssl_options main configuration option.
1724 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1726 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1728 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1730 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1731 without a peer certificate, leading to a segfault because of an
1732 assumption that peers always have certificates. Be a little more
1733 paranoid. Problem reported by Martin Tscholak.
1735 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1736 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1737 NB: ClamAV planning to remove STREAM in "middle of 2010".
1738 CL also introduces -bmalware, various -d+acl logging additions and
1739 more caution in buffer sizes.
1741 PP/09 Implemented reverse_ip expansion operator.
1743 PP/10 Bugzilla 937: provide a "debug" ACL control.
1745 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1747 PP/12 Bugzilla 973: Implement --version.
1749 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1751 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1753 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1755 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1756 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1757 ignore trailing whitespace.
1759 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1761 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1764 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1765 Notification from Dr Andrew Aitchison.
1767 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1768 ExtendedDetectionInfo response format.
1769 Notification from John Horne.
1771 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1774 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1775 XSL and documented dependency on system catalogs, with examples of how
1778 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1781 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1782 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1783 configuration file which is writeable by the Exim user or group.
1785 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1786 of configuration files to cover files specified with the -C option if
1787 they are going to be used with root privileges, not just the default
1790 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1791 option (effectively making it always true).
1793 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1794 files to be used while preserving root privileges.
1796 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1797 that rogue child processes cannot use them.
1799 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1800 run-time user, instead of root.
1802 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1803 Exim run-time user without dropping privileges.
1805 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1806 result string, instead of calling string_vformat() twice with the same
1809 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1810 for other users. Others should always drop root privileges if they use
1811 -C on the command line, even for a whitelisted configure file.
1813 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1815 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1821 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1822 $data_path, and $header_path variables; fixed documentation bugs and
1825 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1826 exipick to access non-standard spools, including the "frozen" queue
1829 NM/01 Bugzilla 965: Support mysql stored procedures.
1830 Patch from Alain Williams
1832 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1834 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1835 Patch from Andreas Metzler
1837 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1838 Patch from Kirill Miazine
1840 NM/05 Bugzilla 671: Added umask to procmail example.
1842 JJ/03 installed exipick 20100323.0, fixing doc bug
1844 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1845 directory. Notification and patch from Dan Rosenberg.
1847 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1849 TK/02 Improve log output when DKIM signing operation fails.
1851 MH/01 Treat the transport option dkim_domain as a colon separated
1852 list, not as a single string, and sign the message with each element,
1853 omitting multiple occurences of the same signer.
1855 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1856 Bugzilla 985, 986. Patch by Simon Arlott
1858 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1859 Patch by Simon Arlott
1861 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1862 MBX locking. Notification from Dan Rosenberg.
1868 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1870 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1872 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1874 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1876 NM/04 Bugzilla 915: Flag broken perl installation during build.
1882 TK/01 Added patch by Johannes Berg that expands the main option
1883 "spamd_address" if it starts with a dollar sign.
1885 TK/02 Write list of recipients to X-Envelope-Sender header when building
1886 the mbox-format spool file for content scanning (suggested by Jakob
1889 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1890 (http://www.dcc-servers.net/) support via dccifd. Activated by
1891 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1893 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1894 by Mark Daniel Reidel <mr@df.eu>.
1896 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1897 When building exim an external PCRE library is now needed -
1898 PCRE is a system library on the majority of modern systems.
1899 See entry on PCRE_LIBS in EDITME file.
1901 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1902 conversation. Added nologin parameter to request.
1903 Patch contributed by Kirill Miazine.
1905 TF/01 Do not log submission mode rewrites if they do not change the address.
1907 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1909 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1910 log files in place. Contributed by Roberto Lima.
1912 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1914 TF/03 Bugzilla 615: When checking the local_parts router precondition
1915 after a local_part_suffix or local_part_prefix option, Exim now
1916 does not use the address's named list lookup cache, since this
1917 contains cached lookups for the whole local part.
1919 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1920 Robert Millan. Documentation is in experimental-spec.txt.
1922 TF/04 Bugzilla 668: Fix parallel build (make -j).
1924 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1926 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1927 Patch provided by Jan Srzednicki.
1929 TF/05 Leading white space used to be stripped from $spam_report which
1930 wrecked the formatting. Now it is preserved.
1932 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1933 that they are available at delivery time.
1935 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1937 TF/08 TLS error reporting now respects the incoming_interface and
1938 incoming_port log selectors.
1940 TF/09 Produce a more useful error message if an SMTP transport's hosts
1941 setting expands to an empty string.
1943 NM/06 Bugzilla 744: EXPN did not work under TLS.
1944 Patch provided by Phil Pennock.
1946 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1947 Patch provided by Richard Godbee.
1949 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1950 acl_smtp_notquit, added index entry.
1952 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1953 Patch provided by Eugene Bujak.
1955 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1956 accept(). Patch provided by Maxim Dounin.
1958 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1959 Patch provided by Phil Pennock.
1961 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1963 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1964 Patch provided by Brad "anomie" Jorsch.
1966 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1967 Patch provided by Dean Brooks.
1969 TK/05 Add native DKIM support (does not depend on external libraries).
1971 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1972 Patch provided by Graeme Fowler.
1974 NM/16 Bugzilla 851: Documentation example syntax fix.
1976 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1978 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1981 NM/19 Bugzilla 745: TLS version reporting.
1982 Patch provided by Phil Pennock.
1984 NM/20 Bugzilla 167: bool: condition support.
1985 Patch provided by Phil Pennock.
1987 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1988 clients. Patch provided by Phil Pennock.
1990 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1991 Patch provided by Brad "anomie" Jorsch.
1993 NM/23 Bugzilla 687: Fix misparses in eximstats.
1994 Patch provided by Heiko Schlittermann.
1996 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1997 Patch provided by Heiko Schlittermann.
1999 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2000 plus update to original patch.
2002 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2004 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2005 Patch provided by David Brownlee.
2007 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2009 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2011 NM/30 Bugzilla 888: TLS documentation bugfixes.
2013 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2015 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2016 Unlike the original bugzilla I have changed all shell scripts in src tree.
2018 NM/33 Bugzilla 898: Transport filter timeout fix.
2019 Patch by Todd Rinaldo.
2021 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2022 Patch by Serge Demonchaux.
2024 NM/35 Bugzilla 39: Base64 decode bug fixes.
2025 Patch by Jakob Hirsch.
2027 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2029 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2031 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2033 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2039 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2040 ALT-N's libdkim that I have put here:
2041 http://duncanthrax.net/exim-experimental/
2043 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2044 were called 'true' and it seems that C99 defines that as a reserved
2045 keyword to be used with 'bool' variable types. That means you could
2046 not include C99-style headers which use bools without triggering
2047 build errors in sieve.c.
2049 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2050 as mailq or other aliases. Changed the --help handling significantly
2051 to do whats expected. exim_usage() emits usage/help information.
2053 SC/01 Added the -bylocaldomain option to eximstats.
2055 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2057 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2059 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2065 PH/01 Another patch from the Sieve maintainer.
2067 PH/02 When an IPv6 address is converted to a string for single-key lookup
2068 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2069 dots are used instead of colons so that keys in lsearch files need not
2070 contain colons. This was done some time before quoting was made available
2071 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2072 (notated using the quote facility) so as to distinguish them from IPv4
2073 keys. This meant that lookups for IP addresses in host lists did not work
2074 for iplsearch lookups.
2076 This has been fixed by arranging for IPv6 addresses to be expressed with
2077 colons if the lookup type is iplsearch. This is not incompatible, because
2078 previously such lookups could never work.
2080 The situation is now rather anomalous, since one *can* have colons in
2081 ordinary lsearch keys. However, making the change in all cases is
2082 incompatible and would probably break a number of configurations.
2084 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2087 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2088 conversion specification without a maximum field width, thereby enabling
2089 a rogue spamd server to cause a buffer overflow. While nobody in their
2090 right mind would setup Exim to query an untrusted spamd server, an
2091 attacker that gains access to a server running spamd could potentially
2092 exploit this vulnerability to run arbitrary code as the Exim user.
2094 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2095 $primary_hostname instead of what libspf2 thinks the hosts name is.
2097 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2098 a directory entry by the name of the lookup key. Previously, if a
2099 symlink pointed to a non-existing file or a file in a directory that
2100 Exim lacked permissions to read, a lookup for a key matching that
2101 symlink would fail. Now it is enough that a matching directory entry
2102 exists, symlink or not. (Bugzilla 503.)
2104 PH/03 The body_linecount and body_zerocount variables are now exported in the
2107 PH/04 Added the $dnslist_matched variable.
2109 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2110 This means they are set thereafter only if the connection becomes
2113 PH/06 Added the client_condition to authenticators so that some can be skipped
2114 by clients under certain conditions.
2116 PH/07 The error message for a badly-placed control=no_multiline_responses left
2117 "_responses" off the end of the name.
2119 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2121 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2122 (without spaces) instead of just copying the configuration text.
2124 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2126 PH/11 Added $max_received_linelength.
2128 PH/12 Added +ignore_defer and +include_defer to host lists.
2130 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2131 way in which PCRE > 7.0 is built.
2133 PH/14 Implemented queue_only_load_latch.
2135 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2136 MAIL command. The effect was to mangle the value on 64-bit systems.
2138 PH/16 Another patch from the Sieve maintainer.
2140 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2142 PH/18 If a system quota error occurred while trying to create the file for
2143 a maildir delivery, the message "Mailbox is full" was not appended to the
2144 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2145 only to a quota excession during the actual writing of the file.
2147 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2148 characters?) which causes problems in log lines. The DN values are now
2149 passed through string_printing() before being added to log lines.
2151 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2152 and InterBase are left for another time.)
2154 PH/21 Added message_body_newlines option.
2156 PH/22 Guard against possible overflow in moan_check_errorcopy().
2158 PH/23 POSIX allows open() to be a macro; guard against that.
2160 PH/24 If the recipient of an error message contained an @ in the local part
2161 (suitably quoted, of course), incorrect values were put in $domain and
2162 $local_part during the evaluation of errors_copy.
2168 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2169 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2172 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2173 issue a MAIL command.
2175 PH/02 In an ACL statement such as
2177 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2179 if a client was not listed at all, or was listed with a value other than
2180 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2181 the condition was not true (as it should be), so access was not denied.
2182 The bug was that the ! inversion was incorrectly passed on to the second
2183 item. This has been fixed.
2185 PH/03 Added additional dnslists conditions == and =& which are different from
2186 = and & when the dns lookup returns more than one IP address.
2188 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2189 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2191 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2192 FSYNC, which compiles an option called disable_fsync that allows for
2193 bypassing fsync(). The documentation is heavily laced with warnings.
2195 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2197 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2198 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2199 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2200 including adding "make clean"; (3) Added -fPIC when compiling the test
2201 dynamically loaded module, to get rid of a warning.
2203 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2204 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2205 = 0s. The bug is otherwise harmless.
2207 PH/07 There was a bug in the dovecot authenticator such that the value of
2208 $auth1 could be overwritten, and so not correctly preserved, after a
2209 successful authentication. This usually meant that the value preserved by
2210 the server_setid option was incorrect.
2212 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2214 PH/09 Installed PCRE release 7.0.
2216 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2217 run for batched SMTP input. It is now run at the start of every message
2218 in the batch. While fixing this I discovered that the process information
2219 (output by running exiwhat) was not always getting set for -bs and -bS
2220 input. This is fixed, and it now also says "batched" for BSMTP.
2222 PH/11 Added control=no_pipelining.
2224 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2225 patch, slightly modified), and move the expansion of helo_data till after
2226 the connection is made in the smtp transport (so it can use these
2229 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2231 PH/14 Added log_selector = +pid.
2233 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2235 PH/16 Add ${if forany and ${if forall.
2237 PH/17 Added dsn_from option to vary the From: line in DSNs.
2239 PH/18 Flush SMTP output before performing a callout, unless control =
2240 no_callout_flush is set.
2242 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2243 was true (the default) a successful delivery failed to delete the retry
2244 item, thus causing premature timeout of the address. The bug is now
2247 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2249 PH/21 Long custom messages for fakedefer and fakereject are now split up
2250 into multiline responses in the same way that messages for "deny" and
2251 other ACL rejections are.
2253 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2254 with slight modification.
2256 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2257 draft, changing the syntax and factoring some duplicate code".
2259 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2260 for deliveries of the second and subsequent messages over the same SMTP
2263 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2264 ${reduce, with only minor "tidies".
2266 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2268 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2269 expansion side effects.
2271 PH/27 When a message times out after an over-quota error from an Exim-imposed
2272 quota, the bounce message says "mailbox is full". This message was not
2273 being given when it was a system quota that was exceeded. It now should
2276 MH/03 Made $recipients available in local_scan(). local_scan() already has
2277 better access to the recipient list through recipients_list[], but
2278 $recipients can be useful in postmaster-provided expansion strings.
2280 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2281 in the case of a MAIL command with additional options following the
2282 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2283 were accidentally chopped off.
2285 PH/29 SMTP synchronization checks are implemented when a command is read -
2286 there is a check that no more input is waiting when there shouldn't be
2287 any. However, for some commands, a delay in an ACL can mean that it is
2288 some time before the response is written. In this time, more input might
2289 arrive, invalidly. So now there are extra checks after an ACL has run for
2290 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2291 pipelining has not been advertised.
2293 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2295 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2296 returned in the SMTP response when smtp_return_error_details was false.
2297 This has been fixed.
2299 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2300 instead of the C I/O that was originally supplied, because problems were
2301 reported on Solaris.
2303 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2304 Exim which did not show up earlier: it was assuming that a call to
2305 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2306 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2307 was a macro that became an assignment, so it seemed to work. This has
2308 changed to a proper function call with a void return, hence the compile
2309 error. Exim's code has been fixed.
2311 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2314 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2316 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2318 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2319 because it turns out to be too big or there is a timeout) there is no
2320 "Completed" line in the log. When some messages of this type were
2321 selected by exigrep, they were listed as "not completed". Others were
2322 picked up by some special patterns. I have improved the selection
2323 criteria to be more general.
2325 PH/38 The host_find_failed option in the manualroute router can now be set
2326 to "ignore", to completely ignore a host whose IP address cannot be
2327 found. If all hosts are ignored, the behaviour is controlled by the new
2328 host_all_ignored option.
2330 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2331 homing or because of multiple MX records with /mx) generated more than
2332 one IP address, and the following item turned out to be the local host,
2333 all the secondary addresses of the first item were incorrectly removed
2334 from the list, along with the local host and any following hosts (which
2335 is what is supposed to happen).
2337 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2338 whoever called Exim into the -H file. In the case of the daemon it was
2339 behaving confusingly. When first started, it used values for whoever
2340 started the daemon, but after a SIGHUP it used the Exim user (because it
2341 calls itself on a restart). I have changed the code so that it now always
2344 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2345 message are rejected with the same error (e.g. no authentication or bad
2346 sender address), and a DATA command is nevertheless sent (as can happen
2347 with PIPELINING or a stupid MUA), the error message that was given to the
2348 RCPT commands is included in the rejection of the DATA command. This is
2349 intended to be helpful for MUAs that show only the final error to their
2352 PH/42 Another patch from the Sieve maintainer.
2354 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2355 Eximstats - Fixed some broken HTML links and added missing column headers
2357 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2358 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2360 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2362 PH/43 Yet another patch from the Sieve maintainer.
2364 PH/44 I found a way to check for a TCP/IP connection going away before sending
2365 the response to the final '.' that terminates a message, but only in the
2366 case where the client has not sent further data following the '.'
2367 (unfortunately, this is allowed). However, in many cases there won't be
2368 any further data because there won't be any more messages to send. A call
2369 to select() can be used: if it shows that the input is "ready", there is
2370 either input waiting, or the socket has been closed. An attempt to read
2371 the next input character can distinguish the two cases. Previously, Exim
2372 would have sent an OK response which the client would never have see.
2373 This could lead to message repetition. This fix should cure that, at
2374 least in a lot of common cases.
2376 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2377 advertised in response to EHLO.
2383 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2384 fixed by 4.65/MH/01 (is this a record?) are fixed:
2386 (i) An empty string was always treated as zero by the numeric comparison
2387 operators. This behaviour has been restored.
2389 (ii) It is documented that the numeric comparison operators always treat
2390 their arguments as decimal numbers. This was broken in that numbers
2391 starting with 0 were being interpreted as octal.
2393 While fixing these problems I realized that there was another issue that
2394 hadn't been noticed. Values of message_size_limit (both the global option
2395 and the transport option) were treated as octal if they started with 0.
2396 The documentation was vague. These values are now always treated as
2397 decimal, and I will make that clear in the documentation.
2403 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2404 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2407 MH/01 Don't check that the operands of numeric comparison operators are
2408 integers when their expansion is in "skipping" mode (fixes bug
2409 introduced by 4.64-PH/07).
2411 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2412 child addresses, Exim now panics and dies. Previously, because the count
2413 is held in a short int, deliveries were likely to be lost. As such a
2414 large number of recipients for a single message is ridiculous
2415 (performance will be very, very poor), I have chosen to impose a limit
2416 rather than extend the field.
2422 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2423 leftover -K file (the existence of which was triggered by #402).
2424 While we were at it, introduced process PID as part of the -K
2425 filename. This should rule out race conditions when creating
2428 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2429 processing considerably. Previous code took too long for large mails,
2430 triggering a timeout which in turn triggers #401.
2432 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2433 in the DK code in transports.c. sendfile() is not really portable,
2434 hence the _LINUX specificness.
2436 TF/01 In the add_headers option to the mail command in an Exim filter,
2437 there was a bug that Exim would claim a syntax error in any
2438 header after the first one which had an odd number of characters
2441 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2442 callout verification, Exim cached a "reject" for the entire domain. This
2443 is correct for most verifications, but it is not correct for a recipient
2444 verification with use_sender or use_postmaster set, because in that case
2445 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2446 case of MAIL FROM:<> rejection from other early rejections (e.g.
2447 rejection of HELO). When verifying a recipient using a non-null MAIL
2448 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2449 Whatever the result of the callout, the value of the domain cache is
2450 left unchanged (for any other kind of callout, getting as far as trying
2451 RCPT means that the domain itself is ok).
2453 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2456 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2457 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2458 session, the daemon ignores ECONNECT errors and logs others; it now
2459 ignores EPIPE as well.
2461 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2462 (quoted-printable decoding).
2464 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2465 later the small subsequent patch to fix an introduced bug.
2467 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2469 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2471 PH/08 An error is now given if message_size_limit is specified negative.
2473 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2474 to be given (somewhat) arbitrary names.
2476 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2479 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2480 miscellaneous code fixes
2482 PH/10 Added the log_reject_target ACL modifier to specify where to log
2485 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2486 hostname. This is wrong, because it relates to the incoming message (and
2487 probably the interface on which it is arriving) and not to the outgoing
2488 callout (which could be using a different interface). This has been
2489 changed to use the value of the helo_data option from the smtp transport
2490 instead - this is what is used when a message is actually being sent. If
2491 there is no remote transport (possible with a router that sets up host
2492 addresses), $smtp_active_hostname is used.
2494 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2495 tweaks were necessary in order to get it to work (see also 21 below):
2496 (a) The code assumed that strncpy() returns a negative number on buffer
2497 overflow, which isn't the case. Replaced with Exim's string_format()
2499 (b) There were several signed/unsigned issues. I just did the minimum
2500 hacking in of casts. There is scope for a larger refactoring.
2501 (c) The code used strcasecmp() which is not a standard C function.
2502 Replaced with Exim's strcmpic() function.
2503 (d) The code set only $1; it now sets $auth1 as well.
2504 (e) A simple test gave the error "authentication client didn't specify
2505 service in request". It would seem that Dovecot has changed its
2506 interface. Fortunately there's a specification; I followed it and
2507 changed what the client sends and it appears to be working now.
2509 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2512 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2513 address A is aliased to B and C, where B exists and C does not. Without
2514 -v the output is "A verified" because verification stops after a
2515 successful redirection if more than one address is generated. However,
2516 with -v the child addresses are also verified. Exim was outputting "A
2517 failed to verify" and then showing the successful verification for C,
2518 with its parentage. It now outputs "B failed to verify", showing B's
2519 parentage before showing the successful verification of C.
2521 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2522 look up a TXT record in a specific list after matching in a combined
2525 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2526 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2527 they consult the DNS. I had assumed they would set it the way they
2528 wanted; and indeed my experiments on Linux seem to show that in some
2529 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2530 To be on the safe side, however, I have now made the interface to
2531 host_find_byname() similar to host_find_bydns(), with an argument
2532 containing the DNS resolver options. The host_find_byname() function now
2533 sets these options at its start, just as host_find_bydns() does. The smtp
2534 transport options dns_qualify_single and dns_search_parents are passed to
2535 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2536 of host_find_byname() use the default settings of RES_DEFNAMES
2537 (qualify_single) but not RES_DNSRCH (search_parents).
2539 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2540 spool_read_header() do less string testing, by means of a preliminary
2541 switch on the second character of optional "-foo" lines. (This is
2542 overdue, caused by the large number of possibilities that now exist.
2543 Originally there were few.) While I was there, I also converted the
2544 str(n)cmp tests so they don't re-test the leading "-" and the first
2545 character, in the hope this might squeeze out yet more improvement.
2547 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2548 flag allowing group syntax was set by the header_syntax check but not
2549 turned off, possible causing trouble later; (2) The flag was not being
2550 set at all for the header_verify test, causing "group"-style headers to
2551 be rejected. I have now set it in this case, and also caused header_
2552 verify to ignore an empty address taken from a group. While doing this, I
2553 came across some other cases where the code for allowing group syntax
2554 while scanning a header line wasn't quite right (mostly, not resetting
2555 the flag correctly in the right place). These bugs could have caused
2556 trouble for malformed header lines. I hope it is now all correct.
2558 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2559 with the "reply" argument non-NULL. The code, however (which originally
2560 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2561 but it didn't always do it. This confused somebody who was copying the
2562 code for some other use. I have removed all the tests.
2564 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2565 feature that was used to support insecure browsers during the U.S. crypto
2566 embargo. It requires special client support, and Exim is probably the
2567 only MTA that supported it -- and would never use it because real RSA is
2568 always available. This code has been removed, because it had the bad
2569 effect of slowing Exim down by computing (never used) parameters for the
2570 RSA_EXPORT functionality.
2572 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2573 authenticator to fail if there's a tab character in the incoming data
2574 (there should never be unless someone is messing about, as it's supposed
2575 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2576 if the connection is using TLS or if the remote IP is the same as the
2577 local IP, and the "valid-client-cert option" if a client certificate has
2580 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2581 authenticators. This can be used for authorization after authentication
2582 succeeds. (In the case of plaintext, it servers for both authentication
2585 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2586 if any retry times were supplied.
2588 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2589 connection, where there is no client IP address to check. In this
2590 situation, the verify now always succeeds.
2592 PH/25 Applied John Jetmore's -Mset patch.
2594 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2596 PH/27 In a string expansion for a processed (not raw) header when multiple
2597 headers of the same name were present, leading whitespace was being
2598 removed from all of them, but trailing whitespace was being removed only
2599 from the last one. Now trailing whitespace is removed from each header
2600 before concatenation. Completely empty headers in a concatenation (as
2601 before) are ignored.
2603 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2604 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2606 PH/29 [Removed. This was a change that I later backed out, and forgot to
2607 correct the ChangeLog entry (that I had efficiently created) before
2608 committing the later change.]
2610 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2611 address errors (4xx response to RCPT) over the same connection as other
2612 messages routed to the same hosts. Such deliveries are always "forced",
2613 so retry times are not inspected. This resulted in far too many retries
2614 for the affected addresses. The effect occurred only when there were more
2615 hosts than the hosts_max_try setting in the smtp transport when it had
2616 the 4xx errors. Those hosts that it had tried were not added to the list
2617 of hosts for which the message was waiting, so if all were tried, there
2618 was no problem. Two fixes have been applied:
2620 (i) If there are any address or message errors in an SMTP delivery, none
2621 of the hosts (tried or untried) are now added to the list of hosts
2622 for which the message is waiting, so the message should not be a
2623 candidate for sending over the same connection that was used for a
2624 successful delivery of some other message. This seems entirely
2625 reasonable: after all the message is NOT "waiting for some host".
2626 This is so "obvious" that I'm not sure why it wasn't done
2627 previously. Hope I haven't missed anything, but it can't do any
2628 harm, as the worst effect is to miss an optimization.
2630 (ii) If, despite (i), such a delivery is accidentally attempted, the
2631 routing retry time is respected, so at least it doesn't keep
2632 hammering the server.
2634 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2635 in ${readsocket because some servers need this prod.
2637 PH/32 Added some extra debug output when updating a wait-xxx database.
2639 PH/33 The hint "could be header name not terminated by colon", which has been
2640 given for certain expansion errors for a long time, was not being given
2641 for the ${if def:h_colon_omitted{... case.
2643 PH/34 The spec says: "With one important exception, whenever a domain list is
2644 being scanned, $domain contains the subject domain." There was at least
2645 one case where this was not true.
2647 PH/35 The error "getsockname() failed: connection reset by peer" was being
2648 written to the panic log as well as the main log, but it isn't really
2649 panic-worthy as it just means the connection died rather early on. I have
2650 removed the panic log writing for the ECONNRESET error when getsockname()
2653 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2654 runs only) independently of the message's sender address. This meant
2655 that, if the 4xx error was in fact related to the sender, a different
2656 message to the same recipient with a different sender could confuse
2657 things. In particular, this can happen when sending to a greylisting
2658 server, but other circumstances could also provoke similar problems.
2659 I have changed the default so that the retry time for these errors is now
2660 based a combination of the sender and recipient addresses. This change
2661 can be overridden by setting address_retry_include_sender=false in the
2664 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2665 remote server are returned as part of bounce messages. This was not
2666 happening for LMTP over a pipe (the lmtp transport), but now it is the
2667 same for both kinds of LMTP.
2669 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2670 in header lines that were in fact CNAMEs. This is no longer the case.
2672 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2673 and queue runs started by the daemon processed all messages. This has
2674 been fixed so that -R and -S can now usefully be given with -q<time>.
2676 PH/40 Import PCRE release 6.7 (fixes some bugs).
2678 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2680 PH/42 Give an error if -q is specified more than once.
2682 PH/43 Renamed the variables $interface_address and $interface_port as
2683 $received_ip_address and $received_port, to make it clear that these
2684 values apply to message reception, and not to the outgoing interface when
2685 a message is delivered. (The old names remain recognized, of course.)
2687 PH/44 There was no timeout on the connect() call when using a Unix domain
2688 socket in the ${readsocket expansion. There now is.
2690 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2691 be meaningful with "accept".
2693 SC/01 Eximstats V1.43
2694 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2696 SC/02 Eximstats V1.44
2697 Use a glob alias rather than an array ref in the generated
2698 parser. This improves both readability and performance.
2700 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2701 Collect SpamAssassin and rejection statistics.
2702 Don't display local sender or destination tables unless
2703 there is data to show.
2704 Added average volumes into the top table text output.
2706 SC/04 Eximstats V1.46
2707 Collect data on the number of addresses (recipients)
2708 as well as the number of messages.
2710 SC/05 Eximstats V1.47
2711 Added 'Message too big' to the list of mail rejection
2712 reasons (thanks to Marco Gaiarin).
2714 SC/06 Eximstats V1.48
2715 Mainlog lines which have GMT offsets and are too short to
2716 have a flag are now skipped.
2718 SC/07 Eximstats V1.49 (Alain Williams)
2719 Added the -emptyok flag.
2721 SC/08 Eximstats V1.50
2722 Fixes for obtaining the IP address from reject messages.
2724 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2725 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2726 whitespace changes from 4.64-PH/27
2728 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2731 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2732 are found, allow negative numbers in numeric criteria)
2734 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2736 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2737 to match changes made in 4.64-PH/43
2739 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2741 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2742 "domain" is actually an IP address, and force a failure. This locks out
2743 those revolvers/nameservers that support "A-for-A" lookups, in
2744 contravention of the specifications.
2746 PH/48 When a host name was looked up from an IP address, and the subsequent
2747 forward lookup of the name timed out, the host name was left in
2748 $sender_host_name, contrary to the specification.
2750 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2751 restricted to single-key lookups, Exim was not diagnosing an error if
2752 * or *@ was used with a query-style lookup.
2754 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2756 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2757 long ago, but noone interested enough thought of it. Let's just say that
2758 the "1.1" means that there are some new functions that weren't there at
2759 some point in the past.
2761 PH/51 Error processing for expansion failure of helo_data from an smtp
2762 transport during callout processing was broken.
2764 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2765 tested/used via the -bh/-bhc/-bs options.
2767 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2768 bug, fixed in subsequent PCRE releases).
2770 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2771 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2773 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2779 SC/01 Use a glob alias rather than an array ref in eximstats generated
2780 parser. This improves both readability and performance.
2782 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2783 Don't display local sender or destination tables in eximstats unless
2784 there is data to show.
2785 Added average volumes into the eximstats top table text output.
2787 SC/03 Collect data on the number of addresses (recipients) as well
2788 as the number of messages in eximstats.
2790 TF/01 Correct an error in the documentation for the redirect router. Exim
2791 does (usually) call initgroups() when daemonizing.
2793 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2794 with consistent privilege compared to when running as a daemon.
2796 TF/03 Note in the spec that $authenticated_id is not set for local
2797 submissions from trusted users.
2799 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2800 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2802 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2803 by adding some example configuration directives to the default
2804 configuration file. A little bit of work is required to uncomment the
2805 directives and define how usernames and passwords are checked, but
2806 there is now a framework to start from.
2808 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2809 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2810 without this. I don't know how relevant this is to other LDAP libraries.
2812 PH/02 Add the verb name to the "unknown ACL verb" error.
2814 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2816 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2818 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2819 directory not expanded when it should be if an expanded home directory
2820 was set for the address (which is overridden by the transport).
2822 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2825 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2826 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2827 because it is too late at that time, and has no effect.
2829 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2830 security issue with \' (bugzilla #107). I could not use the
2831 PQescapeStringConn() function, because it needs a PGconn value as one of
2834 PH/08 When testing addresses using -bt, indicate those final addresses that
2835 are duplicates that would not cause an additional delivery. At least one
2836 person was confused, thinking that -bt output corresponded to deliveries.
2837 (Suppressing duplicates isn't a good idea as you lose the information
2838 about possibly different redirections that led to the duplicates.)
2840 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2841 systems where poll() doesn't work, in particular OS X.
2843 PH/10 Added more information to debugging output for retry time not reached.
2845 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2846 operations in malware.c.
2848 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2851 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2852 syslog_duplication set false, log lines that would normally be written
2853 both the the main log and to the reject log were not written to syslog at
2856 PH/14 In the default configuration, change the use of "message" in ACL warn
2857 statements to "add_header".
2859 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2860 not followed by a command (e.g. "seen endif").
2862 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2863 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2866 PH/17 Added extra conditions to the default value of delay_warning_condition
2870 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2871 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2872 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2875 The Auto-Submitted: and various List- headers are standardised, whereas I
2876 don't think Precedence: ever was.
2878 PH/18 Refactored debugging code in route_finduser() to show more information,
2879 in particular, the error code if getpwnam() issues one.
2881 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2882 This is apparently needed in addition to the PH/07 change above to avoid
2883 any possible encoding problems.
2885 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2886 but not after initializing Perl.
2888 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2889 output them only if debugging. By default they are written stderr,
2890 apparently, which is not desirable.
2892 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2895 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2898 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2900 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2901 authenticated or an ident call has been made. Suppress the default
2902 values for $authenticated_id and $authenticated_sender (but permit -oMai
2903 and -oMas) when testing with -bh.
2905 PH/24 Re-jigged the order of the tests in the default configuration so that the
2906 tests for valid domains and recipients precede the DNS black list and CSA
2907 tests, on the grounds that those ones are more expensive.
2909 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2910 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2911 This bug exists in every version of Exim that I still have, right back to
2914 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2915 However, an attempt to turn on case-sensitivity in a regex key by
2916 including (?-i) didn't work because the subject string was already
2917 lowercased, and the effects were non-intuitive. It turns out that a
2918 one-line patch can be used to allow (?-i) to work as expected.
2924 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2925 other effects) broke the use of negated acl sub-conditions.
2927 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2930 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2931 "Deny" causes Exim to reject the incoming connection with a 554 error.
2932 Unfortunately, if there is a major crisis, such as a disk failure,
2933 tcp-wrappers gives "deny", whereas what one would like would be some
2934 kind of temporary error. A kludge has been added to help with this.
2935 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2936 554 error is used if errno is still zero or contains ENOENT (which occurs
2937 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2940 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2942 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2943 errors. Otherwise a message that provokes a temporary error (when other
2944 messages do not) can cause a whole host to time out.
2946 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2947 addresses were routed directly to files or pipes from a redirect router.
2948 File deliveries just didn't batch; pipe deliveries might have suffered
2951 PH/06 A failure to get a lock for a hints database would erroneously always say
2952 "Failed to get write lock", even when it was really a read lock.
2954 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2955 of 0600. This has been changed to use the value of the lockfile_mode
2956 option (which defaults to 0600).
2958 PH/08 Applied small patch from the Sieve maintainer.
2960 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2961 folder from quota calculations, a direct delivery into this folder messed
2962 up the contents of the maildirsize file. This was because the regex was
2963 used only to exclude .Trash (or whatever) when the size of the mailbox
2964 was calculated. There was no check that a delivery was happening into an
2965 excluded directory. This bug has been fixed by ignoring all quota
2966 processing for deliveries into excluded directories.
2968 PH/10 Added the maildirfolder_create_regex option to appendfile.
2974 PH/01 The code for finding all the local interface addresses on a FreeBSD
2975 system running IPv6 was broken. This may well have applied to all BSD
2976 systems, as well as to others that have similar system calls. The broken
2977 code found IPv4 interfaces correctly, but gave incorrect values for the
2978 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2979 that it would not match correctly against @[] and not recognize the IPv6
2982 PH/02 The ipliteral router was not recognizing addresses of the form user@
2983 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2985 PH/03 Added disable_ipv6.
2987 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2988 decoded form, because it is most often used to construct To: headers
2989 lines in autoreplies, and the decoded form may well be syntactically
2990 invalid. However, $reply_address has leading white space removed, and all
2991 newlines turned into spaces so that the autoreply transport does not
2994 PH/05 If group was specified without a user on a router, and no group or user
2995 was specified on a transport, the group from the router was ignored.
2997 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2998 for visible compile-time settings that can be used to change these
2999 numbers, for those that want even more. Backwards compatibility with old
3000 spool files has been maintained. However, going back to a previous Exim
3001 release will lost any variables that are in spool files.
3003 PH/07 Two small changes when running in the test harness: increase delay when
3004 passing a TCP/IP connection to a new process, in case the original
3005 process has to generate a bounce, and remove special handling of
3006 127.0.0.2 (sic), which is no longer necessary.
3008 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3009 be the same on different OS.
3011 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3014 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3015 whether --show-vars was specified or not
3017 JJ/02 exipick: Added support for new ACL variable spool format introduced
3020 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3021 syntactically invalid From: or Reply-to: line, and a filter used this to
3022 generate an autoreply, and therefore failed to obtain an address for the
3023 autoreply, Exim could try to deliver to a non-existent relative file
3024 name, causing unrelated and misleading errors. What now happens is that
3025 it logs this as a hard delivery error, but does not attempt to create a
3028 PH/11 The exinext utility has a -C option for testing purposes, but although
3029 the given file was scanned by exinext itself; it wasn't being passed on
3030 when Exim was called.
3032 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3033 an end-of-file indication when reading a command response.
3035 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3036 compiled. In many other places in Exim, IPv6 addresses are always
3037 recognized, so I have changed this. It also means that IPv4 domain
3038 literals of the form [IPV4:n.n.n.n] are now always recognized.
3040 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3041 used if the router is not running as root, for example, when verifying at
3042 ACL time, or when using -bh. The debugging output from this situation was
3043 non-existent - all you got was a failure to exec. I have made two
3046 (a) Failures to set uid/gid, the current directory, or a process leader
3047 in a subprocess such as that created by queryprogram now generate
3048 suitable debugging output when -d is set.
3050 (b) The queryprogram router detects when it is not running as root,
3051 outputs suitable debugging information if -d is set, and then runs
3052 the subprocess without attempting to change uid/gid.
3054 PH/15 Minor change to Makefile for building test_host (undocumented testing
3057 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3058 additional section of a DNS packet that returns MX or SRV records.
3059 Instead, it always explicitly searches for A/AAAA records. This avoids
3060 major problems that occur when a DNS server includes only records of one
3061 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3062 fixed another bug: if SRV records were looked up and the corresponding
3063 address records were *not* found in the additional section, the port
3064 values from the SRV records were lost.
3066 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3067 using the correct key (the original address) when searching the retry
3068 rules in order to find which one to use for generating the retry hint.
3070 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3071 adding the default one. Similarly, if it contains a Reply-To: header, the
3072 errors_reply_to option, if set, is not used.
3074 PH/19 When calculating a retry time, Exim used to measure the "time since
3075 failure" by looking at the "first failed" field in the retry record. Now
3076 it does not use this if it is later than than the arrival time of the
3077 message. Instead it uses the arrival time. This makes for better
3078 behaviour in cases where some deliveries succeed, thus re-setting the
3079 "first failed" field. An example is a quota failure for a huge message
3080 when small messages continue to be delivered. Without this change, the
3081 "time since failure" will always be short, possible causing more frequent
3082 delivery attempts for the huge message than are intended.
3083 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3085 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3086 $1, $2, $3) because the numerical variables can be reset during some
3087 expansion items (e.g. "match"), thereby losing the authentication data.
3089 PH/21 Make -bV show the size of off_t variables so that the test suite can
3090 decide whether to run tests for quotas > 2G.
3092 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3093 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3094 filecount value is greater than 2G or if a quota value is greater than 2G
3095 on a system where the size of off_t is not greater than 4, a panic error
3098 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3099 never match. The debug and -bh output now contains an explicit error
3100 message indicating a malformed IPv4 address or mask.
3102 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3103 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3104 PH/23 above applies.
3106 PH/25 Do not write to syslog when running in the test harness. The only
3107 occasion when this arises is a failure to open the main or panic logs
3108 (for which there is an explicit test).
3110 PH/26 Added the /no_tell option to "control=freeze".
3112 PH/27 If a host name lookup failed very early in a connection, for example, if
3113 the IP address matched host_lookup and the reverse lookup yielded a name
3114 that did not have a forward lookup, an error message of the form "no IP
3115 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3116 could be logged. Now it outputs the IP address instead of "NULL".
3118 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3119 allows the sender and the authenticated sender to be set when
3120 submitting a message from within Exim. Since child_open_exim() is
3121 documented for local_scan(), the new function should be too.
3123 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3124 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3125 results in an empty string is now treated as unset.
3127 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3129 PH/31 Added sender_verify_fail logging option.
3131 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3132 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3133 tidied the source and removed it altogether.
3135 PH/33 When a queue run was abandoned because the load average was too high, a
3136 log line was always written; now it is written only if the queue_run log
3137 selector is set. In addition, the log line for abandonment now contains
3138 information about the queue run such as the pid. This is always present
3139 in "start" and "stop" lines but was omitted from the "abandon" line.
3141 PH/34 Omit spaces between a header name and the colon in the error message that
3142 is given when verify = headers_syntax fails (if there are lots of them,
3143 the message gets confusing).
3145 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3146 names, as there are now some PTR records that contain slashes. This check
3147 is only to protect against broken name servers that fall over on strange
3148 characters, so the fact that it applies to all lookups doesn't matter.
3150 PH/36 Now that the new test suite is complete, we can remove some of the
3151 special code in Exim that was needed for the old test suite. For example,
3152 sorting DNS records because real resolvers return them in an arbitrary
3153 order. The new test suite's fake resolver always returns records in the
3156 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3157 bounces) except when queue_only is set, to avoid logging races between
3158 the different processes.
3160 PH/38 Panic-die if .include specifies a non-absolute path.
3162 PH/39 A tweak to the "H" retry rule from its user.
3164 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3165 a label. They prevented compilation on older perls.
3167 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3168 a warning to be raised on newish perls.
3170 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3171 on queue. Changes to match documented behaviour of showing count of
3172 messages matching specified criteria.
3174 PH/40 Changed the default ident timeout from 30s to 5s.
3176 PH/41 Added support for the use of login_cap features, on those BSD systems
3177 that have them, for controlling the resources used by pipe deliveries.
3179 PH/42 The content-scanning code uses fopen() to create files in which to put
3180 message data. Previously it was not paying any attention to the mode of
3181 the files. Exim runs with umask(0) because the rest of the code creates
3182 files with open(), and sets the required mode explicitly. Thus, these
3183 files were ending up world-writeable. This was not a big issue, because,
3184 being within the spool directory, they were not world-accessible. I have
3185 created a function called modefopen, which takes an additional mode
3186 argument. It sets umask(777), creates the file, chmods it to the required
3187 mode, then resets the umask. All the relevant calls to fopen() in the
3188 content scanning code have been changed to use this function.
3190 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3191 to 24 hours. This avoids potential overflow problems when processing G
3192 and H retry rules. I suspect nobody ever tinkers with this value.
3194 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3196 PH/45 When the plaintext authenticator is running as a client, the server's
3197 challenges are checked to ensure they are valid base64 strings. By
3198 default, the authentication attempt is cancelled if an invalid string is
3199 received. Setting client_ignore_invalid_base64 true ignores these errors.
3200 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3201 they are received. Thus, the responses can be made to depend on the
3202 challenges. If an invalid string is ignored, an empty string is placed in
3205 PH/46 Messages that are created by the autoreply transport now contains a
3206 References: header, in accordance with RFCs 2822 and 3834.
3208 PH/47 Added authenticated_sender_force to the smtp transport.
3210 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3212 PH/49 Installed latest patch from the Sieve maintainer.
3214 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3215 was also set, the appendfile transport was unnecessarily scanning a
3216 directory of message files (e.g. for maildir delivery) to find the count
3217 of files (along with the size), even though it did not need this
3218 information. It now does the scan only if it needs to find either the
3219 size of the count of files.
3221 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3223 PH/52 Two bugs concerned with error handling when the smtp transport is
3226 (i) Exim was not creating retry information for temporary errors given
3227 for individual recipients after the DATA command when the smtp transport
3228 was used in LMTP mode. This meant that they could be retried too
3229 frequently, and not timed out correctly.
3231 (ii) Exim was setting the flag that allows error details to be returned
3232 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3233 recipients that were returned after the DATA command.
3235 PH/53 This is related to PH/52, but is more general: for any failing address,
3236 when detailed error information was permitted to be returned to the
3237 sender, but the error was temporary, then after the final timeout, only
3238 "retry timeout exceeded" was returned. Now it returns the full error as
3239 well as "retry timeout exceeded".
3241 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3242 do this, and (what is worse) MTAs that accept it.
3244 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3245 will now be deprecated.
3247 PH/56 New os.c-cygwin from the Cygwin maintainer.
3249 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3250 formats (previously only available in exim formats via -bpr, -bpru,
3251 and -bpra. Now also available in native and exiqgrep formats)
3253 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3254 with very large, slow to parse queues
3256 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3258 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3260 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3261 responses to authentication challenges, though it was showing the
3262 challenges; (ii) I've removed the CR characters from the debug output for
3265 PH/58 Allow for the insertion of a newline as well as a space when a string
3266 is turned into more than one encoded-word during RFC 2047 encoding. The
3267 Sieve code now uses this.
3269 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3270 data_4xx, lost_connection, tls_required.
3272 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3273 message was being sent as an SMTP response.
3275 PH/61 Add -l and -k options to exicyclog.
3277 PH/62 When verifying, if an address was redirected to one new address, so that
3278 verification continued, and the new address failed or deferred after
3279 having set something in $address_data, the value of $address_data was not
3280 passed back to the ACL. This was different to the case when no
3281 redirection occurred. The value is now passed back in both cases.
3283 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3284 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3285 use login_cap.h, so on its own it isn't the distinguishing feature. The
3286 new name refers directly to the setclassresources() function.
3288 PH/65 Added configuration files for NetBSD3.
3290 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3292 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3293 is preferred over IPv4.
3295 PH/68 The bounce_return_message and bounce_return_body options were not being
3296 honoured for bounces generated during the reception of non-SMTP messages.
3297 In particular, this applied to messages rejected by the ACL. This bug has
3298 been fixed. However, if bounce_return_message is true and bounce_return_
3299 body is false, the headers that are returned for a non-SMTP message
3300 include only those that have been read before the error was detected.
3301 (In the case of an ACL rejection, they have all been read.)
3303 PH/69 The HTML version of the specification is now built in a directory called
3304 spec_html instead of spec.html, because the latter looks like a path with
3305 a MIME-type, and this confuses some software.
3307 PH/70 Catch two compiler warnings in sieve.c.
3309 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3310 function verify_get_ident() calls ip_connect() to connect a socket, but
3311 if the "connect()" function timed out, ip_connect() used to close the
3312 socket. However, verify_get_ident() also closes the socket later, and in
3313 between Exim writes to the log, which may get opened at this point. When
3314 the socket was closed in ip_connect(), the log could get the same file
3315 descriptor number as the socket. This naturally causes chaos. The fix is
3316 not to close the socket in ip_connect(); the socket should be closed by
3317 the function that creates it. There was only one place in the code where
3318 this was missing, in the iplookup router, which I don't think anybody now
3319 uses, but I've fixed it anyway.
3321 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3322 well as to direct DNS lookups. Otherwise the handling of names in host
3323 lists is inconsistent and therefore confusing.
3329 PH/01 Two changes to the default runtime configuration:
3331 (1) Move the checks for relay_from_hosts and authenticated clients from
3332 after to before the (commented out) DNS black list checks.
3334 (2) Add control=submission to the relay_from_hosts and authenticated
3335 clients checks, on the grounds that messages accepted by these
3336 statements are most likely to be submissions.
3338 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3340 (1) Generate an error if the third argument for the ${prvs expansion is
3343 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3346 (3) Reset the variables that are obtained from the first argument of
3347 ${prvscheck and used in the second argument before leaving the code,
3348 because their memory is reclaimed, so using them afterwards may do
3351 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3352 one (it's much easier than Tom thought :-).
3354 (5) Because of (4), we can now allow for the use of $prvscheck_result
3355 inside the third argument.
3357 PH/03 For some reason, the default setting of PATH when running a command from
3358 a pipe transport was just "/usr/bin". I have changed it to
3361 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3362 anything to be listed in the output from -bV.
3364 PH/05 When a filter generated an autoreply, the entire To: header line was
3365 quoted in the delivery log line, like this:
3367 => >A.N.Other <ano@some.domain> <original@ddress> ...
3369 This has been changed so that it extracts the operative address. There
3370 may be more than one such address. If so, they are comma-separated, like
3373 => >ano@some.domain,ona@other.domain <original@ddress> ...
3375 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3376 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3377 not being looked up in the rDNS to get a host name, Exim was showing the
3378 IP address twice in Received: lines, even though the IP addresses were
3379 identical. For example:
3381 Received: from [1.2.3.4] (helo=[1.2.3.4])
3383 However, if the real host name was known, it was omitting the HELO data
3384 if it matched the actual IP address. This has been tidied up so that it
3385 doesn't show the same IP address twice.
3387 PH/07 When both +timestamp and +memory debugging was on, the value given by
3388 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3389 called by the debug printing, thereby overwriting the timestamp buffer.
3390 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3392 PH/08 When the original message was included in an autoreply transport, it
3393 always said "this is a copy of the message, including all the headers",
3394 even if body_only or headers_only was set. It now gives an appropriate
3397 PH/09 Applied a patch from the Sieve maintainer which:
3399 o fixes some comments
3400 o adds the (disabled) notify extension core
3401 o adds some debug output for the result of if/elsif tests
3402 o points to the current vacation draft in the documentation
3403 and documents the missing references header update
3407 o fixes a bug in processing the envelope test (when testing
3408 multiple envelope elements, the last element determined the
3411 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3412 Electronic Mail") by including:
3414 Auto-submitted: auto-generated
3416 in the messages that it generates (bounce messages and others, such as
3417 warnings). In the case of bounce messages for non-SMTP messages, there was
3418 also a typo: it was using "Auto_submitted" (underscore instead of
3419 hyphen). Since every message generated by Exim is necessarily in response
3420 to another message, thes have all been changed to:
3422 Auto-Submitted: auto-replied
3424 in accordance with these statements in the RFC:
3426 The auto-replied keyword:
3428 - SHOULD be used on messages sent in direct response to another
3429 message by an automatic process,
3431 - MUST NOT be used on manually-generated messages,
3433 - MAY be used on Delivery Status Notifications (DSNs) and Message
3434 Disposition Notifications (MDNs),
3436 - MUST NOT be used on messages generated by automatic or periodic
3437 processes, except for messages which are automatic responses to
3440 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3441 to the default Received: header definition.
3443 PH/12 Added log selector acl_warn_skipped (default on).
3445 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3446 variables because (a) they are in the wrong storage pool and (b) even if
3447 they were copied, it wouldn't work properly because of the caching.
3449 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3450 checking when decoding. Apparently there are clients that generate
3451 overlong encoded strings. Why am I not surprised?
3453 PH/15 If the first argument of "${if match_address" was not empty, but did not
3454 contain an "@" character, Exim crashed. Now it writes a panic log message
3455 and treats the condition as false.
3457 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3459 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3460 extension "envelope-auth". The code is finished and in agreement with
3461 other implementations, but there is no documentation so far and in fact,
3462 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3463 not changing the active code.
3465 Print executed "if" and "elsif" statements when debugging is used. This
3466 helps a great deal to understand what a filter does.
3468 Document more things not specified clearly in RFC3028. I had all this
3469 sorted out, when out of a sudden new issues came to my mind. Oops."
3471 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3474 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3475 canonical form (as documented). However, after a host name lookup from
3476 the IP address, check_host() was doing a simple string comparison with
3477 addresses acquired from the DNS when checking that the found name did
3478 have the original IP as one of its addresses. Since any found IPv6
3479 addresses are likely to be in abbreviated form, the comparison could
3480 fail. Luckily, there already exists a function for doing the comparison
3481 by converting both addresses to binary, so now that is used instead of
3482 the text comparison.
3484 PH/20 There was another similar case to PH/19, when a complete host name was
3485 given in a host list; looking up its IP address could give an abbreviated
3486 form, whereas the current host's name might or might not be abbreviated.
3487 The same fix has been applied.
3493 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3494 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3497 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3498 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3500 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3502 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3503 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3504 Also, for "Auto-Submitted", treat anything other than "no" as
3505 non-personal, in accordance with RFC 3834. (Previously it treated
3506 anything starting "auto-" as non-personal.)
3508 TF/01 The control=submission/name=... option had a problem with syntax
3509 errors if the name included a slash character. The /name= option
3510 now slurps the rest of the string, so it can include any characters
3511 but it must come last in the list of options (after /sender_retain
3514 PH/05 Some modifications to the interface to the fake nameserver for the new
3522 TK/01 Added the "success_on_redirect" address verification option. See
3523 NewStuff for rationale and an example.
3525 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3527 PH/02 Patch to exigrep to allow it to work on syslog lines.
3529 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3530 fread() to skip over the body file's header line, because in Cygwin the
3531 header line is locked and is inaccessible.
3533 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3534 co-exist for some time) to make it clear that it is the Exim ID that is
3535 referenced, not the Message-ID: header line.
3537 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3538 string_format() function, because snprintf() does not exist on all
3541 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3542 ${stat: expansion item.
3544 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3545 protocol synchronization error", to keep the pedants happy.
3547 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3548 well as for IRIX systems, when gcc is being used. See the host.c source
3551 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3553 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3556 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3557 transport and to the smtp transport in LMTP mode.
3559 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3561 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3562 run a filter in a subprocess. This could lead to confusion in subsequent
3563 lookups in the parent process. There should also be a search_tidyup() at
3564 the end of the subprocess.
3566 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3567 only if the host matched helo_try_verify_hosts, which caused the
3568 verification to occur when the EHLO/HELO command was issued. The ACL just
3569 tested the remembered result. Now, if a previous verification attempt has
3570 not happened, "verify = helo" does it there and then.
3572 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3574 TK/03 Fix log output including CR from clamd.
3576 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3577 exist provoked a memory error which could cause a segfault.
3579 PH/15 Installed PCRE 6.2
3581 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3583 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3584 of the problem. Specifically, suggested +O2 rather than +O1 for the
3587 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3589 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3590 record, the port was not correctly logged when the outgoing_port log
3591 selector was set (it logged the transort's default port).
3593 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3594 fallback_hosts, and "hosts" in the smtp transport.
3596 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3597 host errors such as "Connection refused".
3599 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3600 authentication with radiusclient 0.4.9:
3602 - Error returned from rc_read_config was caught wrongly
3603 - Username/password not passed on to radius server due to wrong length.
3605 The presumption is that some radiusclient API changes for 4.51/PH/17
3606 were not taken care of correctly. The code is still untested by me (my
3607 Linux distribution still has 0.3.2 of radiusclient), but it was
3608 contributed by a Radius user.
3610 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3611 expanding the "port" option of the smtp transport.
3613 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3614 while reading a MIME header. Thanks to Tom Hughes for a patch.
3616 PH/24 Include config.h inside local_scan.h so that configuration settings are
3619 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3620 that in an ACL for RCPT (for example), you can examine exactly what was
3623 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3624 commands, but it was not correctly comparing the address with the actual
3625 client host address. Thus, it would show the EHLO address in Received:
3626 header lines when this was not necessary.
3628 PH/27 Added the % operator to ${eval:}.
3630 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3631 it should be ignoring failures (because with -C, for example, it has lost
3632 privilege). It wasn't ignoring creation failures other than "already
3635 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3638 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3639 because an input line was too long, either on its own, or by virtue of
3640 too many continuations, the temporary file was not being removed, and the
3641 return code was incorrect.
3643 PH/31 Missing "BOOL" in function definition in filtertest.c.
3645 PH/32 Applied Sieve patches from the maintainer.
3647 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3649 PH/33 Added "verify = not_blind".
3651 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3652 Local/Makefile (with some defaults set). These are used in built scripts
3653 such as exicyclog, but they have never been used in the exim_install
3654 script (though there are many overriding facilities there). I have
3655 arranged that the exim_install script now takes note of these two
3658 PH/35 Installed configuration files for Dragonfly.
3660 PH/36 When a locally submitted message by a trusted user did not contain a
3661 From: header, and the sender address was obtained from -f or from an SMTP
3662 MAIL command, and the trusted user did not use -F to supply a sender
3663 name, $originator_name was incorrectly used when constructing a From:
3664 header. Furthermore, $originator_name was used for submission mode
3665 messages from external hosts without From: headers in a similar way,
3666 which is clearly wrong.
3668 PH/37 Added control=suppress_local_fixups.
3670 PH/38 When log_selector = +received_sender was set, and the addition of the
3671 sender made the log line's construction buffer exactly full, or one byte
3672 less than full, an overflow happened when the terminating "\n" was
3675 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3676 when the result of a list match is failure because a DNS lookup failed.
3678 PH/40 RM_COMMAND is now used in the building process.
3680 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3681 the "build-* directories that it finds.
3683 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3684 address in a domain literal was a prefix of an interface address.
3686 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3687 when verifying a sender address, unless rewrite_headers is false.
3689 PH/44 Wrote a long comment about why errors_to addresses are verified as
3690 recipients, not senders.
3692 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3693 the ratelimit ACL was added.
3695 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3697 PH/46 Added extra information about PostgreSQL errors to the error string.
3699 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3700 avoiding the need to install special zones in a real server. This is
3701 backwards compatible; if it can't find the fake resolver, it drops back.
3702 Thus, both old and new test suites can be run.
3704 TF/02 Added util/ratelimit.pl
3706 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3707 clock is set back in time.
3709 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3710 Candler <B.Candler@pobox.com>.
3712 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3713 OK for addresses that are the result of redirections.
3715 PH/48 A number of further additions for the benefit of the new test suite,
3716 including a fake gethostbyname() that interfaces to the fake DNS resolver
3719 TF/06 The fix for widen_domains has also been applied to qualify_single and
3720 search_parents which are the other dnslookup options that can cause
3723 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3726 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3728 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3729 service" if it encountered an MX record with an empty target hostname.
3730 The message is now "an MX or SRV record indicated no SMTP service".
3732 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3733 if the DNS of the sending site is misconfigured. This is quite a
3734 common situation. This change restores the behaviour of treating a
3735 helo verification defer as a failure.
3737 PH/51 If self=fail was set on a router, the bounce message did not include the
3738 actual error message.
3744 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3746 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3747 command itself ended in error, the underlying message about the transport
3748 filter timeout was being overwritten with the pipe command error. Now the
3749 underlying error message should be appended to the second error message.
3751 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3753 PH/02 Reduce the amount of output that "make" produces by default. Full output
3754 can still be requested.
3756 PH/03 The warning log line about a condition test deferring for a "warn" verb
3757 was being output only once per connection, rather than after each
3758 occurrence (because it was using the same function as for successful
3759 "warn" verbs). This seems wrong, so I have changed it.
3761 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3762 it should not have, which might have caused a crash in the right
3763 circumstances, but probably never did.
3765 PH/04 Installed a modified version of Tony Finch's patch to make submission
3766 mode fix the return path as well as the Sender: header line, and to
3767 add a /name= option so that you can make the user's friendly name appear
3770 TF/03 Added the control = fakedefer ACL modifier.
3772 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3773 Mark Lowes for thorough testing.
3775 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3777 TK/03 Merged latest SRS patch from Miles Wilton.
3779 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3780 with the definition in sysexits.h (which is #included earlier).
3781 Fortunately, Exim does not actually use EX_OK. The code used to try to
3782 preserve the sysexits.h value, by assuming that macro definitions were
3783 scanned for macro replacements. I have been disabused of this notion,
3784 so now the code just undefines EX_OK before #including unistd.h.
3786 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3787 in the smtp transport. When a block could not be written in a single
3788 write() function, the timeout was being re-applied to each part-write.
3789 This seems wrong - if the receiver was accepting one byte at a time it
3790 would take for ever. The timeout is now adjusted when this happens. It
3791 doesn't have to be particularly precise.
3793 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3794 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3796 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3797 without a domain if the check to <postmaster@domain> fails.
3799 SC/01 Eximstats: added -xls and the ability to specify output files
3800 (patch written by Frank Heydlauf).
3802 SC/02 Eximstats: use FileHandles for outputting results.
3804 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3806 SC/04 Eximstats: fixed display of large numbers with -nvr option
3808 SC/05 Eximstats: fixed merging of reports with empty tables.
3810 SC/06 Eximstats: added the -include_original_destination flag
3812 SC/07 Eximstats: removed tabs and trailing whitespace.
3814 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3816 TK/06 MBOX spool code: Add real "From " MBOX separator line
3817 so the .eml file is really in mbox format (even though
3818 most programs do not really care). Patch from Alex Miller.
3820 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3821 The latter is generated from $received_to and is only set if the
3822 message has one envelope recipient. SA can use these headers,
3823 obviously out-of-the-box. Patch from Alex Miller.
3825 PH/08 The ${def test on a variable was returning false if the variable's
3826 value was "0", contrary to what the specification has always said!
3827 The result should be true unless the variable is empty.
3829 PH/09 The syntax error of a character other than { following "${if
3830 def:variable_name" (after optional whitespace) was not being diagnosed.
3831 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3832 accidental colon was present, for example, could give incorrect results.
3834 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3835 result is used (not including appendfile, where other changes are about
3838 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3839 This involved changing a lot of size variables from int to off_t. It
3840 should work with maildirs and everything.
3842 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3843 spamd dying while we are connected to it.
3845 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3848 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3849 function for BDB 4.3.
3851 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3853 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3854 the sources for PCRE have been re-arranged and more files are now
3857 PH/15 The code I had for printing potentially long long variables in PH/11
3858 above was not the best (it lost precision). The length of off_t variables
3859 is now inspected at build time, and an appropriate printing format (%ld
3860 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3861 to be "long long int" or "long int". This is needed for the internal
3862 formatting function string_vformat().
3864 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3865 the configuration file to be ":syslog", then the script "guesses" where
3866 the logs files are, rather than using the compiled in default. In our
3867 case the guess is not the same as the compiled default, so the script
3868 suddenly stopped working when I started to use syslog. The patch checks
3869 to see if log_file_path is "". If so, it attempts to read it from exim
3870 with no configuration file to get the compiled in version, before it
3871 falls back to the previous guessing code."
3873 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3874 implementing BATV in an Exim configuration. See NewStuff for the gory
3877 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3878 Makefile that are specific to HP-UX.
3880 PH/18 If the "use_postmaster" option was set for a recipient callout together
3881 with the "random" option, the postmaster address was used as the MAIL
3882 FROM address for the random test, but not for the subsequent recipient
3883 test. It is now used for both.
3885 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3886 patch removes a few documentation additions to RFC 3028, because the
3887 latest draft now contains them. It adds the new en;ascii-case comparator
3888 and a new error check for 8bit text in MIME parts. Comparator and
3889 require names are now matched exactly. I enabled the subaddress
3890 extension, but it is not well tested yet (read: it works for me)."
3892 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3893 rework some of the code of TK/09 above to avoid the hardwired use of
3894 "%lld" and "long long". Replaced the call to snprintf() with a call to
3897 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3898 records point to non-existent hosts", "retry timeout exceeded", and
3899 "retry time not reached for any host after a long failure period".
3901 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3902 experimental DomainKeys support:
3904 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3905 (2) On an error such as an illegally used "control", the wrong name for
3906 the control was given.
3908 These problems did NOT occur unless DomainKeys support was compiled.
3910 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3912 PH/24 Added ${if match_ip condition.
3914 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3915 databases so that it will be absolutely obvious if a crash occurs in the
3916 DB library. This is a regular occurrence (often caused by mis-matched
3919 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3920 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3921 picked up on a user's system that detects such things. There doesn't seem
3922 to be a gcc warning option for this - only an attribute that has to be
3923 put on the function's prototype. It seems that in Fedora Core 4 they have
3924 set this on a number of new functions. No doubt there will be more in due
3927 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3928 specify a transport. However, if an address that was verified by such a
3929 router was the subject of a callout, Exim crashed because it tried to
3930 read the rcpt_include_affixes from the non-existent transport. Now it
3931 just assumes that the setting of that option is false. This bug was
3932 introduced by 4.51/PH/31.
3934 PH/28 Changed -d+all to exclude +memory, because that information is very
3935 rarely of interest, but it makes the output a lot bigger. People tend to
3936 do -d+all out of habit.
3938 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3939 code in os-type was giving problems when libc.so lives in lib64, like on
3942 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3943 aren't the modern standard, and it seems that some systems' include files
3944 don't always have them. Exim was already checking for some of the newer
3945 ones like T_AAAA, and defining it itself. I've added checks for all the
3946 record types that Exim uses.
3948 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3949 not automatically generating a new one, as it is supposed to. This
3950 prevented TLS from working. If the file did exist, but contained invalid
3951 data, a new version was generated, as expected. It was only the case of a
3952 non-existent file that was broken.
3954 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3955 with a change in libdomainkeys > 0.64.
3957 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3958 from DNS. If the selector record carries the flag, it now has
3959 precedence over the domain-wide flag.
3961 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3963 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3964 the use of a port name that isn't defined in /etc/services) occurred, the
3965 message was deferred as in a normal delivery, and thus remained on the
3966 spool, instead of being failed because of the mua_wrapper setting. This
3967 is now fixed, and I tidied up some of the mua_wrapper messages at the
3970 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3971 the messages in a hash of arrays rather than using individual hashes.
3972 This is a bit cleaner and results in dramatic memory savings, albeit
3973 at a slight CPU cost.
3975 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3976 as requested by Marc Sherman.
3978 SC/10 Eximstats: added histograms for user specified patterns as requested
3981 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3983 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3984 fopen() in the content-scanning modules that did not already have it.
3990 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3991 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3993 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3995 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3997 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3998 to test Sieve filters that use "vacation".
4000 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4001 that changes the way the GnuTLS parameters are stored in the cache file.
4002 The new format can be generated externally. For backward compatibility,
4003 if the data in the cache doesn't make sense, Exim assumes it has read an
4004 old-format file, and it generates new data and writes a new file. This
4005 means that you can't go back to an older release without removing the
4008 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4009 work if there are any delivery delays because "one_time" forces the
4010 parent to be marked "delivered", so its unseen clone is never tried
4011 again. For this reason, Exim now forbids the simultaneous setting of
4014 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4015 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4016 entry if you want to know the details. The fix, however, neglected to
4017 consider the case when local delivery batching is involved. The test for
4018 "previously delivered" was not happening when checking to see if an
4019 address could be batched with a previous (undelivered) one; under
4020 certain circumstances this could lead to multiple deliveries to the same
4023 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4024 in its include files, and this causes problems building Exim.
4026 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4027 header_syntax) but Exim was just ignoring anything given after a slash.
4028 In particular, this caused confusion with an attempt to use "verify =
4029 reverse_host_lookup/defer_ok". An error is now given when options are
4030 supplied for verify items that do not have them. (Maybe reverse_host_
4031 lookup should have a defer_ok option, but that's a different point.)
4033 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4034 defined by RFC 821) to 2048, because there were problems with some AUTH
4035 commands, and RFC 1869 says the size should be increased for extended
4036 SMTP commands that take arguments.
4038 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4041 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4042 "unknown" error; now it says that the functionality isn't in the binary.
4044 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4045 an address' error message when a string expansion fails (syntax or
4046 whatever). Otherwise the password may appear in the log. Following change
4047 PH/42 below, there is no longer a chance of it appearing in a bounce
4050 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4052 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4053 of its addresses was ever tried. (Bugzilla bug #2.)
4055 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4056 the result incorrectly in the debug output. (It correctly added a newline
4057 to what was transported.)
4059 TF/01 Added $received_time.
4061 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4062 commented out examples of how to interface to a virus scanner and to
4063 SpamAssassin. Also added commented examples of av_scanner and
4064 spamd_address settings.
4066 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4067 and controls are allowed in which ACLs. There were a couple of minor
4068 errors. Some of the entries in the conditions table (which is a table of
4069 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4070 negation of where the condition IS allowed.
4072 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4074 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4075 header file does not have a version number, so I've had to invent a new
4076 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4077 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4078 radiusclient), but it was contributed by a Radius user.
4080 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4081 files or directories, for OpenSSL.
4083 PH/19 When an Exim process that is running as root has to create an Exim log
4084 file, it does so in a subprocess that runs as exim:exim so as to get the
4085 ownership right at creation (otherwise, other Exim processes might see
4086 the file with the wrong ownership). There was no test for failure of this
4087 fork() call, which would lead to the process getting stuck as it waited
4088 for a non-existent subprocess. Forks do occasionally fail when resources
4089 run out. I reviewed all the other calls to fork(); they all seem to check
4092 PH/20 When checking for unexpected SMTP input at connect time (before writing
4093 the banner), Exim was not dealing correctly with a non-positive return
4094 from the read() function. If the client had disconnected by this time,
4095 the result was a log entry for a synchronization error with an empty
4096 string after "input=" when read() returned zero. If read() returned -1
4097 (an event I could not check), uninitialized data bytes were printed.
4098 There were reports of junk text (parts of files, etc) appearing after
4101 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4103 PH/22 Added support for macro redefinition, and (re)definition in between
4104 driver and ACL definitions.
4106 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4107 forgetting to use the resulting value; it was using the unexpanded value.
4109 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4110 hadn't been configured. The fix is from Juergen Kreileder, who
4111 understands it better than I do:
4113 "Here's what I see happening with three configured cyrus_sasl
4114 authenticators configured (plain, login, cram-md5):
4116 On startup auth_cyrus_sasl_init() gets called for each of these.
4117 This means three calls to sasl_listmech() without a specified mech_list.
4118 => SASL tests which mechs of all available mechs actually work
4119 => three warnings about OTP not working
4120 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4122 With the patch, sasl_listmech() also gets called three times. But now
4123 SASL's mech_list option is set to the server_mech specified in the the
4124 authenticator. Or in other words, the answer from sasl_listmech()
4125 gets limited to just the mech you're testing for (which is different
4127 => the return list contains just 'plain' or 'login', 'cram-md5' or
4128 nothing depending on the value of ob->server_mech.
4130 I've just tested the patch: Authentication still works fine,
4131 unavailable mechs specified in the exim configuration are still
4132 caught, and the auth.log warnings about OTP are gone."
4134 PH/25 When debugging is enabled, the contents of the command line are added
4135 to the debugging output, even when log_selector=+arguments is not
4138 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4139 answer is "GNU", and only if the return is "GNU/something" is the answer
4142 PH/27 $acl_verify_message is now set immediately after the failure of a
4143 verification in an ACL, and so is available in subsequent modifiers. In
4144 particular, the message can be preserved by coding like this:
4146 warn !verify = sender
4147 set acl_m0 = $acl_verify_message
4149 Previously, $acl_verify_message was set only while expanding "message"
4150 and "log_message" when a very denied access.
4152 PH/28 Modified OS/os.c-Linux with
4154 -#ifndef OS_LOAD_AVERAGE
4155 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4157 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4158 nomenclature these days.)
4160 PH/29 Installed patch from the Sieve maintainer that adds the options
4161 sieve_useraddress and sieve_subaddress to the redirect router.
4163 PH/30 In these circumstances:
4164 . Two addresses routed to the same list of hosts;
4165 . First host does not offer TLS;
4166 . First host accepts first address;
4167 . First host gives temporary error to second address;
4168 . Second host offers TLS and a TLS session is established;
4169 . Second host accepts second address.
4170 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4171 and peerdn, if requested) that were in fact used only for the second
4174 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4175 attention to any local part prefix or suffix that was matched by the
4176 router that accepted the address. It now behaves in the same way as it
4177 does for delivery: the affixes are removed from the local part unless
4178 rcpt_include_affixes is set on the transport.
4180 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4181 timeout during the DATA phase of an incoming message.
4183 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4184 applied a patch sanctioned by the Sieve maintainer.
4186 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4187 the uid or gid is negative. A case of a negative gid caused this to be
4188 noticed. The fix allows for either to be negative.
4190 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4191 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4194 PH/36 The change PH/12 above was broken. Fixed it.
4196 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4197 the grounds that routing the same address twice would always produce the
4198 same answer. This might have been true once, but it is certainly no
4199 longer true now. Routing a child address may depend on the previous
4200 routing that produced that child. Some complicated redirection strategies
4201 went wrong when messages had multiple recipients, and made Exim's
4202 behaviour dependent on the order in which the addresses were given.
4204 I have moved the duplicate checking until after the routing is complete.
4205 Exim scans the addresses that are assigned to local and remote
4206 transports, and removes any duplicates. This means that more work will be
4207 done, as duplicates will always all be routed, but duplicates are
4208 presumably rare, so I don't expect this is of any significance.
4210 For deliveries to pipes, files, and autoreplies, the duplicate checking
4211 still happens during the routing process, since they are not going to be
4214 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4215 It corrects a timeout issue with spamd. This is Ian's comment: "The
4216 background is that sometimes spamd either never reads data from a
4217 connection it has accepted, or it never writes response data. The exiscan
4218 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4219 blindly assumes that writes won't block so it may never time out."
4221 PH/39 Allow G after quota size as well as K and M.
4223 PH/40 The value set for $authenticated_id in an authenticator may not contain
4224 binary zeroes or newlines because the value is written to log lines and
4225 to spool files. There was no check on this. Now the value is run through
4226 the string_printing() function so that such characters are converted to
4227 printable escape sequences.
4229 PH/41 $message_linecount is a new variable that contains the total number of
4230 lines in the message. Compare $body_linecount, which is the count for the
4233 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4234 bounce and delay warning messages, except in certain special cases, which
4237 (a) An SMTP error message from a remote host;
4238 (b) A message specified in a :fail: redirection;
4239 (c) A message specified in a "fail" command in a system filter;
4240 (d) A message specified in a FAIL return from the queryprogram router;
4241 (e) A message specified by the cannot_route_message router option.
4243 In these cases only, Exim does include the error details in bounce and
4244 warning messages. There are also a few cases where bland messages such
4245 as "unrouteable address" or "local delivery error" are given.
4247 PH/43 $value is now also set for the "else" part of a ${run expansion.
4249 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4250 being worked on, but at least Exim now implements the latest version to
4253 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4254 process to complete was treated as a delivery failure, a timeout while
4255 writing the message to the pipe was logged, but erroneously treated as a
4256 successful delivery. Such timeouts include transport filter timeouts. For
4257 consistency with the overall process timeout, these timeouts are now
4258 treated as errors, giving rise to delivery failures by default. However,
4259 there is now a new Boolean option for the pipe transport called
4260 timeout_defer, which, if set TRUE, converts the failures into defers for
4261 both kinds of timeout. A transport filter timeout is now identified in
4264 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4265 systems where "make" and "gmake" are different, calling "gmake" at top
4266 level broke things. I've arranged for the value of $(MAKE) to be passed
4267 from the Makefile to this script so that it can call the same version of
4271 A note about Exim versions 4.44 and 4.50
4272 ----------------------------------------
4274 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4275 changes of various kinds. As a consequence, a big documentation update was
4276 needed. This delayed the release for rather longer than seemed good, especially
4277 in the light of a couple of (minor) security issues. Therefore, the changes
4278 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4279 release. So 4.44 and 4.50 are in effect two different branches that both start
4282 I have left the 4.50 change log unchanged; it contains all the changes since
4283 4.43. The change log for 4.44 is below; many of its items are identical to
4284 those for 4.50. This seems to be the most sensible way to preserve the
4285 historical information.
4291 1. Minor wording change to the doc/README.SIEVE file.
4293 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4294 computation of the current number of files was incorrect.
4296 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4297 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4300 4. Give more explanation in the error message when the command for a transport
4301 filter fails to execute.
4303 5. There are several places where Exim runs a non-Exim command in a
4304 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4305 was being done only for the command run by the queryprogram router. It is
4306 now done for all such subprocesses. The other cases are: ${run, transport
4307 filters, and the commands run by the lmtp and pipe transports.
4309 6. Added CONFIGURE_GROUP build-time option.
4311 7. Some older OS have a limit of 256 on the maximum number of file
4312 descriptors. Exim was using setrlimit() to set 1000 as a large value
4313 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4314 systems. I've change it so that if it can't get 1000, it tries for 256.
4316 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4317 was an oversight, and furthermore, ever since the addition of extra
4318 controls (e.g. 4.43/32), the checks on when to allow different forms of
4319 "control" were broken. There should now be diagnostics for all cases when a
4320 control that does not make sense is encountered.
4322 9. Added the /retain_sender option to "control=submission".
4324 10. $recipients is now available in the predata ACL (oversight).
4326 11. Tidy the search cache before the fork to do a delivery from a message
4327 received from the command line. Otherwise the child will trigger a lookup
4328 failure and thereby defer the delivery if it tries to use (for example) a
4329 cached ldap connection that the parent has called unbind on.
4331 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4332 of $address_data from the recipient verification was clobbered by the
4333 sender verification.
4335 13. The value of address_data from a sender verification is now available in
4336 $sender_address_data in subsequent conditions in the ACL statement.
4338 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4340 15. Added a new option "connect=<time>" to callout options, to set a different
4343 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4344 was its contents. (It was OK if the option was not defined at all.)
4346 17. A "Completed" log line is now written for messages that are removed from
4347 the spool by the -Mrm option.
4349 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4350 information about exactly what failed.
4352 19. Added -dd to debug only the daemon process.
4354 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4355 handles timeouts, both on the server side and network timeouts. Renamed the
4356 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4358 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4359 It is now set to "smtps".
4361 22. $host_address is now set to the target address during the checking of
4362 ignore_target_hosts.
4364 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4365 being passed; this would have caused $sender_host_name to have been used if
4366 matching the list had actually called for a host name (not very likely,
4367 since this list is usually IP addresses). A host name is now passed as
4370 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4371 code that specifies a non-restarting handler (typically sigaction() in
4372 modern systems) in an attempt to fix a rare and obscure crash bug.
4374 25. Narrowed the window for a race in the daemon that could cause it to ignore
4375 SIGCHLD signals. This is not a major problem, because they are used only to
4376 wake it up if nothing else does.
4378 26. A malformed maildirsize file could cause Exim to calculate negative values
4379 for the mailbox size or file count. Odd effects could occur as a result.
4380 The maildirsize information is now recalculated if the size or filecount
4383 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4384 support for a long time. Removed HAVE_SYS_VFS_H.
4386 28. Installed the latest version of exipick from John Jetmore.
4388 29. In an address list, if the pattern was not a regular expression, an empty
4389 subject address (from a bounce message) matched only if the pattern was an
4390 empty string. Non-empty patterns were not even tested. This was the wrong
4391 because it is perfectly reasonable to use an empty address as part of a
4392 database query. An empty address is now tested by patterns that are
4393 lookups. However, all the other forms of pattern expect the subject to
4394 contain a local part and a domain, and therefore, for them, an empty
4395 address still always fails if the pattern is not itself empty.
4397 30. Exim went into a mad DNS loop when attempting to do a callout where the
4398 host was specified on an smtp transport, and looking it up yielded more
4399 than one IP address.
4401 31. Re-factored the code for checking spool and log partition space into a
4402 function that finds that data and another that does the check. The former
4403 is then used to implement four new variables: $spool_space, $log_space,
4404 $spool_inodes, and $log_inodes.
4406 32. The RFC2047 encoding function was originally intended for short strings
4407 such as real names; it was not keeping to the 75-character limit for
4408 encoded words that the RFC imposes. It now respects the limit, and
4409 generates multiple encoded words if necessary. To be on the safe side, I
4410 have increased the buffer size for the ${rfc2047: expansion operator from
4413 33. It is now permitted to omit both strings after an "if" condition; if the
4414 condition is true, the result is "true". As before, when the second string
4415 is omitted, a false condition yields an empty string. This makes it less
4416 cumbersome to write custom ACL and router conditions.
4418 34. Failure to deliver a bounce message always caused it to be frozen, even if
4419 there was an errors_to setting on the router. The errors_to setting is now
4422 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4423 canonical form (fully expanded) before being placed in
4424 $sender_host_address.
4426 36. The table in the code that translates DNS record types into text (T_A to
4427 "A" for instance) was missing entries for NS and CNAME. It is just possible
4428 that this could have caused confusion if both these types were looked up
4429 for the same domain, because the text type is used as part of Exim's
4430 per-process caching. But the chance of anyone hitting this buglet seems
4433 37. The dnsdb lookup has been extended in a number of ways.
4435 (1) There is a new type, "zns", which walks up the domain tree until it
4436 finds some nameserver records. It should be used with care.
4438 (2) There is a new type, "mxh", which is like "mx" except that it returns
4439 just the host names, not the priorities.
4441 (3) It is now possible to give a list of domains (or IP addresses) to be
4442 looked up. The behaviour when one of the lookups defers can be
4443 controlled by a keyword.
4445 (4) It is now possible to specify the separator character for use when
4446 multiple records are returned.
4448 38. The dnslists ACL condition has been extended: it is now possible to supply
4449 a list of IP addresses and/or domains to be looked up in a particular DNS
4452 39. Added log_selector=+queue_time_overall.
4454 40. When running the queue in the test harness, wait just a tad after forking a
4455 delivery process, to get repeatability of debugging output.
4457 41. Include certificate and key file names in error message when GnuTLS fails
4458 to set them up, because the GnuTLS error message doesn't include the name
4459 of the failing file when there is a problem reading it.
4461 42. Allow both -bf and -bF in the same test run.
4463 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4465 44. The "Exiscan patch" is now merged into the mainline Exim source.
4467 45. Sometimes the final signoff response after QUIT could fail to get
4468 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4469 < 0 before doing a fflush(). This bug looks as though it goes back to the
4470 introduction of TLS in release 3.20, but "sometimes" must have been rare
4471 because the tests only now provoked it.
4473 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4474 (this can affect the format of dates).
4476 47. exim_tidydb, when checking for the continued existence of a message for
4477 which it has found a message-specific retry record, was not finding
4478 messages that were in split spool directories. Consequently, it was
4479 deleting retry records that should have stayed in existence.
4481 48. Steve fixed some bugs in eximstats.
4483 49. The SPA authentication driver was not abandoning authentication and moving
4484 on to the next authenticator when an expansion was forced to fail,
4485 contradicting the general specification for all authenticators. Instead it
4486 was generating a temporary error. It now behaves as specified.
4488 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4489 (the order specifies the preference for clients). The order is now AES256,
4490 AES128, 3DES, ARCFOUR128.
4492 51. Small patch to Sieve code - explicitly set From: when generating an
4495 52. Exim crashed if a remote delivery caused a very long error message to be
4496 recorded - for instance if somebody sent an entire SpamAssassin report back
4497 as a large number of 550 error lines. This bug was coincidentally fixed by
4498 increasing the size of one of Exim's internal buffers (big_buffer) that
4499 happened as part of the Exiscan merge. However, to be on the safe side, I
4500 have made the code more robust (and fixed the comments that describe what
4503 53. Now that there can be additional text after "Completed" in log lines (if
4504 the queue_time_overall log selector is set), a one-byte patch to exigrep
4505 was needed to allow it to recognize "Completed" as not the last thing in
4508 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4509 patch that reportedly fixes this has been added. I am not expert enough to
4510 create a test for it. This is what the patch creator wrote:
4512 "I found a little strange behaviour of ldap code when working with
4513 Windows 2003 AD Domain, where users was placed in more than one
4514 Organization Units. When I tried to give exim partial DN, the exit code
4515 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4516 But simultaneously result of request was absolutely normal ldap result,
4517 so I produce this patch..."
4519 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4520 REFERENCE, so I have modified the code to exclude the patch when that macro
4523 55. Some experimental protocols are using DNS PTR records for new purposes. The
4524 keys for these records are domain names, not reversed IP addresses. The
4525 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4526 leaves it alone. Component reversal etc. now happens only for IP addresses.
4529 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4531 57. Double the size of the debug message buffer (to 2048) so that more of very
4532 long debug lines gets shown.
4534 58. The exicyclog utility now does better if the number of log files to keep
4535 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4537 59. Two changes related to the smtp_active_hostname option:
4539 (1) $smtp_active_hostname is now available as a variable.
4540 (2) The default for smtp_banner uses $smtp_active_hostname instead
4541 of $primary_hostname.
4543 60. The host_aton() function is supposed to be passed a string that is known
4544 to be a valid IP address. However, in the case of IPv6 addresses, it was
4545 not checking this. This is a hostage to fortune. Exim now panics and dies
4546 if the condition is not met. A case was found where this could be provoked
4547 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4548 components; fortuitously, this particular loophole had already been fixed
4549 by change 4.50/55 above.
4551 If there are any other similar loopholes, the new check in host_aton()
4552 itself should stop them being exploited. The report I received stated that
4553 data on the command line could provoke the exploit when Exim was running as
4554 exim, but did not say which command line option was involved. All I could
4555 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4556 running as the user.
4559 61. There was a buffer overflow vulnerability in the SPA authentication code
4560 (which came originally from the Samba project). I have added a test to the
4561 spa_base64_to_bits() function which I hope fixes it.
4564 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4565 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4567 63. The daemon start-up calls getloadavg() while still root for those OS that
4568 need the first call to be done as root, but it missed one case: when
4569 deliver_queue_load_max is set with deliver_drop_privilege. This is
4570 necessary for the benefit of the queue runner, because there is no re-exec
4571 when deliver_drop_privilege is set.
4573 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4574 This has been fixed.
4576 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4577 was in use, was not putting the data itself into the right store pool;
4578 consequently, it could be overwritten for a subsequent message in the same
4579 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4582 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4584 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4585 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4586 the function were treating the return as a boolean value, which happened to
4587 work because 0=false and not-0=true, but is not correct code.
4589 68. The host_aton() function was not handling scoped IPv6 addresses (those
4590 with, for example, "%eth0" on the end) correctly.
4592 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4593 negated items (that is, ~something) in unsigned ints. Some compilers
4594 apparently mutter when there is no cast.
4596 70. If an address verification called from an ACL failed, and did not produce a
4597 user-specific message (i.e. there was only a "system" message), nothing was
4598 put in $acl_verify_message. In this situation, it now puts the system
4601 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4602 session; change 4.31/43 added the unwanted input to the log line - except
4603 that it did not do this in the start of session case. It now does.
4605 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4606 This is wrong and can cause the other end to generate a synchronization
4607 error if it is another Exim or anything else that does the synchronization
4608 check. A QUIT command is no longer sent after a timeout.
4610 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4611 during host lookups.
4613 74. The defer_ok option of callout verification was not working if it was used
4614 when verifying addresses in header lines, that is, for this case:
4616 verify = header_sender/callout=defer_ok
4618 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4619 those file descriptors could be used for SMTP connections. If anything
4620 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4621 could be sent to the SMTP client, causing chaos. The daemon now opens
4622 stdin, stdout, and stderr to /dev/null when it puts itself into the
4625 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4626 log by default. The user can override this with suitable Perl magic.
4628 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4629 the log message when discard triggers, was not working for the DATA ACL or
4630 for the non-SMTP ACL.
4632 78. Error message wording change in sieve.c.
4634 79. If smtp_accept_max_per_host was set, the number of connections could be
4635 restricted to fewer than expected, because the daemon was trying to set up
4636 a new connection before checking whether the processes handling previous
4637 connections had finished. The check for completed processes is now done
4638 earlier. On busy systems, this bug wouldn't be noticed because something
4639 else would have woken the daemon, and it would have reaped the completed
4642 80. If a message was submitted locally by a user whose login name contained one
4643 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4644 It caused a spool format error. I have fixed the spool reading code. A
4645 related problem was that the "from" clause in the Received: line became
4646 illegal because of the space(s). It is now covered by ${quote_local_part.
4648 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4651 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4653 83. Patch from Sieve maintainer to fix unterminated string problem in
4654 "vacation" handling.
4656 84. Some minor changes to the Linux configuration files to help with other
4657 OS variants using glibc.
4659 85. One more patch for Sieve to update vacation handling to latest spec.
4662 ----------------------------------------------------
4663 See the note above about the 4.44 and 4.50 releases.
4664 ----------------------------------------------------
4670 1. Change 4.43/35 introduced a bug that caused file counts to be
4671 incorrectly computed when quota_filecount was set in an appendfile
4674 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4675 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4678 3. Give more explanation in the error message when the command for a transport
4679 filter fails to execute.
4681 4. There are several places where Exim runs a non-Exim command in a
4682 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4683 was being done only for the command run by the queryprogram router. It is
4684 now done for all such subprocesses. The other cases are: ${run, transport
4685 filters, and the commands run by the lmtp and pipe transports.
4687 5. Some older OS have a limit of 256 on the maximum number of file
4688 descriptors. Exim was using setrlimit() to set 1000 as a large value
4689 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4690 systems. I've change it so that if it can't get 1000, it tries for 256.
4692 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4693 was an oversight, and furthermore, ever since the addition of extra
4694 controls (e.g. 4.43/32), the checks on when to allow different forms of
4695 "control" were broken. There should now be diagnostics for all cases when a
4696 control that does not make sense is encountered.
4698 7. $recipients is now available in the predata ACL (oversight).
4700 8. Tidy the search cache before the fork to do a delivery from a message
4701 received from the command line. Otherwise the child will trigger a lookup
4702 failure and thereby defer the delivery if it tries to use (for example) a
4703 cached ldap connection that the parent has called unbind on.
4705 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4706 of $address_data from the recipient verification was clobbered by the
4707 sender verification.
4709 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4710 was its contents. (It was OK if the option was not defined at all.)
4712 11. A "Completed" log line is now written for messages that are removed from
4713 the spool by the -Mrm option.
4715 12. $host_address is now set to the target address during the checking of
4716 ignore_target_hosts.
4718 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4719 being passed; this would have caused $sender_host_name to have been used if
4720 matching the list had actually called for a host name (not very likely,
4721 since this list is usually IP addresses). A host name is now passed as
4724 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4725 code that specifies a non-restarting handler (typically sigaction() in
4726 modern systems) in an attempt to fix a rare and obscure crash bug.
4728 15. Narrowed the window for a race in the daemon that could cause it to ignore
4729 SIGCHLD signals. This is not a major problem, because they are used only to
4730 wake it up if nothing else does.
4732 16. A malformed maildirsize file could cause Exim to calculate negative values
4733 for the mailbox size or file count. Odd effects could occur as a result.
4734 The maildirsize information is now recalculated if the size or filecount
4737 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4738 support for a long time. Removed HAVE_SYS_VFS_H.
4740 18. Updated exipick to current release from John Jetmore.
4742 19. Allow an empty sender to be matched against a lookup in an address list.
4743 Previously the only cases considered were a regular expression, or an
4746 20. Exim went into a mad DNS lookup loop when doing a callout where the
4747 host was specified on the transport, if the DNS lookup yielded more than
4750 21. The RFC2047 encoding function was originally intended for short strings
4751 such as real names; it was not keeping to the 75-character limit for
4752 encoded words that the RFC imposes. It now respects the limit, and
4753 generates multiple encoded words if necessary. To be on the safe side, I
4754 have increased the buffer size for the ${rfc2047: expansion operator from
4757 22. Failure to deliver a bounce message always caused it to be frozen, even if
4758 there was an errors_to setting on the router. The errors_to setting is now
4761 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4762 canonical form (fully expanded) before being placed in
4763 $sender_host_address.
4765 24. Updated eximstats to version 1.33
4767 25. Include certificate and key file names in error message when GnuTLS fails
4768 to set them up, because the GnuTLS error message doesn't include the name
4769 of the failing file when there is a problem reading it.
4771 26. Expand error message when OpenSSL has problems setting up cert/key files.
4774 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4775 (this can affect the format of dates).
4777 28. exim_tidydb, when checking for the continued existence of a message for
4778 which it has found a message-specific retry record, was not finding
4779 messages that were in split spool directories. Consequently, it was
4780 deleting retry records that should have stayed in existence.
4782 29. eximstats updated to version 1.35
4783 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4784 1.35 - bugfix such that pie charts by volume are generated correctly
4786 30. The SPA authentication driver was not abandoning authentication and moving
4787 on to the next authenticator when an expansion was forced to fail,
4788 contradicting the general specification for all authenticators. Instead it
4789 was generating a temporary error. It now behaves as specified.
4791 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4792 (the order specifies the preference for clients). The order is now AES256,
4793 AES128, 3DES, ARCFOUR128.
4795 31. Small patch to Sieve code - explicitly set From: when generating an
4798 32. Exim crashed if a remote delivery caused a very long error message to be
4799 recorded - for instance if somebody sent an entire SpamAssassin report back
4800 as a large number of 550 error lines. This bug was coincidentally fixed by
4801 increasing the size of one of Exim's internal buffers (big_buffer) that
4802 happened as part of the Exiscan merge. However, to be on the safe side, I
4803 have made the code more robust (and fixed the comments that describe what
4806 33. Some experimental protocols are using DNS PTR records for new purposes. The
4807 keys for these records are domain names, not reversed IP addresses. The
4808 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4809 leaves it alone. Component reversal etc. now happens only for IP addresses.
4812 34. The host_aton() function is supposed to be passed a string that is known
4813 to be a valid IP address. However, in the case of IPv6 addresses, it was
4814 not checking this. This is a hostage to fortune. Exim now panics and dies
4815 if the condition is not met. A case was found where this could be provoked
4816 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4817 components; fortuitously, this particular loophole had already been fixed
4818 by change 4.50/55 or 4.44/33 above.
4820 If there are any other similar loopholes, the new check in host_aton()
4821 itself should stop them being exploited. The report I received stated that
4822 data on the command line could provoke the exploit when Exim was running as
4823 exim, but did not say which command line option was involved. All I could
4824 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4825 running as the user.
4828 35. There was a buffer overflow vulnerability in the SPA authentication code
4829 (which came originally from the Samba project). I have added a test to the
4830 spa_base64_to_bits() function which I hope fixes it.
4833 36. The daemon start-up calls getloadavg() while still root for those OS that
4834 need the first call to be done as root, but it missed one case: when
4835 deliver_queue_load_max is set with deliver_drop_privilege. This is
4836 necessary for the benefit of the queue runner, because there is no re-exec
4837 when deliver_drop_privilege is set.
4839 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4840 was in use, was not putting the data itself into the right store pool;
4841 consequently, it could be overwritten for a subsequent message in the same
4842 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4845 38. Sometimes the final signoff response after QUIT could fail to get
4846 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4847 < 0 before doing a fflush(). This bug looks as though it goes back to the
4848 introduction of TLS in release 3.20, but "sometimes" must have been rare
4849 because the tests only now provoked it.
4855 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4856 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4857 Now it returns an integer. A number of calls were still expecting a T/F
4858 return. Fortuitously, in all cases, the tests worked in OK situations,
4859 which is the norm. However, things would have gone wrong on any write
4860 failures on the smtp file descriptor. This function is used when sending
4861 messages over SMTP and also when doing verify callouts.
4863 2. When Exim is called to do synchronous delivery of a locally submitted
4864 message (the -odf or -odi options), it no longer closes stderr before doing
4867 3. Implemented the mua_wrapper option.
4869 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4871 5. Implemented the functions header_remove(), header_testname(),
4872 header_add_at_position(), and receive_remove_recipient(), and exported them
4875 6. If an ACL "warn" statement specified the addition of headers, Exim already
4876 inserted X-ACL-Warn: at the start if there was no header name. However, it
4877 was not making this test for the second and subsequent header lines if
4878 there were newlines in the string. This meant that an invalid header could
4879 be inserted if Exim was badly configured.
4881 7. Allow an ACL "warn" statement to add header lines at the start or after all
4882 the Received: headers, as well as at the end.
4884 8. Added the rcpt_4xx retry error code.
4886 9. Added postmaster_mailfrom=xxx to callout verification option.
4888 10. Added mailfrom=xxxx to the callout verification option, for verify=
4891 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4892 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4894 12. Inserted some casts to stop certain compilers warning when using pointer
4895 differences as field lengths or precisions in printf-type calls (mostly
4896 affecting debugging statements).
4898 13. Added optional readline() support for -be (dynamically loaded).
4900 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4901 same clock tick as a message's arrival, so that its received time was the
4902 same as the "first fail" time on the retry record, and that message
4903 remained on the queue past the ultimate address timeout, every queue runner
4904 would try a delivery (because it was past the ultimate address timeout) but
4905 after another failure, the ultimate address timeout, which should have then
4906 bounced the address, did not kick in. This was a "< instead of <=" error;
4907 in most cases the first failure would have been in the next clock tick
4908 after the received time, and all would be well.
4910 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4911 being recognized when the domain list was tested by the match_domain
4912 condition in an expansion string.
4914 16. Added the ${str2b64: operator.
4916 17. Exim was always calling setrlimit() to set a large limit for the number of
4917 processes, without checking whether the existing limit was already
4918 adequate. (It did check for the limit on file descriptors.) Furthermore,
4919 errors from getrlimit() and setrlimit() were being ignored. Now they are
4920 logged to the main and panic logs, but Exim does carry on, to try to do its
4921 job under whatever limits there are.
4923 18. Imported PCRE 5.0.
4925 19. Trivial typo in log message " temporarily refused connection" (the leading
4928 20. If the log selector return_path_on_delivery was set and an address was
4929 redirected to /dev/null, the delivery process crashed because it assumed
4930 that a return path would always be set for a "successful" delivery. In this
4931 case, the whole delivery is bypassed as an optimization, and therefore no
4934 21. Internal re-arrangement: the function for sending a challenge and reading
4935 a response while authentication was assuming a zero-terminated challenge
4936 string. It's now changed to take a pointer and a length, to allow for
4937 binary data in such strings.
4939 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4941 23. Exim was not respecting finduser_retries when seeking the login of the
4942 uid under which it was called; it was always trying 10 times. (The default
4943 setting of finduser_retries is zero.) Also, it was sleeping after the final
4944 failure, which is pointless.
4946 24. Implemented tls_on_connect_ports.
4948 25. Implemented acl_smtp_predata.
4950 26. If the domain in control=submission is set empty, Exim assumes that the
4951 authenticated id is a complete email address when it generates From: or
4952 Sender: header lines.
4954 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4955 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4956 chown and chgrp in /bin and hostname in /usr/bin.
4958 28. Exim was keeping the "process log" file open after each use, just as it
4959 does for the main log. This opens the possibility of it remaining open for
4960 long periods when the USR1 signal hits a daemon. Occasional processlog
4961 errors were reported, that could have been caused by this. Anyway, it seems
4962 much more sensible not to leave this file open at all, so that is what now
4965 29. The long-running daemon process does not normally write to the log once it
4966 has entered its main loop, and it closes the log before doing so. This is
4967 so that log files can straightforwardly be renamed and moved. However,
4968 there are a couple of unusual error situations where the daemon does write
4969 log entries, and I had neglected to close the log afterwards.
4971 30. The text of an SMTP error response that was received during a remote
4972 delivery was being truncated at 512 bytes. This is too short for some of
4973 the long messages that one sometimes sees. I've increased the limit to
4976 31. It is now possible to make retry rules that apply only when a message has a
4977 specific sender, in particular, an empty sender.
4979 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4980 it possible to be selective about when SMTP synchronization is enforced.
4982 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4984 32. Implemented hosts_connection_nolog.
4986 33. Added an ACL for QUIT.
4988 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4991 35. Added mailbox_size and mailbox_filecount to appendfile.
4993 36. Added control = no_multiline_responses to ACLs.
4995 37. There was a bug in the logic of the code that waits for the clock to tick
4996 in the case where the clock went backwards by a substantial amount such
4997 that the microsecond fraction of "now" was more than the microsecond
4998 fraction of "then" (but the whole seconds number was less).
5000 38. Added support for the libradius Radius client library this is found on
5001 FreeBSD (previously only the radiusclient library was supported).
5007 1. When certain lookups returned multiple values in the form name=value, the
5008 quoting of the values was not always being done properly. Specifically:
5009 (a) If the value started with a double quote, but contained no whitespace,
5011 (b) If the value contained whitespace other than a space character (i.e.
5012 tabs or newlines or carriage returns) it was not quoted.
5013 This fix has been applied to the mysql and pgsql lookups by writing a
5014 separate quoting function and calling it from the lookup code. The fix
5015 should probably also be applied to nisplus, ibase and oracle lookups, but
5016 since I cannot test any of those, I have not disturbed their existing code.
5018 2. A hit in the callout cache for a specific address caused a log line with no
5019 reason for rejecting RCPT. Now it says "Previous (cached) callout
5020 verification failure".
5022 3. There was an off-by-one bug in the queryprogram router. An over-long
5023 return line was truncated at 256 instead of 255 characters, thereby
5024 overflowing its buffer with the terminating zero. As well as fixing this, I
5025 have increased the buffer size to 1024 (and made a note to document this).
5027 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5028 when Exim is waiting for an SMTP response from a remote server, Exim
5029 restarts its select() call on the socket, thereby resetting its timeout.
5030 This is not a problem when such interrupts are rare. Somebody set up a cron
5031 job to run exiwhat every 2 minutes, which is less than the normal select()
5032 timeout (5 or 10 minutes). This meant that the select() timeout never
5033 kicked in because it was always reset. I have fixed this by comparing the
5034 time when an interrupt arrives with the time at the start of the first call
5035 to select(). If more time than the timeout has elapsed, the interrupt is
5036 treated as a timeout.
5038 5. Some internal re-factoring in preparation for the addition of Sieve
5039 extensions (by MH). In particular, the "personal" test is moved to a
5040 separate function, and given an option for scanning Cc: and Bcc: (which is
5041 not set for Exim filters).
5043 6. When Exim created an email address using the login of the caller as the
5044 local part (e.g. when creating a From: or Sender: header line), it was not
5045 quoting the local part when it contained special characters such as @.
5047 7. Installed new OpenBSD configuration files.
5049 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5050 try to make them clearer.
5052 9. Callout options, other than the timeout value, were being ignored when
5053 verifying sender addresses in header lines. For example, when using
5055 verify = header_sender/callout=no_cache
5057 the cache was (incorrectly) being used.
5059 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5060 only the Cygwin environment.
5062 11. When return_path_on_delivery was set as a log selector, if different remote
5063 addresses in the same message used different return paths and parallel
5064 remote delivery occurred, the wrong values would sometimes be logged.
5065 (Whenever a remote delivery process finished, the return path value from
5066 the most recently started remote delivery process was logged.)
5068 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5069 lines when AUTH and/or TLS are in use. This is the "received protocol"
5070 field. Exim used to use "asmtp" for authenticated SMTP, without any
5071 indication (in the protocol name) for TLS use. Now it follows the RFC and
5072 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5073 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5074 names appear in log lines as well as in Received: header lines.
5076 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5077 extensions, and comparison tests, and to fix some bugs.
5079 14. Changes to the "personal" filter test:
5081 (1) The test was buggy in that it was just doing the equivalent of
5082 "contains" tests on header lines. For example, if a user's address was
5083 anne@some.where, the "personal" test would incorrectly be true for
5085 To: susanne@some.where
5087 This test is now done by extracting each address from the header in turn,
5088 and checking the entire address. Other tests that are part of "personal"
5089 are now done using regular expressions (for example, to check local parts
5090 of addresses in From: header lines).
5092 (2) The list of non-personal local parts in From: addresses has been
5093 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5094 taken from the Sieve specification recommendations.
5096 (3) If the message contains any header line starting with "List-" it is
5097 treated as non-personal.
5099 (4) The test for "circular" in the Subject: header line has been removed
5100 because it now seems ill-conceived.
5102 15. Minor typos in src/EDITME comments corrected.
5104 16. Installed latest exipick from John Jetmore.
5106 17. If headers_add on a router specified a text string that was too long for
5107 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5108 of string_sprintf() is now avoided.
5110 18. $message_body_size was not set (it was always zero) when running the DATA
5111 ACL and the local_scan() function.
5113 19. For the "mail" command in an Exim filter, no default was being set for
5114 the once_repeat time, causing a random time value to be used if "once" was
5115 specified. (If the value happened to be <= 0, no repeat happened.) The
5116 default is now 0s, meaning "never repeat". The "vacation" command was OK
5117 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5118 (I found it when inspecting the code).
5120 20. There is now an overall timeout for performing a callout verification. It
5121 defaults to 4 times the callout timeout, which applies to individual SMTP
5122 commands during the callout. The overall timeout applies when there is more
5123 than one host that can be tried. The timeout is checked before trying the
5124 next host. This prevents very long delays if there are a large number of
5125 hosts and all are timing out (e.g. when the network connections are timing
5126 out). The value of the overall timeout can be changed by specifying an
5127 additional sub-option for "callout", called "maxwait". For example:
5129 verify = sender/callout=5s,maxwait=20s
5131 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5132 to the end before writing, but this should make it even safer).
5134 22. Exim was forgetting that it had advertised PIPELINING for the second and
5135 subsequent messages on an SMTP connection. It was also not resetting its
5136 memory on STARTTLS and an internal HELO.
5138 23. When Exim logs an SMTP synchronization error within a session, it now
5139 records whether PIPELINING has been advertised or not.
5141 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5142 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5143 rather than long int.
5145 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5147 26. Added the never_mail option to autoreply.
5153 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5154 crash if the getsockname() call failed; this can happen if a connection is
5155 closed very soon after it is established. The problem was simply in the
5156 order in which certain operations were done, causing Exim to try to write
5157 to the SMTP stream before it had set up the file descriptor. The bug has
5158 been fixed by making things happen in the correct order.
5164 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5165 before the connection was closed, thus losing the rejection response.
5167 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5168 some early Solaris releases, but causes trouble in current releases where
5169 socklen_t is defined.
5171 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5174 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5177 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5178 that releasing the top of it at the end releases what was used for sub-
5179 expansions (unless the block got too big). However, discard this block if
5180 the first thing is a variable or header, so that we can use its block when
5181 it is dynamic (useful for very large $message_headers, for example).
5183 6. Lookups now cache *every* query, not just the most recent. A new, separate
5184 store pool is used for this. It can be recovered when all lookup caches are
5185 flushed. Lookups now release memory at the end of their result strings.
5186 This has involved some general refactoring of the lookup sources.
5188 7. Some code has been added to the store_xxx() functions to reduce the amount
5189 of flapping under certain conditions.
5191 8. log_incoming_interface used to affect only the <= reception log lines. Now
5192 it causes the local interface and port to be added to several more SMTP log
5193 lines, for example "SMTP connection from", and rejection lines.
5195 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5197 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5199 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5200 could be overwritten at the end of the current message (or the start of a
5201 new message if it was set in a HELO ACL). The value is now preserved for
5202 the duration of the SMTP connection.
5204 12. If a transport had a headers_rewrite setting, and a matching header line
5205 contained an unqualified address, that address was qualified, even if it
5206 did not match any rewriting rules. The underlying bug was that the values
5207 of the flags that permit the existence of unqualified sender and recipient
5208 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5209 non-local messages, and by -bnq for local messages) were not being
5210 preserved with the message after it was received.
5212 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5213 "next input=" as part of the text comprising the host identity instead of
5214 the correct text. The code was using the same buffer for two different
5215 strings. However, depending on which order the printing function evaluated
5216 its arguments, the bug did not always show up. Under Linux, for example, my
5217 test suite worked just fine.
5219 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5220 doesn't work with some older versions of Perl. It has been changed to "my",
5221 which in any case is probably the better facility to use.
5223 15. A really picky compiler found some instances of statements for creating
5224 error messages that either had too many or two few arguments for the format
5227 16. The size of the buffer for calls to the DNS resolver has been increased
5228 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5229 for addresses that have a lot of PTR records. This alleviates a problem; it
5230 does not fully solve it.
5232 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5233 buffer now truncates the list and logs the incident, which is the same
5234 action as happens when Exim is looking up a host name and its aliases.
5235 Previously in this situation something unpredictable would happen;
5236 sometimes it was "internal error: store_reset failed".
5238 18. If a server dropped the connection unexpectedly when an Exim client was
5239 using GnuTLS and trying to read a response, the client delivery process
5240 crashed while trying to generate an error log message.
5242 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5243 string, for example:
5245 warn message = H1: something\nH2: something
5247 the text was added as a single header line from Exim's point of view
5248 though it ended up OK in the delivered message. However, searching for the
5249 second and subsequent header lines using $h_h2: did not work. This has been
5250 fixed. Similarly, if a system filter added multiple headers in this way,
5251 the routers could not see them.
5253 20. Expanded the error message when iplsearch is called with an invalid key to
5254 suggest using net-iplsearch in a host list.
5256 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5257 ACLs are no longer actually imposed (and a message to that effect is
5260 22. If a "gecos" field in a passwd entry contained escaped characters, in
5261 particular, if it contained a \" sequence, Exim got it wrong when building
5262 a From: or a Sender: header from that name. A second bug also caused
5263 incorrect handling when an unquoted " was present following a character
5264 that needed quoting.
5266 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5267 was not being matched caselessly.
5269 24. Arranged for all hyphens in the exim.8 source to be escaped with
5272 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5273 properly. Recipient callout cache records were still being keyed to include
5274 the sender, even when use_sender was set false. This led to far more
5275 callouts that were necessary. The sender is no longer included in the key
5276 when use_sender is false.
5278 26. Added "control = submission" modifier to ACLs.
5280 27. Added the ${base62d: operator to decode base 62 numbers.
5282 28. dnsdb lookups can now access SRV records.
5284 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5285 the configuration file.
5287 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5288 (-v) mode. This makes the output for a verbose queue run more intelligible.
5290 31. Added a use_postmaster feature to recipient callouts.
5292 32. Added the $body_zerocount variable, containing the number of binary zero
5293 bytes in the message body.
5295 33. The time of last modification of the "new" subdirectory is now used as the
5296 "mailbox time last read" when there is a quota error for a maildir
5299 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5301 35. Added +ignore_unknown as a special item in host lists.
5303 36. Code for decoding IPv6 addresses in host lists is now included, even if
5304 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5305 address was recognized as an IP address, but was then not correctly decoded
5306 into binary, causing unexpected and incorrect effects when compared with
5313 1. Very minor rewording of debugging text in manualroute to say "list of
5314 hosts" instead of "hostlist".
5316 2. If verify=header_syntax was set, and a header line with an unqualified
5317 address (no domain) and a large number of spaces between the end of the
5318 name and the colon was received, the reception process suffered a buffer
5319 overflow, and (when I tested it) crashed. This was caused by some obsolete
5320 code that should have been removed. The fix is to remove it!
5322 3. When running in the test harness, delay a bit after writing a bounce
5323 message to get a bit more predictability in the log output.
5325 4. Added a call to search_tidyup() just before forking a reception process. In
5326 theory, someone could use a lookup in the expansion of smtp_accept_max_
5327 per_host which, without the tidyup, could leave open a database connection.
5329 5. Added the variables $recipient_data and $sender_data which get set from a
5330 lookup success in an ACL "recipients" or "senders" condition, or a router
5331 "senders" option, similar to $domain_data and $local_part_data.
5333 6. Moved the writing of debug_print from before to after the "senders" test
5336 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5337 problems for message scanning, either using a data ACL, or using
5338 local_scan() because the Received: header was not generated till after they
5339 were called (in order to set the time as the time of reception completion).
5340 I have revised the way this works. The header is now generated after the
5341 body is received, but before the ACL or local_scan() are called. After they
5342 are run, the timestamp in the header is updated.
5348 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5349 before starting a queue runner without re-exec. This happened only when
5350 deliver_drop_privilege was set or when the Exim user was set to root. The
5351 effect of the bug was that timeouts during subsequent deliveries caused
5352 crashes instead of being properly handled. The handler is now left at its
5353 default (and expected) setting.
5355 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5356 message, again when deliver_drop_privilege is set or Exim is run as root.
5357 The bug described in (1) was not present in this case, but the tidying up
5358 of the other signals was missing. I have made the two cases consistent.
5360 3. The ignore_target_hosts setting on a manualroute router was being ignored
5361 for hosts that were looked up using the /MX notation.
5363 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5366 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5367 operated on the sender address. After changing the $sender_address to <>
5368 for the sender address verify, Exim was re-instated it as the original
5369 (before rewriting) address, but remembering that it had rewritten it, so it
5370 wasn't rewriting it again. This bug also had the effect of breaking the
5371 sender address verification caching when the sender address was rewritten.
5373 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5374 This has been changed so that if the ip literal address matches
5375 ignore_target_hosts, the router declines.
5377 7. Added expansion conditions match_domain, match_address, and match_local_
5378 part (NOT match_host).
5380 8. The placeholder for the Received: header didn't have a length field set.
5382 9. Added code to Exim itself and to exim_lock to test for a specific race
5383 condition that could lead to file corruption when using MBX delivery. The
5384 issue is with the lockfile that is created in /tmp. If this file is removed
5385 after a process has opened it but before that process has acquired a lock,
5386 there is the potential for a second process to recreate the file and also
5387 acquire a lock. This could lead to two Exim processes writing to the file
5388 at the same time. The added code performs the same test as UW imapd; it
5389 checks after acquiring the lock that its file descriptor still refers to
5390 the same named file.
5392 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5393 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5396 11. Added the smtp_active_hostname option. If used, this will typically be made
5397 to depend on the incoming interface address. Because $interface_address is
5398 not set up until the daemon has forked a reception process, error responses
5399 that can happen earlier (such as "too many connections") no longer contain
5402 12. If an expansion in a condition on a "warn" statement fails because a lookup
5403 defers, the "warn" statement is abandoned, and the next ACL statement is
5404 processed. Previously this caused the whole ACL to be aborted.
5406 13. Added the iplsearch lookup type.
5408 14. Added ident_timeout as a log selector.
5410 15. Added tls_certificate_verified as a log selector.
5412 16. Added a global option tls_require_ciphers (compare the smtp transport
5413 option of the same name). This controls incoming TLS connections.
5415 17. I finally figured out how to make tls_require_ciphers do a similar thing
5416 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5417 before starting the TLS session.
5419 18. Tabs are now shown as \t in -bP output.
5421 19. If the log selector return_path_on_delivery was set, Exim crashed when
5422 bouncing a message because it had too many Received: header lines.
5424 20. If two routers both had headers_remove settings, and the first one included
5425 a superfluous trailing colon, the final name in the first list and the
5426 first name in the second list were incorrectly joined into one item (with a
5427 colon in the middle).
5433 1. Added -C and -D options to the exinext utility, mainly to make it easier
5434 to include in the automated testing, but these could be helpful when
5435 multiple configurations are in use.
5437 2. The exinext utility was not formatting the output nicely when there was
5438 an alternate port involved in the retry record key, nor when there was a
5439 message id as well (for retries that were specific to a specific message
5440 and a specific host). It was also confused by IPv6 addresses, because of
5441 the additional colons they contain. I have fixed the IPv4 problem, and
5442 patched it up to do a reasonable job for IPv6.
5444 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5445 delivery, the log line now contains "pipelined" if PIPELINING was used.
5447 4. An SMTP transport process used to panic and die if the bind() call to set
5448 an explicit outgoing interface failed. This has been changed; it is now
5449 treated in the same way as a connect() failure.
5451 5. A reference to $sender_host_name in the part of a conditional expansion
5452 that was being skipped was still causing a DNS lookup. This no longer
5455 6. The def: expansion condition was not recognizing references to header lines
5456 that used bh_ and bheader_.
5458 7. Added the _cache feature to named lists.
5460 8. The code for checking quota_filecount in the appendfile transport was
5461 allowing one more file than it should have been.
5463 9. For compatibility with Sendmail, the command line option
5471 and sets the incoming protocol and host name (for trusted callers). The
5472 host name and its colon can be omitted when only the protocol is to be set.
5473 Note the Exim already has two private options, -pd and -ps, that refer to
5474 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5475 "s", but I don't think that's a major issue.
5477 10. A number of refactoring changes to the code, none of which should affect
5480 (a) The number of logging options was getting close to filling up the
5481 32-bit word that was used as a bit map. I have split them into two classes:
5482 those that are passed in the argument to log_write(), and those that are
5483 only ever tested independently outside of that function. These are now in
5484 separate 32-bit words, so there is plenty of room for expansion again.
5485 There is no change in the user interface or the logging behaviour.
5487 (b) When building, for example, log lines, the code previously used a
5488 macro that called string_cat() twice, in order to add two strings. This is
5489 not really sufficiently general. Furthermore, there was one instance where
5490 it was actually wrong because one of the argument was used twice, and in
5491 one call a function was used. (As it happened, calling the function twice
5492 did not affect the overall behaviour.) The macro has been replaced by a
5493 function that can join an arbitrary number of extra strings onto a growing
5496 (c) The code for expansion conditions now uses a table and a binary chop
5497 instead of a serial search (which was left over from when there were very
5498 few conditions). Also, it now recognizes conditions like "pam" even when
5499 the relevant support is not compiled in: a suitably worded error message is
5500 given if an attempt is made to use such a condition.
5502 11. Added ${time_interval:xxxxx}.
5504 12. A bug was causing one of the ddress fields not to be passed back correctly
5505 from remote delivery subprocesses. The field in question was not being
5506 subsequently used, so this caused to problems in practice.
5508 13. Added new log selectors queue_time and deliver_time.
5510 14. Might have fixed a bug in maildirsizefile handling that threw up
5511 "unexpected character" debug warnings, and recalculated the data
5512 unnecessarily. In any case, I expanded the warning message to give more
5515 15. Added the message "Restricted characters in address" to the statements in
5516 the default ACL that block characters like @ and % in local parts.
5518 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5519 Three changes have been made:
5521 (a) There was a serious bug; a negative response to MAIL caused the whole
5522 recipient domain to be cached as invalid, thereby blocking all messages
5523 to all local parts at the same domain, from all senders. This bug has
5524 been fixed. The domain is no longer cached after a negative response to
5525 MAIL if the sender used is not empty.
5527 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5530 (c) A new callout option, "use_sender" has been added for people who want
5531 the modified behaviour.
5537 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5540 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5541 indeed breaks things for older releases.
5543 3. Added additional logging to the case where there is a problem reading data
5544 from a filter that is running in a subprocess using a pipe, in order to
5545 try to track down a specific problem.
5547 4. Testing facility fudge: when running in the test harness and attempting
5548 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5549 getting "No route to host". Convert this to a timeout.
5551 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5554 6. Some OS don't have socklen_t but use size_t instead. This affects the
5555 fifth argument of getsockopt() amongst other things. This is now
5556 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5557 can be set for individual OS. I have set it for SunOS5, OSF1, and
5558 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5559 some earlier ones do not.
5561 7. Change 4.30/15 was not doing the test caselessly.
5563 8. The standard form for an IPv6 address literal was being rejected by address
5564 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5565 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5566 this, as well as the form without the "IPv6" on the front (but only when
5567 address literals are enabled, of course).
5569 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5571 10. Exim crashed if a message with an empty sender address specified by -f
5572 encountered a router with an errors_to setting. This could be provoked only
5573 by a command such as
5577 where an empty string was supplied; "<>" did not hit this bug.
5579 11. Installed PCRE release 4.5.
5581 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5582 remained set. It is now erased.
5584 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5585 times from message ids (which are base 36 rather than the normal 62).
5587 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5588 were being counted as actual protocol errors, and logged if the log
5589 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5590 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5591 and DATA following a set of rejected RCPTs do not count as protocol errors.
5592 In other words, Exim assumes they were pipelined, though this may not
5593 actually be the case. Of course, in all cases the client gets an
5594 appropriate error code.
5596 15. If a lookup fails in an ACL condition, a message about the failure may
5597 be available; it is used if testing the ACL cannot continue, because most
5598 such messages specify what the cause of the deferral is. However, some
5599 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5600 that caused an old message to be retained and used if a later statement
5601 caused a defer, replacing the real cause of the deferral.
5603 16. If an IP address had so many PTR records that the DNS lookup buffer
5604 was not large enough to hold them, Exim could crash while trying to process
5605 the truncated data. It now detects and logs this case.
5607 17. Further to 4.21/58, another change has been made: if (and only if) the
5608 first line of a message (the first header line) ends with CRLF, a bare LF
5609 in a subsequent header line has a space inserted after it, so as not to
5610 terminate the header.
5612 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5613 unnecessarily, used atoi() instead of strtol(), and didn't check the
5614 termination when getting file sizes from file names by regex.
5616 19. Completely re-implemented the support for maildirsize files, in the light
5617 of a number of problems with the previous contributed implementation
5618 (4.30/29). In particular:
5620 . If the quota is zero, the maildirsize file is maintained, but no quota is
5623 . If the maildir directory does not exist, it is created before any attempt
5624 to write a maildirsize file.
5626 . The quota value in the file is just a cache; if the quota is changed in
5627 the transport, the new value overrides.
5629 . A regular expression is available for excluding directories from the
5632 20. The autoreply transport checks the characters in options that define the
5633 message's headers; it allows continued headers, but it was checking with
5634 isspace() after an embedded newline instead of explicitly looking for a
5637 21. If all the "regular" hosts to which an address was routed had passed their
5638 expiry times, and had not reached their retry times, the address was
5639 bounced, even if fallback hosts were defined. Now Exim should go on to try
5642 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5643 equivalent code in the SMTP transport. Some hosts send humungous responses
5644 to HELO/EHLO, more than 1024 it seems.
5646 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5647 gives compiler warnings in some environments. I've now done it "properly",
5650 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5651 (because of problems with the built-in one) was declared to return uschar *
5652 instead of char *, causing compiler failure.
5654 25. Fixed a file descriptor leak when processing alias/forward files.
5656 26. Fixed a minor format string issue in dbfn.c.
5658 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5660 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5661 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5664 29. When a sender address is verified, it is cached, to save repeating the test
5665 when there is more than one recipient in a message. However, when the
5666 verification involves a callout, it is possible for different callout
5667 options to be set for different recipients. It is too complicated to keep
5668 track of this in the cache, so now Exim always runs a verification when a
5669 callout is required, relying on the callout cache for the optimization.
5670 The overhead is duplication of the address routing, but this should not be
5673 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5674 to be verified with callout=postmaster, and the main callout worked but the
5675 postmaster check failed, the verification correctly failed. However, if a
5676 subsequent RCPT command asked for sender verification *without* the
5677 postmaster check, incorrect caching caused this verification also to fail,
5680 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5681 it was not caching the DNS options (qualify_single, search_parents) that
5682 were used when the lookup failed. A subsequent lookup with different
5683 options therefore always gave the same answer, though there were cases
5684 where it should not have. (Example: a "domains = !$mx_any" option on a
5685 dnslookup router: the "domains" option is always processed without any
5686 widening, but the router might have qualify_single set.) Now Exim uses the
5687 cached value only when the same options are set.
5689 32. Added John Jetmore's "exipick" utility to the distribution.
5691 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5692 than a timeout (e.g. a certificate is required, and is not provided), an
5693 Exim server now closes the connection immediately. Previously it waited for
5694 the client to close - but if the client is SSL, it seems that they each
5695 wait for each other, leading to a delay before one of them times out.
5697 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5698 maintained 0.8.x compatibility because I don't think many are using it, and
5699 it is clearly obsolete.
5701 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5704 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5705 client certificate was expired. A simple patch fixes this, though I don't
5706 understand the full logic of why the verify callback is called multiple
5709 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5710 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5711 which causes problems with some clients (such as the Certicom SSL Plus
5712 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5713 disables the coutermeasure allowing Eudora to connect."
5715 38. Exim was not checking that a write() to a log file succeeded. This could
5716 lead to Bad Things if a log got too big, in particular if it hit a file
5717 size limit. Exim now panics and dies if it cannot write to a log file, just
5718 as it does if it cannot open a log file.
5720 39. Modified OS/Makefile-Linux so that it now contains
5722 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5724 The two -D definitions ensure that Exim is compiled with large file
5725 support, which makes it possible to handle log files that are bigger than
5728 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5729 instance) a domain was checked against a named list that involved a lookup,
5730 causing $domain_data to be set, then another domain was checked against the
5731 same list, then the first domain was re-checked, the value of $domain_data
5732 after the final check could be wrong. In particular, if the second check
5733 failed, it could be set empty. This bug probably also applied to
5736 41. The strip_trailing_dot option was not being applied to the address given
5737 with the -f command-line option.
5739 42. The code for reading a message's header from the spool was incrementing
5740 $received_count, but never initializing it. This meant that the value was
5741 incorrect (doubled) while delivering a message in the same process in which
5742 it was received. In the most common configuration of Exim, this never
5743 happens - a fresh exec is done - but it can happen when
5744 deliver_drop_privilege is set.
5746 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5747 it now includes up to 150 characters of the unexpected data in the log
5750 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5751 and building data strings. The size of both of these buffers was 10 000
5752 bytes - far larger than anybody would *ever* want, thought I. Needless to
5753 say, somebody hit the limit. I have increased the maximum line length to
5754 20 000 and the maximum data length of concatenated lines to 100 000. I have
5755 also fixed two bugs, because there was no checking on these buffers. Tsk,
5756 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5757 buffer is too small.
5759 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5760 lsearch lookups. Now it does.
5762 46. When parsing a route_list item in a manualroute router, a fixed-length
5763 buffer was used for the list of hosts. I made this 1024 bytes long,
5764 thinking that nobody would ever have a list of hosts that long. Wrong.
5765 Somebody had a whole pile of complicated expansion conditions, and the
5766 string was silently truncated, leading to an expansion error. It turns out
5767 that it is easier to change to an unlimited length (owing to other changes
5768 that have happened since this code was originally written) than to build
5769 structure for giving a limitation error. The length of the item that
5770 expands into the list of hosts is now unlimited.
5772 47. The lsearch lookup could not handle data where the length of text line was
5773 more than 4095 characters. Such lines were truncated, leading to shortened
5774 data being returned. It should now handle lines of any length.
5776 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5777 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5780 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5781 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5782 no longer happens in comment lines. A list of replacements is now placed
5783 at the head of all of the source files, except those whose only change is
5784 to replace PERL_COMMAND in the very first #! line.
5786 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5787 messages on the queue, with a bottom-up merge sort, using code contributed
5788 by Michael Haardt. This should make operations like -bp somewhat faster on
5789 large queues. It won't affect queue runners, except when queue_run_in_order
5792 51. Installed eximstats 1.31 in the distribution.
5794 52. Added support for SRV lookups to the dnslookup router.
5796 53. If an ACL referred to $message_body or $message_body_end, the value was not
5797 reset for any messages that followed in the same SMTP session.
5799 54. The store-handling optimization for building very long strings was not
5800 differentiating between the different store pools. I don't think this
5801 actually made any difference in practice, but I've tidied it.
5803 55. While running the routers to verify a sender address, $sender_address
5804 was still set to the sender address. This is wrong, because when routing to
5805 send a bounce to the sender, it would be empty. Therefore, I have changed
5806 it so that, while verifying a sender address, $sender_address is set to <>.
5807 (There is no change to what happens when verifying a recipient address.)
5809 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5810 target A or AAAA records (if not already returned) without resetting the
5811 qualify_single or search_parents options of the DNS resolver. These are
5812 inappropriate in this case because the targets of MX and SRV records must
5813 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5814 target that, when qualified, matched something in the local domain. These
5815 two options are now turned off when doing these lookups.
5817 57. It seems that at least some releases of Reiserfs (which does not have the
5818 concept of a fixed number of inodes) returns zero and not -1 for the
5819 number of available inodes. This interacted badly with check_spool_inodes,
5820 which assumed that -1 was the "no such thing" setting. What I have done is
5821 to check that the total number of inodes is greater than zero before doing
5822 the test of how many are available.
5824 58. When a "warn" ACL statement has a log_message modifier, the message is
5825 remembered, and not repeated. This is to avoid a lot of repetition when a
5826 message has many recipients that cause the same warning to be written.
5827 However, Exim was preserving the list of already written lines for an
5828 entire SMTP session, which doesn't seem right. The memory is now reset if a
5829 new message is started.
5831 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5832 debugging output unless log_rewrite was also set.
5834 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5835 of (int)(handle) when we know that handle contains (void *)(-1).
5837 61. The Exim daemon panic-logs an error return when it closes the incoming
5838 connection. However "connection reset by peer" seems to be common, and
5839 isn't really an error worthy of noting specially, so that particular error
5842 62. When Exim is trying to find all the local interfaces, it used to panic and
5843 die if the ioctl to get the interface flags failed. However, it seems that
5844 on at least one OS (Solaris 9) it is possible to have an interface that is
5845 included in the list of interfaces, but for which you get a failure error
5846 for this call. This happens when the interface is not "plumbed" into a
5847 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5848 failure of the "get flags" call assumes that the interface is down.
5850 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5851 makes life easier for people who are doing arithmetic on fields extracted
5852 from dates, where you often get leading zeros that should not be
5853 interpreted as octal.
5855 64. Added qualify_domain to the redirect router, to override the global
5858 65. If a pathologically long header line contained very many addresses (the
5859 report of this problem mentioned 10 000) and each of them was rewritten,
5860 Exim could use up a very large amount of memory. (It kept on making new
5861 copies of the header line as it rewrote, and never released the old ones.)
5862 At the expense of a bit more processing, the header rewriting function has
5863 been changed so that it no longer eats memory in this way.
5865 66. The generation of the Received: header has been moved from the time that a
5866 message starts to be received, to the time that it finishes. The timestamp
5867 in the Received: header should now be very close to that of the <= log
5868 line. There are two side-effects of this change:
5870 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5871 logged header lines no longer include the local Received: line, because
5872 it has not yet been created. The same applies to a copy of the message
5873 that is returned to a non-SMTP sender when a message is rejected.
5875 (b) When a filter file is tested using -bf, no additional Received: header
5876 is added to the test message. After some thought, I decided that this
5879 This change does not affect the value of $received_for. It is still set
5880 after address rewriting, but before local_scan() is called.
5882 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5884 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5885 gave an unhelpful panic error message, and a defer error. I have managed to
5886 change this behaviour so that it now rejects any supplied certificate,
5887 which seems right, as the list of acceptable certificates is empty.
5889 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5890 gave an unhelpful defer error. I have not managed to make this reject any
5891 supplied certificates, but the error message it gives is "no certificate
5892 supplied", which is not helpful.
5894 70. exigrep's output now also includes lines that are not associated with any
5895 message, but which match the given pattern. Implemented by a patch from
5896 Martin Sluka, which also tidied up the Perl a bit.
5898 71. Recipient callout verification, like sender verification, was using <> in
5899 the MAIL FROM command. This isn't really the right thing, since the actual
5900 sender may affect whether the remote host accepts the recipient or not. I
5901 have changed it to use the actual sender in the callout; this means that
5902 the cache record is now keyed on a recipient/sender pair, not just the
5903 recipient address. There doesn't seem to be a real danger of callout loops,
5904 since a callout by the remote host to check the sender would use <>.
5905 [SEE ABOVE: changed after hitting problems.]
5907 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5908 temporary errors. However, in the case of such a code being given after
5909 the end of a data transmission (i.e. after ".") Exim was failing to write
5910 a retry record for the message. (Yes, there was some broken host that was
5911 actually sending 8xx at this point.)
5913 73. An unknown lookup type in a host list could cause Exim to panic-die when
5914 the list was checked. (An example that provoked this was putting <; in the
5915 middle of a list instead of at the start.) If this happened during a DATA
5916 ACL check, a -D file could be left lying around. This kind of configuration
5917 error no longer causes Exim to die; instead it causes a defer error. The
5918 incident is still logged to the main and panic logs.
5920 74. Buglet left over from Exim 3 conversion. The message "too many messages
5921 in one connection" was written to the rejectlog but not the mainlog, except
5922 when address rewriting (yes!) was being logged.
5924 75. Added write_rejectlog option.
5926 76. When a system filter was run not as root (that is, when system_filter_user
5927 was set), the values of the $n variables were not being returned to the
5928 main process; thus, they were not subsequently available in the $sn
5931 77. Added +return_path_on_delivery log selector.
5933 78. A connection timeout was being treated differently from recipients deferred
5934 when testing hosts_max_try with a message that was older than the host's
5935 retry timeout. (The host should not be counted, thus allowing all hosts to
5936 be tried at least once before bouncing.) This may have been the cause of an
5937 occasionally reported bug whereby a message would remain on the queue
5938 longer than the retry timeout, but would be bounced if a delivery was
5939 forced. I say "may" because I never totally pinned down the problem;
5940 setting up timeout/retry tests is difficult. See also the next item.
5942 79. The ultimate address timeout was not being applied to errors that involved
5943 a combination of host plus message (for example, a timeout on a MAIL
5944 command). When an address resolved to a number of possible hosts, and they
5945 were not all tried for each delivery (e.g. because of hosts_max_try), a
5946 message could remain on the queue longer than the retry timeout.
5948 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5951 81. Fixed an obscure SMTP outgoing bug which required at least the following
5952 conditions: (a) there was another message waiting for the same server;
5953 (b) the server returned 5xx to all RCPT commands in the first message so
5954 that the message was not completed; (c) the server dropped the connection
5955 or gave a negative response to the RSET that Exim sends to abort the
5956 transaction. The observed case was a dropped connection after DATA that had
5957 been sent in pipelining mode. That is, the server had advertised PIPELINING
5958 but was not implementing it correctly. The effect of the bug was incorrect
5959 behaviour, such as trying another host, and this could lead to a crash.
5965 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5966 and daemon.c were passed as pointers to ints; they should have been
5967 pointers to socklen_t variables (which are typically unsigned ints).
5969 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5972 3. Fixed a really odd bug that affected only the testing scheme; patching a
5973 certain fixed string in the binary changed the value of another string that
5974 happened to be identical to the end of the original first string.
5976 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5977 name", it returns that address as the IP address. On some operating
5978 systems (e.g. Solaris), it also passes back the IP address string as the
5979 "host name". However, on others (e.g. Linux), it passes back an empty
5980 string. Exim wasn't checking for this, and was changing the host name to an
5981 empty string, assuming it had been canonicalized.
5983 5. Although rare, it is permitted to have more than one PTR record for a given
5984 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5985 all the names associated with an address, because they do in Solaris.
5986 However, it seems that they do not in Linux for data that comes from the
5987 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5988 I found this out when I moved to a new Linux workstation and tried to run
5989 the Exim test suite.
5991 To get round this problem I have changed the code so that it now does its
5992 own call to the DNS to look up PTR records when searching for a host name.
5993 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5994 addresses that are only in /etc/hosts are still found.
5996 This behaviour is, however, controlled by an option called host_lookup_
5997 order, which defaults to "bydns:byaddr". If people want to use the other
5998 order, or indeed, just use one or the other means of lookup, they can
5999 specify it in this variable.
6001 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6002 some operating systems, this comes back from gethostbyaddr() as an empty
6003 string, and this is what Exim used to test for. However, it seems that in
6004 other systems, "." is yielded. Exim now tests for this case too.
6006 7. The values of check_spool_space and check_log_space are now held internally
6007 as a number of kilobytes instead of an absolute number of bytes. If a
6008 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6009 kilobyte. This means that much larger values can be stored.
6011 8. Exim monitor: an attempt to get the action menu when not actually pointing
6012 at a message produces an empty menu entitled "No message selected". This
6013 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6014 no entries in it ("Shell widget menu has zero width and/or height"). So I
6015 have added a single, blank menu entry in this case.
6017 9. Added ${quote_local_part.
6019 10. MIME decoding is now applied to the contents of Subject: header lines when
6022 11. Now that a reference to $sender_host_address automatically causes a reverse
6023 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6024 host lookup before query-style lookups in lists that might use this
6025 variable. This has therefore been abolished, and the "net-" prefix is no
6026 longer necessary for query-style lookups.
6028 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6029 have been a typo for LFLAGS, so it has been changed.
6031 13. The install script calls Exim with "-C /dev/null" in order to find the
6032 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6033 to be output. However, since Exim outputs its version number before the
6034 error, it didn't break the script. It just looked ugly. I fixed this by
6035 always allowing "-C /dev/null" if the caller is root.
6037 14. Ignore overlarge ACL variable number when reading spool file - insurance
6038 against a later release with more variables having written the file.
6040 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6041 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6042 this, as well as the form without the "IPv6" on the front.
6044 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6045 OS/Makefile-Darwin file.
6047 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6048 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6050 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6051 variable. However, some OS use ints for inodes. I've added cast to long int
6052 to get rid of the compiler warning.
6054 19. I had forgotten to lock out "/../" in configuration file names when
6055 ALT_CONFIG_PREFIX was set.
6057 20. Routers used for verification do not need to specify transports. However,
6058 if such a router generated a host list, and callout was configured, Exim
6059 crashed, because it could not find a port number from the (non-existent)
6060 transport. It now assumes port 25 in this circumstance.
6062 21. Added the -t option to exigrep.
6064 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6065 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6066 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6067 them is set, LOOKUP_LSEARCH is forced.
6069 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6071 24. Added sender and host information to the "rejected by local_scan()" log
6072 line; previously there was no indication of these.
6074 25. Added .include_if_exists.
6076 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6077 Linux. It turns out that not all file systems support this. Apparently some
6078 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6079 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6080 not supported on the file descriptor, is now ignored when Exim is trying to
6081 sync a directory. This applies only to Linux.
6083 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6085 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6086 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6089 29. Added support for maildirsize files from supplied patch (modified a bit).
6091 30. The use of :fail: followed by an empty string could lead Exim to respond to
6092 sender verification failures with (e.g.):
6094 550 Verification failed for <xxx>
6095 550 Sender verify failed
6097 where the first response line was missing the '-' that indicates it is not
6098 the final line of the response.
6100 31. The loop for finding the name of the user that called Exim had a hardwired
6101 limit of 10; it now uses the value of finduser_retries, which is used for
6102 all other user lookups.
6104 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6107 33. Exim was neglecting to zero errno before one call of strtol() when
6108 expanding a string and expecting an integer value. On some systems this
6109 resulted in spurious "integer overflow" errors. Also, it was casting the
6110 result into an int without checking.
6112 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6113 did not work. The code looks as if it has *never* worked, though it appears
6114 to have been documented since at least release 1.62. I have made it work.
6116 35. The "timeout_DNS" error in retry rules, also documented since at least
6117 1.62, also never worked. As it isn't clear exactly what this means, and
6118 clearly it isn't a major issue, I have abolished the feature by treating it
6119 as "timeout", and writing a warning to the main and panic logs.
6121 36. The display of retry rules for -brt wasn't always showing the error code
6124 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6125 timeout_connect_A, timeout_connect_MX.
6127 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6128 to the empty sender.
6130 39. The daemon was not analysing the content of -oX till after it had closed
6131 stderr and disconnected from the controlling terminal. This meant that any
6132 syntax errors were only noted on the panic log, and the return code from
6133 the command was 0. By re-arranging the code a little, I've made the
6134 decoding happen first, so such errors now appear on stderr, and the return
6135 code is 1. However, the actual setting up of the sockets still happens in
6136 the disconnected process, so errors there are still only recorded on the
6139 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6140 connections (as happens on some IP stacks) was logged at start up time as
6141 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6142 it from "IPv6 and IPv4", which means that two separate sockets are being
6145 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6146 says whether AF_INET or AF_INET6 was passed as an argument.
6148 42. Exiwhat output was messed up when time zones were included in log
6151 43. Exiwhat now gives more information about the daemon's listening ports,
6152 and whether -tls-on-connect was used.
6154 44. The "port" option of the smtp transport is now expanded.
6156 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6157 silently ignored. Now an error message is written to the main and panic
6160 46. There's a new ACL modifier called "logwrite" which writes to a log file
6161 as soon as it is encountered.
6163 47. Added $local_user_uid and $local_user_gid at routing time.
6165 48. Exim crashed when trying to verify a sender address that was being
6168 49. Exim was recognizing only a space character after ".include". It now also
6169 recognizes a tab character.
6171 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6172 extracting the relevant information from the specification. The man page no
6173 longer contains scrambled data for the -d option, and I've added a section
6174 at the front about calling Exim under different names.
6176 51. Added "extra_headers" argument to the "mail" command in filter files.
6178 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6181 53. Installed eximstats 1.29.
6183 54. Added transport_filter_timeout as a generic transport option.
6185 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6186 Cc: header lines. This was required by RFC 822, but it not required by RFC
6189 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6190 incoming messages that did not have them. Now it does so only if the
6191 message originates locally, that is, if there is no associated remote host
6192 address. When Resent- header lines are present, this applies to the Resent-
6193 lines rather than the non-Resent- lines.
6195 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6196 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6198 58. Messages for configuration errors now include the name of the main
6199 configuration file - useful now that there may be more than one file in a
6200 list (.included file names were always shown).
6202 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6203 for those rare installations that do not start the daemon as root or run it
6204 setuid root. I've cut out the call to initgroups() if the daemon is not
6207 60. The Exim user and group can now be bound into the binary as text strings
6208 that are looked up at the start of Exim's processing.
6210 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6212 62. Added $mailstore_basename variable.
6214 63. Installed patch to sieve.c from Michael Haardt.
6216 64. When Exim failed to open the panic log after failing to open the main log,
6217 the original message it was trying to log was written to stderr and debug
6218 output, but if they were not available (the usual case in production), it
6219 was lost. Now it is written to syslog before the two lines that record the
6220 failures to open the logs.
6222 65. Users' Exim filters run in subprocesses under the user's uid. It is
6223 possible for a "deliver" command or an alias in a "personal" command to
6224 provoke an address rewrite. If logging of address rewriting is configured,
6225 this fails because the process is not running as root or exim. There may be
6226 a better way of dealing with this, but for the moment (because 4.30 needs
6227 to be released), I have disabled address rewrite logging when running a
6228 filter in a non-root, non-exim process.
6234 1. The buildconfig auxiliary program wasn't quoting the value set for
6235 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6236 not defined. This bug was masked in 4.22 by the effect that was fixed in
6239 2. Some messages that were rejected after a message id was allocated were
6240 shown as "incomplete" by exigrep. It no longer does this for messages that
6241 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6243 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6244 have allow_domain_literals set, the ID did not get logged in the <= line.
6245 Domain literals are now always recognized in Message-ID: header lines.
6247 4. The first argument for a ${extract expansion item is the key name or field
6248 number. Leading and trailing spaces in this item were not being ignored,
6249 causing some misleading effects.
6251 5. When deliver_drop_privilege was set, single queue runner processes started
6252 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6253 same command in the process it spins off) were not dropping privilege.
6255 6. When the daemon running as "exim" started a queue runner, it always
6256 re-executed Exim in the spun-off process. This is a waste of effort when
6257 deliver_drop_privilege is set. The new process now just calls the
6258 queue-runner function directly.
6264 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6267 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6268 ignored. Though the use of -oP was forcing the writing of a pid file, it
6269 was always written to the default place.
6271 3. If the message "no IP address found for host xxxx" is generated during
6272 incoming verification, it is now followed by identification of the incoming
6273 connection (so you can more easily find what provoked it).
6275 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6277 5. Added some features to "harden" Exim a bit more against certain attacks:
6279 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6280 be put in Local/Makefile. This is like the never_users runtime option,
6281 but it cannot be overridden. The default setting is "root".
6283 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6284 prefix string with which any file named in a -C command line option
6287 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6288 is retained for -C and -D only if the caller of Exim is root. Without
6289 it, the exim user may also use -C and -D and retain privilege.
6291 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6292 command line option is disabled.
6294 6. Macro names set by the -D option must start with an upper case letter, just
6295 like macro names defined in the configuration file.
6297 7. Added "dereference=" facility to LDAP.
6299 8. Two instances of the typo "uknown" in the source files are fixed.
6301 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6302 the Configure-Makefile script screwed up while processing it.
6304 10. Incorporated PCRE 4.4.
6306 11. The SMTP synchronization check was not operating right at the start of an
6307 SMTP session. For example, it could not catch a HELO sent before the client
6308 waited for the greeting. There is now a check for outstanding input at the
6309 point when the greeting is written. Because of the duplex, asynchronous
6310 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6311 way, but not yet received, when the check is performed.
6313 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6314 on TCP/IP sockets, because this apparently causes some broken clients to
6317 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6318 unchanged) from the Cygwin maintainer.
6320 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6321 was supported instead of testing for mbx. Effectively a typo.
6323 15. The spa authenticator server code was not checking that the input it
6324 received was valid base64.
6326 16. The debug output line for the "set" modifier in ACLs was not showing the
6327 name of the variable that was being set.
6329 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6331 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6332 lookup on its own. Something else was needed to trigger the lookup. For
6333 example, a match in host_lookup or the need for a host name in a host list.
6334 Now, if $sender_host_name is referenced and the host name has not yet been
6335 looked up, a lookup is performed. If the lookup fails, the variable remains
6336 empty, and $host_lookup_failed is set to "1".
6338 19. Added "eqi" as a case-independent comparison operator.
6340 20. The saslauthd authentication condition could segfault if neither service
6341 nor realm was specified.
6343 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6344 error message that was logged was misleading, and incoming SMTP
6345 connections were dropped. The message is now more accurate, and temporary
6346 errors are given to SMTP connections.
6348 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6349 (except RSET) with a 421 error, until QUIT is received. However, it was
6350 failing to send a response to QUIT.
6352 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6353 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6354 ACL is now run after the helo_try_verify_hosts code.
6356 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6357 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6358 case-independent, so other case variants are also recognized). Apparently
6359 some systems use these upper case variants.
6361 25. If more than two messages were waiting for the same host, and a transport
6362 filter was specified for the transport, Exim sent two messages over the
6363 same TCP/IP connection, and then failed with "socket operation on non-
6364 socket" when it tried to send the third.
6366 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6368 27. The extern definition of crypt16() in expand.c was not being excluded when
6369 the OS had its own crypt16() function.
6371 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6372 as a preferred synonym for return_size_limit, both as an option and as an
6375 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6377 30. Changed the default configuration ACL to relax the local part checking rule
6378 for addresses that are not in any local domains. For these addresses,
6379 slashes and pipe symbols are allowed within local parts, but the sequence
6380 /../ is explicitly forbidden.
6382 31. SPA server authentication was not clearing the challenge buffer before
6385 32. log_message in a "warn" ACL statement was writing to the reject log as
6386 well as to the main log, which contradicts the documentation and doesn't
6387 seem right (because no rejection is happening). So I have stopped it.
6389 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6390 However, I am unable to do any testing of this.
6392 34. Fixed an infelicity in the appendfile transport. When checking directories
6393 for a mailbox, to see if any needed to be created, it was accidentally
6394 using path names with one or more superfluous leading slashes; tracing
6395 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6397 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6398 contents are added to the log line that is written for every discarded
6399 recipient. (Previously a log_message setting was ignored.)
6401 36. The ${quote: operator now quotes the string if it is empty.
6403 37. The install script runs exim in order to find its version number. If for
6404 some reason other than non-existence or emptiness, which it checks, it
6405 could not run './exim', it was installing it with an empty version number,
6406 i.e. as "exim-". This error state is now caught, and the installation is
6409 38. An argument was missing from the function that creates an error message
6410 when Exim fails to connect to the socket for saslauthd authentication.
6411 This could cause Exim to crash, or give a corrupted message.
6413 39. Added isip, isip4, and isip6 to ${if conditions.
6415 40. The ACL variables $acl_xx are now saved with the message, and can be
6416 accessed later in routers, transports, and filters.
6418 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6419 strings in the file are not string-expanded.
6421 42. If a MAIL command specified a SIZE value that was too large to fit into an
6422 int variable, the check against message_size_limit failed. Such values are
6423 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6424 day this will have to be increased, but I don't think I want to be around
6425 when emails are that large.
6432 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6433 iconv() is not standard in FreeBSD.
6435 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6436 IPv6 enabled. The observed symptom was a segmentation fault on return from
6437 the function os_common_find_running_interfaces() in src/os.c.
6439 3. In the check_special_case() function in daemon.c I had used "errno" as an
6440 argument name, which causes warnings on some systems. This was basically a
6441 typo, since it was named "eno" in the comments!
6443 4. The code that waits for the clock to tick (at a resolution of some fraction
6444 of a second) so as to ensure message-id uniqueness was always waiting for
6445 at least one whole tick, when it could have waited for less. [This is
6446 almost certainly not relevant at current processor speeds, where it is
6447 unlikely to ever wait at all. But we try to future-proof.]
6449 5. The function that sleeps for a time interval that includes fractions of a
6450 second contained a race. It did not block SIGALRM between setting the
6451 timer, and suspending (a couple of lines later). If the interval was short
6452 and the sigsuspend() was delayed until after it had expired, the suspension
6453 never ended. On busy systems this could lead to processes getting stuck for
6456 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6457 process, before it forks any delivery processes. The open lookup caching
6458 mechanism meant that the open file or database connection was passed into
6459 the delivery process. The problem was that delivery processes always tidy
6460 up cached lookup data. This could cause a problem for the next delivery
6461 process started by the queue runner, because the external queue runner
6462 process does not know about the closure. So the next delivery process
6463 still has data in the lookup cache. In the case of a file lookup, there was
6464 no problem because closing a file descriptor in a subprocess doesn't affect
6465 the parent. However, if the lookup was caching a connection to a database,
6466 the connection was closed, and the second delivery process was likely to
6467 see errors such as "PGSQL: query failed: server closed the connection
6468 unexpectedly". The problem has been fixed by closing all cached lookups
6469 in a queue runner before running a delivery process.
6471 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6472 seem to have the "const" qualifier which it has on other OS. I've
6475 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6476 *of the same type* (client or server) with the same public name that an
6477 error should be diagnosed.
6479 9. When Exim looked up a host name for an IP address, but failed to find the
6480 original IP address when looking up the host name (a safety check), it
6481 output the message "<ip address> does not match any IP for NULL", which was
6482 confusing, to say the least. The bug was that the host name should have
6483 appeared instead of "NULL".
6485 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6486 user that is built into the binary, and the -C or -D options is used, root
6487 privilege is dropped before the configuration file is read. In addition,
6488 logging is switched to stderr instead of the normal log files. If the
6489 configuration then re-defines the Exim user, the unprivileged environment
6490 is probably not what is expected, so Exim logs a panic warning message (but
6493 However, if deliver_drop_privilege is set, the unprivileged state may well
6494 be exactly what is intended, so the warning has been cut out in that case,
6495 and Exim is allowed to try to write to its normal log files.
6501 1. smtp_return_error_details was not giving details for temporary sender
6502 or receiver verification errors.
6504 2. Diagnose a configuration error if two authenticators have the same public
6507 3. Exim used not to create the message log file for a message until the first
6508 delivery attempt. This could be confusing when incoming messages were held
6509 for policy or load reasons. The message log file is now created at the time
6510 the message is received, and an initial "Received" line is written to it.
6512 4. The automatically generated man page for command line options had a minor
6513 bug that caused no ill effects; however, a more serious problem was that
6514 the procedure for building the man page automatically didn't always
6515 operate. Consequently, release 4.20 contains an out-of-date version. This
6516 shouldn't happen again.
6518 5. When building Exim with embedded Perl support, the script that builds the
6519 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6520 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6522 6. The freeze_tell option was not being used for messages that were frozen on
6523 arrival, either by an ACL or by local_scan().
6525 7. Added the smtp_incomplete_transaction log selector.
6527 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6528 was accepting AUTH without a new EHLO.
6530 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6531 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6534 10. Logging of TCP/IP connections (when configured) now happens in the main
6535 daemon process instead of the child process, so that the TCP/IP connection
6536 count is more accurate (but it can never be perfect).
6538 11. The use of "drop" in a nested ACL was not being handled correctly in the
6539 outer ACL. Now, if condition failure induced by the nested "drop" causes
6540 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6541 or "require"), the connection is dropped.
6543 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6544 that yield "discard" can now be used with an "accept" or a "discard" verb,
6545 but an error is generated for any others (because I can't see a useful way
6546 to define what should happen).
6548 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6549 are now processed in the same way as lines in the Exim configuration file.
6550 In particular, continuation lines are supported.
6552 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6554 15. Added -ti meaning -t -i.
6556 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6557 domains, and warn by logging if others are found.
6559 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6560 returned from GIFCONF when Exim is trying to find the list of interfaces on
6561 a host. The code in os.c has been modified to copy each ifreq to an aligned
6562 structure in all cases.
6564 Also, in some cases, the returned ifreq's were being copied to a 'struct
6565 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6566 means the last couple of bytes of an IPv6 address could be chopped if the
6567 ifreq contained only a normal sockaddr (14 bytes storage).
6569 18. Named domain lists were not supported in the hosts_treat_as_local option.
6570 An entry such as +xxxx was not recognized, and was treated as a literal
6573 19. Ensure that header lines added by a DATA ACL are included in the reject log
6574 if the ACL subsequently rejects the message.
6576 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6577 MD5 (which is deprecated).
6579 21. When testing a filter file using -bf, Exim was writing a message when it
6580 took the sender from a "From " line in the message, but it was not doing so
6581 when it took $return_path from a Return-Path: header line. It now does.
6583 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6584 with a valid header line field name (a series of printing characters
6585 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6587 23. Changed "disc" in the source to "disk" to conform to the documentation and
6588 the book and for uniformity.
6590 24. Ignore Sendmail's -Ooption=value command line item.
6592 25. When execve() failed while trying to run a command in a pipe transport,
6593 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6594 could be confused with a return value of 69 from the command itself. This
6595 has been changed to 127, the value the shell returns if it is asked to run
6596 a non-existent command. The wording for the related log line suggests a
6597 non-existent command as the problem.
6599 26. If received_header_text expands to an empty string, do not add a Received:
6600 header line to the message. (Well, it adds a token one on the spool, but
6601 marks it "old" so that it doesn't get used or transmitted.)
6603 27. Installed eximstats 1.28 (addition of -nt option).
6605 28. There was no check for failure on the call to getsockname() in the daemon
6606 code. This can fail if there is a shortage of resources on the system, with
6607 ENOMEM, for example. A temporary error is now given on failure.
6609 29. Contrary to the C standard, it seems that in some environments, the
6610 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6611 program. Exim now does this explicitly; it affects the formatting of
6612 timestamps using strftime().
6614 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6615 complaints. I've now initialized all the variables, to avoid this.
6617 32. Header lines added by a system filter were not being "seen" during
6618 transport-time rewrites.
6620 33. The info_callback() function passed to OpenSSL is set up with type void
6621 (*)(SSL *, int, int), as described somewhere. However, when calling the
6622 function (actually a macro) that sets it up, the type void(*)() is
6623 expected. I've put in a cast to prevent warnings from picky compilers.
6625 34. If a DNS black list lookup found a CNAME record, but there were no A
6626 records associated with the domain it pointed at, Exim crashed.
6628 35. If a DNS black list lookup returned more than one A record, Exim ignored
6629 all but the first. It now scans all returned addresses if a particular IP
6630 value is being sought. In this situation, the contents of the
6631 $dnslist_value variable are a list of all the addresses, separated by a
6634 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6635 to accept a host name and all its aliases if the forward lookup for any of
6636 them yielded the IP address of the incoming connection. Now it accepts only
6637 those names whose forward lookup yields the correct IP address. Any other
6638 names are discarded. This closes a loophole whereby a rogue DNS
6639 administrator could create reverse DNS records to break through a
6640 wildcarded host restriction in an ACL.
6642 37. If a user filter or a system filter that ran in a subprocess used any of
6643 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6644 the wrong values were passed to the pipe command ($thisaddress had the
6645 value of $0, $0 had the value of $1, etc). This bug was introduced by
6646 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6648 38. Improved the line breaking for long SMTP error messages from ACLs.
6649 Previously, if there was no break point between 40 and 75 characters, Exim
6650 left the rest of the message alone. Two changes have been made: (a) I've
6651 reduced the minimum length to 35 characters; (b) if it can't find a break
6652 point between 35 and 75 characters, it looks ahead and uses the first one
6653 that it finds. This may give the occasional overlong line, but at least the
6654 remaining text gets split now.
6656 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6657 file descriptors might be low, and that setting 1000 would always raise it.
6658 It turns out that in some environments, the limit is already over 1000 and
6659 that lowering it causes trouble. So now Exim takes care not to decrease it.
6661 40. When delivering a message, the value of $return_path is set to $sender_
6662 address at the start of routing (routers may change the value). By an
6663 oversight, this default was not being set up when an address was tested by
6664 -bt or -bv, which affected the outcome if any router or filter referred to
6667 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6668 log only when "message" or "log_message" are set. However, if one of the
6669 conditions was an address verification, or a call to a nested ACL, the
6670 messages generated by the underlying test were being passed through. This
6671 no longer happens. The underlying message is available in $acl_verify_
6672 message for both "message" and "log_message" expansions, so it can be
6673 passed through if needed.
6675 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6676 new expansion $bh_ to give the encoded byte string without charset
6677 translation. Translation happens only if iconv() is available; HAVE_ICONV
6678 indicates this at build time. HEADERS_CHARSET gives the charset to
6679 translate to; headers_charset can change it in the configuration, and
6680 "headers charset" can change it in an individual filter file.
6682 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6683 that creates RFC 2047 encoded "words" labels them as that charset instead
6684 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6685 expansion operator; (ii) when Exim creates a From: line for a local
6686 message; (iii) when a header line is rewritten to include a "phrase" part.
6688 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6689 buggy, causing it to skip the first lines of messages whose message ID
6690 ended in 'D'. This would not have bitten before Exim release 4.14, because
6691 message IDs were unlikely to end in 'D' before then. The effect was to have
6692 incorrect size information for certain domains.
6694 45. #include "config.h" was missing at the start of the crypt16.c module. This
6695 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6698 46. If there was a timeout during a "random" callout check, Exim treated it as
6699 a failure of the random address, and carried on sending RSET and the real
6700 address. If the delay was just some slowness somewhere, the response to the
6701 original RCPT would be taken as a response to RSET and so on, causing
6702 mayhem of various kinds.
6704 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6705 when I implemented it. It didn't allow for the fact that some option values
6706 may legitimately be negative (e.g. size_addition), and it didn't even do
6707 the right test for positive values.
6709 48. Domain names in DNS records are case-independent. Exim always looks them up
6710 in lower case. Some resolvers return domain names in exactly the case they
6711 appear in the zone file, that is, they may contain uppercase letters. Not
6712 all resolvers do this - some return always lower case. Exim was treating a
6713 change of case by a resolver as a change of domain, similar to a widening
6714 of a domain abbreviation. This triggered its re-routing code and so it was
6715 trying to route what was effectively the same domain again. This normally
6716 caused routing to fail (because the router wouldn't handle the domain
6717 twice). Now Exim checks for this case specially, and just changes the
6718 casing of the domain that it ultimately uses when it transmits the message
6721 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6724 50. If a filter generated a file delivery with a non-absolute name (possible if
6725 no home directory exists for the router), the forbid_file option was not
6728 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6729 the existing equality matching.
6731 52. Exim was using ints instead of ino_t variables in some places where it was
6732 dealing with inode numbers.
6734 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6735 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6736 called TMPDIR, and if it finds it is different, it changes its value.
6738 54. The smtp_printf() function is now made available to local_scan() so
6739 additional output lines can be written before returning. There is also an
6740 smtp_fflush() function to enable the detection of a dropped connection.
6741 The variables smtp_input and smtp_batched_input are exported to
6744 55. Changed the default runtime configuration: the message "Unknown user"
6745 has been removed from the ACL, and instead placed on the localuser router,
6746 using the cannot_route_message feature. This means that any verification
6747 failures that generate their own messages won't get overridden. Similarly,
6748 the "Unrouteable address" message that was in the ACL for unverifiable
6749 relay addresses has also been removed.
6751 56. Added hosts_avoid_esmtp to the smtp transport.
6753 57. The exicyclog script was not checking for the esoteric option
6754 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6755 will work only if exicyclog is run under the appropriate euid.
6757 58. Following a discussion on the list, the rules by which Exim recognises line
6758 endings on incoming messages have been changed. The -dropcr and drop_cr
6759 options are now no-ops, retained only for backwards compatibility. The
6760 following line terminators are recognized: LF CRLF CR. However, special
6761 processing applies to CR:
6763 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6764 nor a local message in the state where . is a terminator.
6766 (ii) If a bare CR is encountered in a header line, an extra space is added
6767 after the line terminator so as not to end the header. The reasoning
6768 behind this is that bare CRs in header lines are most likely either
6769 to be mistakes, or people trying to play silly games.
6771 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6772 was being incorrectly given as 18 bytes larger than it should have been.
6773 This is a VOB (very old bug).
6775 60. This may never have affected anything current, but just in case it has:
6776 When the local host is found other than at the start of a list of hosts,
6777 the local host, those with the same MX, and any that follow, are discarded.
6778 When the list in question was part of a longer list of hosts, the following
6779 hosts (not currently being processed) were also being discarded. This no
6780 longer happens. I'm not sure if this situation could ever has previously
6783 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6786 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6787 header. This is something that is recommended in a new Internet Draft, and
6788 is something that is documented as being done by Sendmail. There are two
6789 possible values. For messages generated by the autoreply transport, Exim
6792 Auto-Submitted: auto-replied
6794 whereas for all other generated messages (e.g. bounces) it adds
6796 Auto-Submitted: auto-generated
6798 63. The "personal" condition in filters now includes a test for the
6799 Auto-Submitted: header. If it contains the string "auto-" the message it
6800 not considered personal.
6802 64. Added rcpt_include_affixes as a generic transport option.
6804 65. Added queue_only_override (default true).
6806 66. Added the syslog_duplication option.
6808 67. If what should have been the first header line of a message consisted of
6809 a space followed by a colon, Exim was mis-interpreting it as a header line.
6810 It isn't of course - it is syntactically invalid and should therefore be
6811 treated as the start of the message body. The misbehaviour could have
6812 caused a number of strange effects, including loss of data in subsequent
6813 header lines, and spool format errors.
6815 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6816 client host had authenticated. This control can now be exercised by an ACL
6817 for more flexibility.
6819 69. By default, callouts do not happen when testing with -bh. There is now a
6820 variant, -bhc, which does actually run the callout code, including
6821 consulting and updating the callout cache.
6823 70. Added support for saslauthd authentication, courtesy of Alexander
6826 71. If statvfs() failed on the spool or log directories while checking their
6827 size for availability, Exim confusingly gave the error "space shortage".
6828 Furthermore, in debugging mode it crashed with a floating point exception.
6829 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6830 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6831 serious problem, Exim now writes to the main and panic logs when this
6832 happens, with details of the failure. It then refuses to accept the
6833 incoming message, giving the message "spool directory problem" or "log
6834 directory problem" with a 421 code for SMTP messages.
6836 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6837 0, 1, and 2 exist, because some OS complain for execs without them (see
6838 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6839 descriptors. However, the code omitted to check that the open succeeded,
6840 causing mysterious errors if for some reason the permissions on /dev/null
6841 got screwed. Now Exim writes a message to the main and panic logs, and
6842 bombs out if it can't open /dev/null.
6844 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6845 interact so that it is all more flexible. It is supposed to remain
6846 backwards compatible. Also added extra_local_interfaces.
6848 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6849 to bomb out with an assertion failure - to the client this appears as a
6850 connection drop. This problem occurs in the part of the code that was taken
6851 from the Samba project. Fortunately, the assertion is in a very simple
6852 function, so I have fixed this by reproducing the function inline in the
6853 one place where it is called, and arranging for authentication to fail
6854 instead of killing the process with assert().
6856 75. The SPA client code was not working when the server requested OEM rather
6857 than Unicode encoding.
6859 76. Added code to make require_files with a specific uid setting more usable in
6860 the case where statting the file as root fails - usually a non-root-mounted
6861 NFS file system. When this happens and the failure is EACCES, Exim now
6862 forks a subprocess and does the per-uid checking as the relevant uid.
6864 77. Added process_log_path.
6866 78. If log_file_path was not explicitly set, a setting of check_log_space or
6867 check_log_inodes was ignored.
6869 79. If a space check for the spool or log partitions fails, the incident is now
6870 logged. Of course, in the latter case the data may get lost...
6872 80. Added the %p formatting code to string_format() so that it can be used to
6873 print addresses in debug_print(). Adjusted all the address printing in the
6874 debugging in store.c to use %p rather than %d.
6876 81. There was a concern that a line of code in smtp_in.c could overflow a
6877 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6878 initially expressed, the concern was not well-founded, because trailing
6879 spaces are removed early. However, if the trailing spaces were followed by
6880 a NULL, they did not get removed, so the overflow was possible. Two fixes
6883 (a) I re-wrote the offending code in a cleaner fashion.
6884 (b) If an incoming SMTP command contains a NULL character, it is rejected
6887 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6888 runs initgroups(), so that if the Exim user is in any additional groups,
6889 they will be used during message reception.
6895 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob