1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 self-signed certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the source code
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
102 JH/27 Fix a possible security hole, wherein a process operating with the Exim
103 UID can gain a root shell. Credit to http://www.halfdog.net/ for
104 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
107 JH/28 Enable {spool,log} filesystem space and inode checks as default.
108 Main config options check_{log,spool}_{inodes,space} are now
109 100 inodes, 10MB unless set otherwise in the configuration.
111 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
112 Previously it only applied to the main-section connection policy
115 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
117 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
118 by me. Added RFC7919 DH primes as an alternative.
120 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
121 are not in the system include path.
123 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
124 GnuTLS, when a session startup failed (eg because the client disconnected)
125 Exim did stdio operations after fclose. This was exposed by a recent
126 change which nulled out the file handle after the fclose.
128 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
129 signed directly by the cert-signing cert, rather than an intermediate
130 OCSP-signing cert. This is the model used by LetsEncrypt.
132 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
134 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
135 an incoming connection.
140 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
141 and 3.4.4 - once the server is enabled to respond to an OCSP request
142 it does even when not requested, resulting in a stapling non-aware
143 client dropping the TLS connection.
145 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
146 support variable-length bit vectors. No functional change.
148 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
149 The I= interface field on outgoing lines is now after the H= remote
150 host field, same as incoming lines. There is a separate
151 outgoing_interface log selector which allows you to disable the
154 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
155 If not running log_selector +smtp_connection the mainlog would be held
156 open indefinitely after a "too many connections" event, including to a
157 deleted file after a log rotate. Leave the per net connection logging
158 leaving it open for efficiency as that will be quickly detected by the
159 check on the next write.
161 HS/01 Bug 1671: Fix post transport crash.
162 Processing the wait-<transport> messages could crash the delivery
163 process if the message IDs didn't exist for some reason. When
164 using 'split_spool_directory=yes' the construction of the spool
165 file name failed already, exposing the same netto behaviour.
167 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
168 mime_regex ACL conditions.
170 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
171 to DSN fail messages (bounces): remote IP, remote greeting, remote response
172 to HELO, local diagnostic string.
174 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
175 log line to debug. Even when configured with a tls authenticator many
176 client connections are expected to not authenticate in this way, so
177 an authenticate fail is not an error.
179 HS/02 Add the Exim version string to the process info. This way exiwhat
180 gives some more detail about the running daemon.
182 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
183 matter for fast-change records such as DNSBLs.
185 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
186 retry record, even if constant. There may be multiple transports with
187 different interface settings and the retry behaviour needs to be kept
190 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
192 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
194 JH/10 Bug 840: fix log_defer_output option of pipe transport
196 JH/11 Bug 830: use same host for all RCPTS of a message, even under
197 hosts_randomize. This matters a lot when combined with mua_wrapper.
199 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
200 ${quote_pgsql:<string>} operator.
202 JH/13 Bug 1708: avoid misaligned access in cached lookup.
204 JH/14 Change header file name for freeradius-client. Relevant if compiling
205 with Radius support; from the Gentoo tree and checked under Fedora.
207 JH/15 Bug 1712: Introduce $prdr_requested flag variable
209 JH/16 Bug 1714: Permit an empty string as expansion result for transport
210 option transport_filter, meaning no filtering.
212 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
214 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
215 defaults to "*" (all hosts). The variable is now available when not built
216 with TLS, default unset, mainly to enable keeping the testuite sane.
217 If a server certificate is not supplied (via tls_certificate) an error is
218 logged, and clients will find TLS connections fail on startup. Presumably
219 they will retry in-clear.
220 Packagers of Exim are strongly encouraged to create a server certificate
221 at installation time.
223 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
224 with the $config_file variable.
226 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
227 in transport context, after the attempt, and per-recipient. The latter type
228 is per host attempted. The event data is the error message, and the errno
229 information encodes the lookup type (A vs. MX) used for the (first) host,
230 and the trailing two digits of the smtp 4xx reponse.
232 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
233 to write to mainlog (or rejectlog, paniclog) in the window between file
234 creation and permissions/ownership being changed. Particularly affects
235 installations where exicyclog is run as root, rather than exim user;
236 result is that the running daemon panics and dies.
238 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
240 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
241 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
242 "pri" and "weight". Note that the previous implicit priority given by the
243 list order is no longer honoured.
245 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
248 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
249 by defining SUPPORT_SOCKS.
251 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
252 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
253 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
254 variables are renamed to proxy_{local,external}_{address,port}.
256 JH/27 Move Internationalisation support from Experimental to mainline, enabled
257 for a build by defining SUPPORT_I18N
259 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
260 of the query string, and make ${quote_redis:} do that quoting.
262 JH/29 Move Events support from Experimental to mainline, enabled by default
263 and removable for a build by defining DISABLE_EVENT.
265 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
267 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
268 cached by the daemon.
270 JH/32 Move Redis support from Experimental to mainline, enabled for a build
271 by defining LOOKUP_REDIS. The libhiredis library is required.
273 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
274 keys are given for lookup.
276 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
277 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
278 only supported when built with TLS support. The PolarSSL SHA routines
279 are still used when the TLS library is too old for convenient support.
281 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
282 openssl_options), for security. OpenSSL forces this from version 1.1.0
283 server-side so match that on older versions.
285 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
286 allocation for $value could be released as the expansion processing
287 concluded, but leaving the global pointer active for it.
289 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
290 and to use the domains and local_parts ACL conditions.
292 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
293 incorrectly not doubled on cutthrough transmission, hence seen as a
294 body-termination at the receiving system - resulting in truncated mails.
295 Commonly the sender saw a TCP-level error, and retransmitted the message
296 via the normal store-and-forward channel. This could result in duplicates
297 received - but deduplicating mailstores were liable to retain only the
298 initial truncated version.
300 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
302 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
304 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
305 we're in there, support oversigning also; bug 1309.
307 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
309 HS/04 Add support for keep_environment and add_environment options.
311 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
312 either intentional arithmetic overflow during PRNG, or testing config-
315 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
316 delivery resulted in actual delivery. Cancel cutthrough before DATA
319 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
320 rejects a recipient: pass the reject to the originator.
322 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
323 Many were false-positives and ignorable, but it's worth fixing the
326 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
327 for the new environment-manipulation done at startup. Move the routines
328 from being local to tls.c to being global via the os.c file.
330 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
331 an extract embedded as result-arg for a map, the first arg for extract
332 is unavailable so we cannot tell if this is a numbered or keyed
333 extraction. Accept either.
339 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
342 JH/02 The smtp transport option "multi_domain" is now expanded.
344 JH/03 The smtp transport now requests PRDR by default, if the server offers
347 JH/04 Certificate name checking on server certificates, when exim is a client,
348 is now done by default. The transport option tls_verify_cert_hostnames
349 can be used to disable this per-host. The build option
350 EXPERIMENTAL_CERTNAMES is withdrawn.
352 JH/05 The value of the tls_verify_certificates smtp transport and main options
353 default to the word "system" to access the system default CA bundle.
354 For GnuTLS, only version 3.0.20 or later.
356 JH/06 Verification of the server certificate for a TLS connection is now tried
357 (but not required) by default. The verification status is now logged by
358 default, for both outbound TLS and client-certificate supplying inbound
361 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
364 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
365 Status Notification (bounce) messages are now MIME format per RFC 3464.
366 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
367 under the control of the dsn_advertise_hosts option, and routers may
368 have a dsn_lasthop option.
370 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
371 default, modifiable by a malware= option. The list separator for
372 the options can now be changed in the usual way. Bug 68.
374 JH/10 The smtp_receive_timeout main option is now expanded before use.
376 JH/11 The incoming_interface log option now also enables logging of the
377 local interface on delivery outgoing connections.
379 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
380 if the interface and destination host and port all match.
382 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
385 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
386 Patch from Andrew Lewis.
388 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
389 now supports optional time-restrictions, weighting, and priority
390 modifiers per server. Patch originally by <rommer@active.by>.
392 JH/16 The spamd_address main option now supports a mixed list of local
393 and remote servers. Remote servers can be IPv6 addresses, and
394 specify a port-range.
396 JH/17 Bug 68: The spamd_address main option now supports an optional
397 timeout value per server.
399 JH/18 Bug 1581: Router and transport options headers_add/remove can
400 now have the list separator specified.
402 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
405 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
408 JH/21 Support for the A6 type of dns record is withdrawn.
410 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
411 rather than the verbs used.
413 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
414 from 255 to 1024 chars.
416 JH/24 Verification callouts now attempt to use TLS by default.
418 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
419 are generic router options now. The defaults didn't change.
421 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
422 Original patch from Alexander Shikoff, worked over by JH.
424 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
425 files if ZCAT_COMMAND is not executable.
427 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
429 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
431 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
432 Normally benign, it bites when the pair was led to by a CNAME;
433 modern usage is to not canoicalize the domain to a CNAME target
434 (and we were inconsistent anyway for A-only vs AAAA+A).
436 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
438 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
439 when evaluating $sender_host_dnssec.
441 JH/31 Check the HELO verification lookup for DNSSEC, adding new
442 $sender_helo_dnssec variable.
444 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
446 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
448 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
450 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
451 documented as working, but never had. Support all but $spam_report.
453 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
454 added for tls authenticator.
456 HS/03 Add perl_taintmode main config option
461 TL/01 When running the test suite, the README says that variables such as
462 no_msglog_check are global and can be placed anywhere in a specific
463 test's script, however it was observed that placement needed to be near
464 the beginning for it to behave that way. Changed the runtest perl
465 script to read through the entire script once to detect and set these
466 variables, reset to the beginning of the script, and then run through
467 the script parsing/test process like normal.
469 TL/02 The BSD's have an arc4random API. One of the functions to induce
470 adding randomness was arc4random_stir(), but it has been removed in
471 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
472 function when detected.
474 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
475 cause callback expansion.
477 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
478 syntax errors in an expansion can be treated as a string instead of
479 logging or causing an error, due to the internal use of bool_lax
480 instead of bool when processing it.
482 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
483 server certificates when making smtp deliveries.
485 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
487 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
489 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
491 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
492 Merged patch from Sebastian Wiedenroth.
494 JH/05 Fix results-pipe from transport process. Several recipients, combined
495 with certificate use, exposed issues where response data items split
496 over buffer boundaries were not parsed properly. This eventually
497 resulted in duplicates being sent. This issue only became common enough
498 to notice due to the introduction of conection certificate information,
499 the item size being so much larger. Found and fixed by Wolfgang Breyha.
501 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
502 size buffer was used, resulting in syntax errors when an expansion
505 JH/07 Add support for directories of certificates when compiled with a GnuTLS
506 version 3.3.6 or later.
508 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
509 is EXPERIMENTAL_EVENT, the main-configuration and transport options
510 both become "event_action", the variables become $event_name, $event_data
511 and $event_defer_errno. There is a new variable $verify_mode, usable in
512 routers, transports and related events. The tls:cert event is now also
513 raised for inbound connections, if the main configuration event_action
516 TL/06 In test suite, disable OCSP for old versions of openssl which contained
517 early OCSP support, but no stapling (appears to be less than 1.0.0).
519 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
520 server certificate names available under the smtp transport option
521 "tls_verify_cert_hostname" now do not permit multi-component wildcard
524 JH/10 Time-related extraction expansions from certificates now use the main
525 option "timezone" setting for output formatting, and are consistent
526 between OpenSSL and GnuTLS compilations. Bug 1541.
528 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
529 encoded parameter in the incoming message. Bug 1558.
531 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
532 include certificate info, eximon was claiming there were spoolfile
535 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
537 JH/14 Log delivery-related information more consistently, using the sequence
538 "H=<name> [<ip>]" wherever possible.
540 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
541 are problematic for Debian distribution, omit them from the release
544 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
546 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
549 JH/17 Fix a null-indirection in certextract expansions when a nondefault
550 output list separator was used.
555 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
556 checkers that were complaining about end of non-void function with no
559 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
560 This was a regression intruduced in 4.83 by another bugfix.
562 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
564 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
565 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
571 TF/01 Correctly close the server side of TLS when forking for delivery.
573 When a message was received over SMTP with TLS, Exim failed to clear up
574 the incoming connection properly after forking off the child process to
575 deliver the message. In some situations the subsequent outgoing
576 delivery connection happened to have the same fd number as the incoming
577 connection previously had. Exim would try to use TLS and fail, logging
578 a "Bad file descriptor" error.
580 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
581 utilities have not been installed.
583 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
584 temporary space as the ACL may create new global variables.
586 TL/01 LDAP support uses per connection or global context settings, depending
587 upon the detected version of the libraries at build time.
589 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
590 to extract and use the src ip:port in logging and expansions as if it
591 were a direct connection from the outside internet. PPv2 support was
592 updated based on HAProxy spec change in May 2014.
594 JH/02 Add ${listextract {number}{list}{success}{fail}}.
596 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
597 Properly escape header and check for NULL return.
599 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
602 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
604 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
605 characters in header names, implemented as a verify condition.
606 Contributed by Michael Fischer v. Mollard.
608 TL/05 Rename SPF condition results err_perm and err_temp to standardized
609 results permerror and temperror. Previous values are deprecated but
610 still accepted. In a future release, err_perm and err_temp will be
611 completely removed, which will be a backward incompatibility if the
612 ACL tests for either of these two old results. Patch contributed by
613 user bes-internal on the mailing list.
615 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
617 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
618 selectors, in both main and reject logs.
620 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
623 JH/07 Add malware type "sock" for talking to simple daemon.
625 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
627 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
628 routers/transports under cutthrough routing.
630 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
631 numbers. Touch up "bool" conditional to keep the same definition.
633 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
635 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
637 JH/12 Expand items in router/transport headers_add or headers_remove lists
638 individually rather than the list as a whole. Bug 1452.
640 Required for reasonable handling of multiple headers_ options when
641 they may be empty; requires that headers_remove items with embedded
642 colons must have them doubled (or the list-separator changed).
644 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
645 view the policy declared in the DMARC record. Currently, $dmarc_status
646 is a combined value of both the record presence and the result of the
649 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
651 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
652 dnslookup router and the smtp transport (applying to the forward
655 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
656 of ldap servers used for a specific lookup. Patch provided by Heiko
659 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
660 New variable $lookup_dnssec_authenticated for observability.
662 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
663 Patch submitted by Lars Timman.
665 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
667 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
668 Requires trusted mode and valid format message id, aborts otherwise.
669 Patch contributed by Heiko Schlichting.
671 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
672 certextract with support for various fields. Bug 1358.
674 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
675 is requested by default, modifiable by smtp transport option
678 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
679 operate on certificate variables to give certificate fingerprints
680 Also new ${sha256:cert_variable}.
682 JH/23 The PRDR feature is moved from being Experimental into the mainline.
684 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
685 Christian Aistleitner.
687 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
689 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
690 file. Patch from Wolfgang Breyha.
692 JH/25 Expand the coverage of the delivery $host and $host_address to
693 client authenticators run in verify callout. Bug 1476.
695 JH/26 Port service names are now accepted for tls_on_connect_ports, to
696 align with daemon_smtp_ports. Bug 72.
698 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
699 support and error reporting did not work properly.
701 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
702 and is readable. Patch from Andrew Colin Kissa.
704 TL/14 Enhance documentation of ${run expansion and how it parses the
705 commandline after expansion, particularly in the case when an
706 unquoted variable expansion results in an empty value.
708 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
710 PP/02 Fix internal collision of T_APL on systems which support RFC3123
711 by renaming away from it. Addresses GH issue 15, reported by
714 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
716 TL/15 SECURITY: prevent double expansion in math comparison functions
717 (can expand unsanitized data). Not remotely exploitable.
724 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
726 PP/02 Make -n do something, by making it not do something.
727 When combined with -bP, the name of an option is not output.
729 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
732 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
733 $sender_host_name and config options to manage this, and basic check
736 PP/05 DSCP support for outbound connections and control modifier for inbound.
738 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
739 (Only plugin which currently uses this is kerberos4, which nobody should
740 be using, but we should make it available and other future plugins might
741 conceivably use it, even though it would break NAT; stuff *should* be
742 using channel bindings instead).
744 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
745 name; added for Sendmail compatibility; requires admin caller.
746 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
747 just ignore it); requires trusted caller.
748 Also parse but ignore: -Ac -Am -X<logfile>
751 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
753 TL/02 Add +smtp_confirmation as a default logging option.
755 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
756 Patch by Magnus Holmgren from 2007-02-20.
758 TL/04 Bugzilla 1281 - Spec typo.
759 Bugzilla 1283 - Spec typo.
760 Bugzilla 1290 - Spec grammar fixes.
762 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
764 TL/06 Add Experimental DMARC support using libopendmarc libraries.
766 TL/07 Fix an out of order global option causing a segfault. Reported to dev
767 mailing list by by Dmitry Isaikin.
769 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
771 JH/02 Support "G" suffix to numbers in ${if comparisons.
773 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
775 NM/01 Bugzilla 1197 - Spec typo
776 Bugzilla 1196 - Spec examples corrections
778 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
780 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
781 gnutls_enable_pkcs11, but renamed to more accurately indicate its
784 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
785 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
787 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
788 "acl {{name}{arg}...}", and optional args on acl condition
791 JH/05 Permit multiple router/transport headers_add/remove lines.
793 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
795 JH/07 Avoid using a waiting database for a single-message-only transport.
796 Performance patch from Paul Fisher. Bugzilla 1262.
798 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
801 JH/09 Add $headers_added variable, with content from use of ACL modifier
802 add_header (but not yet added to the message). Bugzilla 199.
804 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
805 Pulled from Bugzilla 817 by Wolfgang Breyha.
807 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
809 (nb: this is the same fix as in Exim 4.80.1)
811 JH/11 Add A= logging on delivery lines, and a client_set_id option on
814 JH/12 Add optional authenticated_sender logging to A= and a log_selector
817 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
819 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
820 advertise SMTP AUTH mechanism to us, instead of a generic
821 protocol violation error. Also, make Exim more robust to bad
822 data from the Dovecot auth socket.
824 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
826 When a queue runner is handling a message, Exim first routes the
827 recipient addresses, during which it prunes them based on the retry
828 hints database. After that it attempts to deliver the message to
829 any remaining recipients. It then updates the hints database using
832 So if a recipient address works intermittently, it can get repeatedly
833 deferred at routing time. The retry hints record remains fresh so the
834 address never reaches the final cutoff time.
836 This is a fairly common occurrence when a user is bumping up against
837 their storage quota. Exim had some logic in its local delivery code
838 to deal with this. However it did not apply to per-recipient defers
839 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
841 This change adds a proper retry rule check during routing so that the
842 final cutoff time is checked against the message's age. We only do
843 this check if there is an address retry record and there is not a
844 domain retry record; this implies that previous attempts to handle
845 the address had the retry_use_local_parts option turned on. We use
846 this as an approximation for the destination being like a local
847 delivery, as in LMTP.
849 I suspect this new check makes the old local delivery cutoff check
850 redundant, but I have not verified this so I left the code in place.
852 TF/02 Correct gecos expansion when From: is a prefix of the username.
854 Test 0254 submits a message to Exim with the header
858 When I ran the test suite under the user fanf2, Exim expanded
859 the header to contain my full name, whereas it should have added
860 a Resent-Sender: header. It erroneously treats any prefix of the
861 username as equal to the username.
863 This change corrects that bug.
865 GF/01 DCC debug and logging tidyup
866 Error conditions log to paniclog rather than rejectlog.
867 Debug lines prefixed by "DCC: " to remove any ambiguity.
869 TF/03 Avoid unnecessary rebuilds of lookup-related code.
871 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
872 Bug spotted by Jeremy Harris; was flawed since initial commit.
873 Would have resulted in OCSP responses post-SNI triggering an Exim
874 NULL dereference and crash.
876 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
878 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
879 Bug detection, analysis and fix by Samuel Thibault.
880 Bugzilla 1331, Debian bug #698092.
882 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
884 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
885 Server implementation by Todd Lyons, client by JH.
886 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
887 config variable "prdr_enable" controls whether the server
888 advertises the facility. If the client requests PRDR a new
889 acl_data_smtp_prdr ACL is called once for each recipient, after
890 the body content is received and before the acl_smtp_data ACL.
891 The client is controlled by bolth of: a hosts_try_prdr option
892 on the smtp transport, and the server advertisement.
893 Default client logging of deliveries and rejections involving
894 PRDR are flagged with the string "PRDR".
896 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
897 fclose(). Diagnosis by Todd Lyons.
899 PP/17 Update configure.default to handle IPv6 localhost better.
900 Patch by Alain Williams (plus minor tweaks).
903 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
904 This is now consistent with GnuTLS, and is now documented: the
905 previous undocumented portable approach to treating the option as
906 unset was to force an expansion failure. That still works, and
907 an empty string is now equivalent.
909 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
910 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
911 not performing validation itself.
913 PP/20 Added force_command boolean option to pipe transport.
914 Patch from Nick Koston, of cPanel Inc.
916 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
919 TF/04 Added udpsend ACL modifer and hexquote expansion operator
921 PP/21 Fix eximon continuous updating with timestamped log-files.
922 Broken in a format-string cleanup in 4.80, missed when I repaired the
923 other false fix of the same issue.
924 Report and fix from Heiko Schlichting.
927 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
928 Report from Prashanth Katuri.
930 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
931 It's SecureTransport, so affects any MacOS clients which use the
932 system-integrated TLS libraries, including email clients.
934 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
935 using a MIME ACL for non-SMTP local injection.
936 Report and assistance in diagnosis by Warren Baker.
938 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
940 JH/16 Fix comparisons for 64b. Bugzilla 1385.
942 TL/09 Add expansion variable $authenticated_fail_id to keep track of
943 last id that failed so it may be referenced in subsequent ACL's.
945 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
948 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
949 ldap library initialization, allowing self-signed CA's to be
950 used. Also properly sets require_cert option later in code by
951 using NULL (global ldap config) instead of ldap handle (per
952 session). Bug diagnosis and testing by alxgomz.
954 TL/12 Enhanced documentation in the ratelimit.pl script provided in
955 the src/util/ subdirectory.
957 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
958 renamed to Transport Post Delivery Action by Jeremy Harris, as
961 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
962 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
963 redis_servers = needs to be configured which will be used by the redis
964 lookup. Patch from Warren Baker, of The Packet Hub.
966 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
968 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
969 hostname or reverse DNS when processing a host list. Used suggestions
970 from multiple comments on this bug.
972 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
974 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
975 Missed a few lines, added it to make the runtest require no keyboard
978 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
979 contains upper case chars. Make router use caseful_local_part.
981 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
982 support when GnuTLS has been built with p11-kit.
988 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
990 This, or similar/improved, will also be change PP/11 of 4.82.
996 PP/01 Handle short writes when writing local log-files.
997 In practice, only affects FreeBSD (8 onwards).
998 Bugzilla 1053, with thanks to Dmitry Isaikin.
1000 NM/01 Bugzilla 949 - Documentation tweak
1002 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1005 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1007 PP/02 Implemented gsasl authenticator.
1009 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1011 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1012 `pkg-config foo` for cflags/libs.
1014 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1015 with rest of GSASL and with heimdal_gssapi.
1017 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1018 `pkg-config foo` for cflags/libs for the TLS implementation.
1020 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1021 properties get this fed in as external SSF. A number of robustness
1022 and debugging improvements to the cyrus_sasl authenticator.
1024 PP/08 cyrus_sasl server now expands the server_realm option.
1026 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1027 Patch by Jeremy Harris.
1029 PP/10 Added dbmjz lookup type.
1031 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1033 PP/12 MAIL args handles TAB as well as SP, for better interop with
1034 non-compliant senders.
1035 Analysis and variant patch by Todd Lyons.
1037 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1038 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1039 Patch from Dirk Mueller <dmueller@suse.com>
1041 PP/13 tls_peerdn now print-escaped for spool files.
1042 Observed some $tls_peerdn in wild which contained \n, which resulted
1043 in spool file corruption.
1045 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1046 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1047 or write after TLS renegotiation, which otherwise led to messages
1050 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1051 as a tracking header (ie: a signed header comes before the signature).
1052 Patch from Wolfgang Breyha.
1054 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1055 comma-sep list; embedded commas doubled.
1057 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1059 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1061 Report and patch from Dmitry Banschikov.
1063 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1064 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1065 failures. We appear to now support TLS1.1+ with Exim.
1067 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1068 lets Exim select keys and certificates based upon TLS SNI from client.
1069 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1070 before an outbound SMTP session. New log_selector, +tls_sni.
1072 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1073 NULL dereference. Report and patch from Alun Jones.
1075 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1076 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1077 Not seeing resolver debug output on NetBSD, but suspect this is a
1078 resolver implementation change.
1080 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1081 Left warnings. Added "eximon gdb" invocation mode.
1083 PP/21 Defaulting "accept_8bitmime" to true, not false.
1085 PP/22 Added -bw for inetd wait mode support.
1087 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1088 locate the relevant includes and libraries. Made this the default.
1090 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1091 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1093 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1094 This may cause build issues on older platforms.
1096 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1097 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1098 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1099 Added SNI support via GnuTLS too.
1100 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1102 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1104 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1105 Applied second patch from Janne, implementing suggestion to default
1106 multiple-strings-in-record handling to match SPF spec.
1108 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1110 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1111 read-only, out of scope).
1112 Patch from Wolfgang Breyha, report from Stuart Northfield.
1114 PP/29 Fix three issues highlighted by clang analyser static analysis.
1115 Only crash-plausible issue would require the Cambridge-specific
1116 iplookup router and a misconfiguration.
1117 Report from Marcin Mirosław.
1119 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1121 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1122 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1123 As part of this, removing so much warning spew let me fix some minor
1124 real issues in debug logging.
1126 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1127 assignment on my part. Fixed.
1129 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1130 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1131 Janne Snabb (who went above and beyond: thank you).
1133 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1134 string otherwise requires a connection and a bunch more work and it's
1135 relatively easy to get wrong. Should also expose TLS library linkage
1138 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1139 64-bit ${eval} (JH/03).
1141 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1142 GNU libc to support some of the 64-bit stuff, should not lead to
1143 conflicts. Defined before os.h is pulled in, so if a given platform
1144 needs to override this, it can.
1146 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1147 protection layer was required, which is not implemented.
1148 Bugzilla 1254, patch from Wolfgang Breyha.
1150 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1151 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1152 tls_dhparam take prime identifiers. Also unbreak combination of
1153 OpenSSL+DH_params+TLSSNI.
1155 PP/39 Disable SSLv2 by default in OpenSSL support.
1161 PP/01 Solaris build fix for Oracle's LDAP libraries.
1162 Bugzilla 1109, patch from Stephen Usher.
1164 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1166 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1169 TF/02 Fix a couple more cases where we did not log the error message
1170 when unlink() failed. See also change 4.74-TF/03.
1172 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1173 lock up or crash if it happened to be inside a call to libc when it
1174 got a SIGUSR1 from exiwhat.
1176 The SIGUSR1 handler appends the current process status to the process
1177 log which is later printed by exiwhat. It used to use the general
1178 purpose logging code to do this, but several functions it calls are
1179 not safe for signals.
1181 The new output code in the SIGUSR1 handler is specific to the process
1182 log, and simple enough that it's easy to inspect for signal safety.
1183 Removing some special cases also simplifies the general logging code.
1184 Removing the spurious timestamps from the process log simplifies
1187 TF/04 Improved ratelimit ACL condition.
1189 The /noupdate option has been deprecated in favour of /readonly which
1190 has clearer semantics. The /leaky, /strict, and /readonly update modes
1191 are mutually exclusive. The update mode is no longer included in the
1192 database key; it just determines when the database is updated. (This
1193 means that when you upgrde Exim will forget old rate measurements.)
1195 Exim now checks that the per_* options are used with an update mode that
1196 makes sense for the current ACL. For example, when Exim is processing a
1197 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1198 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1199 must specify per_mail/readonly. If you omit the update mode it defaults to
1200 /leaky where that makes sense (as before) or /readonly where required.
1202 The /noupdate option is now undocumented but still supported for
1203 backwards compatibility. It is equivalent to /readonly except that in
1204 ACLs where /readonly is required you may specify /leaky/noupdate or
1205 /strict/noupdate which are treated the same as /readonly.
1207 A useful new feature is the /count= option. This is a generalization
1208 of the per_byte option, so that you can measure the throughput of other
1209 aggregate values. For example, the per_byte option is now equivalent
1210 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1212 The per_rcpt option has been generalized using the /count= mechanism
1213 (though it's more complicated than the per_byte equivalence). When it is
1214 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1215 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1216 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1217 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1218 Note that using per_rcpt with a non-readonly update mode in more than
1219 one ACL will cause the recipients to be double-counted. (The per_mail
1220 and per_byte options don't have this problem.)
1222 The handling of very low rates has changed slightly. If the computed rate
1223 is less than the event's count (usually one) then this event is the first
1224 after a long gap. In this case the rate is set to the same as this event's
1225 count, so that the first message of a spam run is counted properly.
1227 The major new feature is a mechanism for counting the rate of unique
1228 events. The new per_addr option counts the number of different
1229 recipients that someone has sent messages to in the last time period. It
1230 behaves like per_rcpt if all the recipient addresses are different, but
1231 duplicate recipient addresses do not increase the measured rate. Like
1232 the /count= option this is a general mechanism, so the per_addr option
1233 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1234 example, measure the rate that a client uses different sender addresses
1235 with the options per_mail/unique=$sender_address. There are further
1236 details in the main documentation.
1238 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1240 TF/06 Removed a few PCRE remnants.
1242 TF/07 Automatically extract Exim's version number from tags in the git
1243 repository when doing development or release builds.
1245 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1246 Bugzilla 879. Patch from Paul Fisher.
1248 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1249 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1252 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1254 PP/05 Variable $av_failed, true if the AV scanner deferred.
1255 Bugzilla 1078. Patch from John Horne.
1257 PP/06 Stop make process more reliably on build failure.
1258 Bugzilla 1087. Patch from Heiko Schlittermann.
1260 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1261 Bugzilla 1089. Patch from Heiko Schlittermann.
1263 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1264 Bugzilla 1131. Patch from Holger Weiß.
1266 PP/09 Handle IPv6 addresses with SPF.
1267 Bugzilla 860. Patch from Wolfgang Breyha.
1269 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1271 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1274 PP/11 match_* no longer expand right-hand-side by default.
1275 New compile-time build option, EXPAND_LISTMATCH_RHS.
1276 New expansion conditions, "inlist", "inlisti".
1278 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1280 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1282 PP/14 fix log_write() format string regression from TF/03.
1283 Bugzilla 1152. Patch from Dmitry Isaikin.
1289 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1291 PP/02 Harmonised TLS library version reporting; only show if debugging.
1292 Layout now matches that introduced for other libraries in 4.74 PP/03.
1294 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1296 PP/04 New "dns_use_edns0" global option.
1298 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1301 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1302 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1304 TK/01 Updated PolarSSL code to 0.14.2.
1305 Bugzilla 1097. Patch from Andreas Metzler.
1307 PP/07 Catch divide-by-zero in ${eval:...}.
1308 Fixes bugzilla 1102.
1310 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1313 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1314 format-string attack -- SECURITY: remote arbitrary code execution.
1316 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1317 time unintentionally subject to list matching rules, letting the header
1318 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1319 arbitrary string expansion). This allowed for information disclosure.
1321 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1322 INT_MIN/-1 -- value coerced to INT_MAX.
1328 NM/01 Workround for PCRE version dependency in version reporting
1331 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1332 This fixes portability to compilers other than gcc, notably
1333 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1335 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1336 makefiles for portability to HP-UX and POSIX correctness.
1338 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1339 Also via indented variable definition in the Makefile.
1340 (Debugging by Oliver Heesakkers).
1342 PP/02 Restore caching of spamd results with expanded spamd_address.
1343 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1345 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1346 Improves build reliability. Fix from: Frank Elsner
1348 NM/02 Fix wide character breakage in the rfc2047 coding
1349 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1351 NM/03 Allow underscore in dnslist lookups
1352 Fixes bug 1026. Patch from Graeme Fowler
1354 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1355 Code patches from Adam Ciarcinski of NetBSD.
1357 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1360 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1361 is logged, to avoid truncation. Patch from John Horne.
1363 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1364 Patch from Jakob Hirsch.
1366 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1367 SQL string expansion failure details.
1368 Patch from Andrey Oktyabrski.
1370 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1371 Patch from Simon Arlott.
1373 PP/09 New lookups functionality failed to compile on old gcc which rejects
1374 extern declarations in function scope.
1375 Patch from Oliver Fleischmann
1377 PP/10 Use sig_atomic_t for flags set from signal handlers.
1378 Check getgroups() return and improve debugging.
1379 Fixed developed for diagnosis in bug 927 (which turned out to be
1382 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1383 Patch from Mark Zealey.
1385 PP/12 Bugzilla 1056: Improved spamd server selection.
1386 Patch from Mark Zealey.
1388 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1389 Based on patch from Heiko Schlittermann.
1391 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1392 Patch from Uwe Doering, sign-off by Michael Haardt.
1394 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1395 variable declaration deep within a block. Bug and patch from
1398 PP/15 lookups-Makefile IRIX compatibilty coercion.
1400 PP/16 Make DISABLE_DKIM build knob functional.
1402 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1403 Patch by Simon Arlott
1405 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1406 variable argument macros. Our copy now differs from upstream.
1412 TF/01 Failure to get a lock on a hints database can have serious
1413 consequences so log it to the panic log.
1415 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1416 controlled using the smtp_confirmation log selector.
1418 TF/03 Include the error message when we fail to unlink a spool file.
1420 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1421 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1422 for maintaining out-of-tree patches for some time.
1424 PP/01 Bugzilla 139: Documentation and portability issues.
1425 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1426 Handle per-OS dynamic-module compilation flags.
1428 PP/02 Let /dev/null have normal permissions.
1429 The 4.73 fixes were a little too stringent and complained about the
1430 permissions on /dev/null. Exempt it from some checks.
1431 Reported by Andreas M. Kirchwitz.
1433 PP/03 Report version information for many libraries, including
1434 Exim version information for dynamically loaded libraries. Created
1435 version.h, now support a version extension string for distributors
1436 who patch heavily. Dynamic module ABI change.
1438 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1439 privilege escalation vulnerability whereby the Exim run-time user
1440 can cause root to append content of the attacker's choosing to
1443 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1446 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1447 If dropping privileges for untrusted macros, we disabled normal logging
1448 on the basis that it would fail; for the Exim run-time user, this is not
1449 the case, and it resulted in successful deliveries going unlogged.
1450 Fixed. Reported by Andreas Metzler.
1456 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1457 only prepend for the Resent-* case. Fixes regression introduced in
1458 Exim 4.70 by NM/22 for Bugzilla 607.
1460 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1461 increasing numbers of administrators be bitten by this.
1463 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1465 PP/03 Bugzilla 994: added openssl_options main configuration option.
1467 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1469 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1471 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1473 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1474 without a peer certificate, leading to a segfault because of an
1475 assumption that peers always have certificates. Be a little more
1476 paranoid. Problem reported by Martin Tscholak.
1478 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1479 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1480 NB: ClamAV planning to remove STREAM in "middle of 2010".
1481 CL also introduces -bmalware, various -d+acl logging additions and
1482 more caution in buffer sizes.
1484 PP/09 Implemented reverse_ip expansion operator.
1486 PP/10 Bugzilla 937: provide a "debug" ACL control.
1488 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1490 PP/12 Bugzilla 973: Implement --version.
1492 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1494 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1496 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1498 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1499 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1500 ignore trailing whitespace.
1502 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1504 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1507 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1508 Notification from Dr Andrew Aitchison.
1510 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1511 ExtendedDetectionInfo response format.
1512 Notification from John Horne.
1514 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1517 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1518 XSL and documented dependency on system catalogs, with examples of how
1521 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1524 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1525 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1526 configuration file which is writeable by the Exim user or group.
1528 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1529 of configuration files to cover files specified with the -C option if
1530 they are going to be used with root privileges, not just the default
1533 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1534 option (effectively making it always true).
1536 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1537 files to be used while preserving root privileges.
1539 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1540 that rogue child processes cannot use them.
1542 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1543 run-time user, instead of root.
1545 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1546 Exim run-time user without dropping privileges.
1548 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1549 result string, instead of calling string_vformat() twice with the same
1552 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1553 for other users. Others should always drop root privileges if they use
1554 -C on the command line, even for a whitelisted configure file.
1556 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1558 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1564 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1565 $data_path, and $header_path variables; fixed documentation bugs and
1568 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1569 exipick to access non-standard spools, including the "frozen" queue
1572 NM/01 Bugzilla 965: Support mysql stored procedures.
1573 Patch from Alain Williams
1575 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1577 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1578 Patch from Andreas Metzler
1580 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1581 Patch from Kirill Miazine
1583 NM/05 Bugzilla 671: Added umask to procmail example.
1585 JJ/03 installed exipick 20100323.0, fixing doc bug
1587 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1588 directory. Notification and patch from Dan Rosenberg.
1590 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1592 TK/02 Improve log output when DKIM signing operation fails.
1594 MH/01 Treat the transport option dkim_domain as a colon separated
1595 list, not as a single string, and sign the message with each element,
1596 omitting multiple occurences of the same signer.
1598 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1599 Bugzilla 985, 986. Patch by Simon Arlott
1601 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1602 Patch by Simon Arlott
1604 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1605 MBX locking. Notification from Dan Rosenberg.
1611 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1613 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1615 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1617 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1619 NM/04 Bugzilla 915: Flag broken perl installation during build.
1625 TK/01 Added patch by Johannes Berg that expands the main option
1626 "spamd_address" if it starts with a dollar sign.
1628 TK/02 Write list of recipients to X-Envelope-Sender header when building
1629 the mbox-format spool file for content scanning (suggested by Jakob
1632 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1633 (http://www.dcc-servers.net/) support via dccifd. Activated by
1634 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1636 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1637 by Mark Daniel Reidel <mr@df.eu>.
1639 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1640 When building exim an external PCRE library is now needed -
1641 PCRE is a system library on the majority of modern systems.
1642 See entry on PCRE_LIBS in EDITME file.
1644 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1645 conversation. Added nologin parameter to request.
1646 Patch contributed by Kirill Miazine.
1648 TF/01 Do not log submission mode rewrites if they do not change the address.
1650 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1652 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1653 log files in place. Contributed by Roberto Lima.
1655 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1657 TF/03 Bugzilla 615: When checking the local_parts router precondition
1658 after a local_part_suffix or local_part_prefix option, Exim now
1659 does not use the address's named list lookup cache, since this
1660 contains cached lookups for the whole local part.
1662 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1663 Robert Millan. Documentation is in experimental-spec.txt.
1665 TF/04 Bugzilla 668: Fix parallel build (make -j).
1667 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1669 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1670 Patch provided by Jan Srzednicki.
1672 TF/05 Leading white space used to be stripped from $spam_report which
1673 wrecked the formatting. Now it is preserved.
1675 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1676 that they are available at delivery time.
1678 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1680 TF/08 TLS error reporting now respects the incoming_interface and
1681 incoming_port log selectors.
1683 TF/09 Produce a more useful error message if an SMTP transport's hosts
1684 setting expands to an empty string.
1686 NM/06 Bugzilla 744: EXPN did not work under TLS.
1687 Patch provided by Phil Pennock.
1689 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1690 Patch provided by Richard Godbee.
1692 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1693 acl_smtp_notquit, added index entry.
1695 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1696 Patch provided by Eugene Bujak.
1698 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1699 accept(). Patch provided by Maxim Dounin.
1701 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1702 Patch provided by Phil Pennock.
1704 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1706 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1707 Patch provided by Brad "anomie" Jorsch.
1709 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1710 Patch provided by Dean Brooks.
1712 TK/05 Add native DKIM support (does not depend on external libraries).
1714 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1715 Patch provided by Graeme Fowler.
1717 NM/16 Bugzilla 851: Documentation example syntax fix.
1719 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1721 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1724 NM/19 Bugzilla 745: TLS version reporting.
1725 Patch provided by Phil Pennock.
1727 NM/20 Bugzilla 167: bool: condition support.
1728 Patch provided by Phil Pennock.
1730 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1731 clients. Patch provided by Phil Pennock.
1733 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1734 Patch provided by Brad "anomie" Jorsch.
1736 NM/23 Bugzilla 687: Fix misparses in eximstats.
1737 Patch provided by Heiko Schlittermann.
1739 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1740 Patch provided by Heiko Schlittermann.
1742 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1743 plus update to original patch.
1745 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1747 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1748 Patch provided by David Brownlee.
1750 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1752 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1754 NM/30 Bugzilla 888: TLS documentation bugfixes.
1756 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1758 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1759 Unlike the original bugzilla I have changed all shell scripts in src tree.
1761 NM/33 Bugzilla 898: Transport filter timeout fix.
1762 Patch by Todd Rinaldo.
1764 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1765 Patch by Serge Demonchaux.
1767 NM/35 Bugzilla 39: Base64 decode bug fixes.
1768 Patch by Jakob Hirsch.
1770 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1772 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1774 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1776 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1782 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1783 ALT-N's libdkim that I have put here:
1784 http://duncanthrax.net/exim-experimental/
1786 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1787 were called 'true' and it seems that C99 defines that as a reserved
1788 keyword to be used with 'bool' variable types. That means you could
1789 not include C99-style headers which use bools without triggering
1790 build errors in sieve.c.
1792 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1793 as mailq or other aliases. Changed the --help handling significantly
1794 to do whats expected. exim_usage() emits usage/help information.
1796 SC/01 Added the -bylocaldomain option to eximstats.
1798 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1800 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1802 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1808 PH/01 Another patch from the Sieve maintainer.
1810 PH/02 When an IPv6 address is converted to a string for single-key lookup
1811 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1812 dots are used instead of colons so that keys in lsearch files need not
1813 contain colons. This was done some time before quoting was made available
1814 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1815 (notated using the quote facility) so as to distinguish them from IPv4
1816 keys. This meant that lookups for IP addresses in host lists did not work
1817 for iplsearch lookups.
1819 This has been fixed by arranging for IPv6 addresses to be expressed with
1820 colons if the lookup type is iplsearch. This is not incompatible, because
1821 previously such lookups could never work.
1823 The situation is now rather anomolous, since one *can* have colons in
1824 ordinary lsearch keys. However, making the change in all cases is
1825 incompatible and would probably break a number of configurations.
1827 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1830 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1831 conversion specification without a maximum field width, thereby enabling
1832 a rogue spamd server to cause a buffer overflow. While nobody in their
1833 right mind would setup Exim to query an untrusted spamd server, an
1834 attacker that gains access to a server running spamd could potentially
1835 exploit this vulnerability to run arbitrary code as the Exim user.
1837 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1838 $primary_hostname instead of what libspf2 thinks the hosts name is.
1840 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1841 a directory entry by the name of the lookup key. Previously, if a
1842 symlink pointed to a non-existing file or a file in a directory that
1843 Exim lacked permissions to read, a lookup for a key matching that
1844 symlink would fail. Now it is enough that a matching directory entry
1845 exists, symlink or not. (Bugzilla 503.)
1847 PH/03 The body_linecount and body_zerocount variables are now exported in the
1850 PH/04 Added the $dnslist_matched variable.
1852 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1853 This means they are set thereafter only if the connection becomes
1856 PH/06 Added the client_condition to authenticators so that some can be skipped
1857 by clients under certain conditions.
1859 PH/07 The error message for a badly-placed control=no_multiline_responses left
1860 "_responses" off the end of the name.
1862 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1864 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1865 (without spaces) instead of just copying the configuration text.
1867 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1869 PH/11 Added $max_received_linelength.
1871 PH/12 Added +ignore_defer and +include_defer to host lists.
1873 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1874 way in which PCRE > 7.0 is built.
1876 PH/14 Implemented queue_only_load_latch.
1878 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1879 MAIL command. The effect was to mangle the value on 64-bit systems.
1881 PH/16 Another patch from the Sieve maintainer.
1883 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1885 PH/18 If a system quota error occurred while trying to create the file for
1886 a maildir delivery, the message "Mailbox is full" was not appended to the
1887 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1888 only to a quota excession during the actual writing of the file.
1890 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1891 characters?) which causes problems in log lines. The DN values are now
1892 passed through string_printing() before being added to log lines.
1894 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1895 and InterBase are left for another time.)
1897 PH/21 Added message_body_newlines option.
1899 PH/22 Guard against possible overflow in moan_check_errorcopy().
1901 PH/23 POSIX allows open() to be a macro; guard against that.
1903 PH/24 If the recipient of an error message contained an @ in the local part
1904 (suitably quoted, of course), incorrect values were put in $domain and
1905 $local_part during the evaluation of errors_copy.
1911 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1912 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1915 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1916 issue a MAIL command.
1918 PH/02 In an ACL statement such as
1920 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1922 if a client was not listed at all, or was listed with a value other than
1923 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1924 the condition was not true (as it should be), so access was not denied.
1925 The bug was that the ! inversion was incorrectly passed on to the second
1926 item. This has been fixed.
1928 PH/03 Added additional dnslists conditions == and =& which are different from
1929 = and & when the dns lookup returns more than one IP address.
1931 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1932 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1934 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1935 FSYNC, which compiles an option called disable_fsync that allows for
1936 bypassing fsync(). The documentation is heavily laced with warnings.
1938 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1940 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1941 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1942 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1943 including adding "make clean"; (3) Added -fPIC when compiling the test
1944 dynamically loaded module, to get rid of a warning.
1946 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1947 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1948 = 0s. The bug is otherwise harmless.
1950 PH/07 There was a bug in the dovecot authenticator such that the value of
1951 $auth1 could be overwritten, and so not correctly preserved, after a
1952 successful authentication. This usually meant that the value preserved by
1953 the server_setid option was incorrect.
1955 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1957 PH/09 Installed PCRE release 7.0.
1959 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1960 run for batched SMTP input. It is now run at the start of every message
1961 in the batch. While fixing this I discovered that the process information
1962 (output by running exiwhat) was not always getting set for -bs and -bS
1963 input. This is fixed, and it now also says "batched" for BSMTP.
1965 PH/11 Added control=no_pipelining.
1967 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1968 patch, slightly modified), and move the expansion of helo_data till after
1969 the connection is made in the smtp transport (so it can use these
1972 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1974 PH/14 Added log_selector = +pid.
1976 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1978 PH/16 Add ${if forany and ${if forall.
1980 PH/17 Added dsn_from option to vary the From: line in DSNs.
1982 PH/18 Flush SMTP output before performing a callout, unless control =
1983 no_callout_flush is set.
1985 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1986 was true (the default) a successful delivery failed to delete the retry
1987 item, thus causing premature timeout of the address. The bug is now
1990 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1992 PH/21 Long custom messages for fakedefer and fakereject are now split up
1993 into multiline reponses in the same way that messages for "deny" and
1994 other ACL rejections are.
1996 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1997 with slight modification.
1999 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2000 draft, changing the syntax and factoring some duplicate code".
2002 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2003 for deliveries of the second and subsequent messages over the same SMTP
2006 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2007 ${reduce, with only minor "tidies".
2009 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2011 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2012 expansion side effects.
2014 PH/27 When a message times out after an over-quota error from an Exim-imposed
2015 quota, the bounce message says "mailbox is full". This message was not
2016 being given when it was a system quota that was exceeded. It now should
2019 MH/03 Made $recipients available in local_scan(). local_scan() already has
2020 better access to the recipient list through recipients_list[], but
2021 $recipients can be useful in postmaster-provided expansion strings.
2023 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2024 in the case of a MAIL command with additional options following the
2025 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2026 were accidentally chopped off.
2028 PH/29 SMTP synchronization checks are implemented when a command is read -
2029 there is a check that no more input is waiting when there shouldn't be
2030 any. However, for some commands, a delay in an ACL can mean that it is
2031 some time before the response is written. In this time, more input might
2032 arrive, invalidly. So now there are extra checks after an ACL has run for
2033 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2034 pipelining has not been advertised.
2036 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2038 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2039 returned in the SMTP response when smtp_return_error_details was false.
2040 This has been fixed.
2042 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2043 instead of the C I/O that was originally supplied, because problems were
2044 reported on Solaris.
2046 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2047 Exim which did not show up earlier: it was assuming that a call to
2048 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2049 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2050 was a macro that became an assignment, so it seemed to work. This has
2051 changed to a proper function call with a void return, hence the compile
2052 error. Exim's code has been fixed.
2054 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2057 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2059 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2061 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2062 because it turns out to be too big or there is a timeout) there is no
2063 "Completed" line in the log. When some messages of this type were
2064 selected by exigrep, they were listed as "not completed". Others were
2065 picked up by some special patterns. I have improved the selection
2066 criteria to be more general.
2068 PH/38 The host_find_failed option in the manualroute router can now be set
2069 to "ignore", to completely ignore a host whose IP address cannot be
2070 found. If all hosts are ignored, the behaviour is controlled by the new
2071 host_all_ignored option.
2073 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2074 homing or because of multiple MX records with /mx) generated more than
2075 one IP address, and the following item turned out to be the local host,
2076 all the secondary addresses of the first item were incorrectly removed
2077 from the list, along with the local host and any following hosts (which
2078 is what is supposed to happen).
2080 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2081 whoever called Exim into the -H file. In the case of the daemon it was
2082 behaving confusingly. When first started, it used values for whoever
2083 started the daemon, but after a SIGHUP it used the Exim user (because it
2084 calls itself on a restart). I have changed the code so that it now always
2087 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2088 message are rejected with the same error (e.g. no authentication or bad
2089 sender address), and a DATA command is nevertheless sent (as can happen
2090 with PIPELINING or a stupid MUA), the error message that was given to the
2091 RCPT commands is included in the rejection of the DATA command. This is
2092 intended to be helpful for MUAs that show only the final error to their
2095 PH/42 Another patch from the Sieve maintainer.
2097 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2098 Eximstats - Fixed some broken HTML links and added missing column headers
2100 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2101 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2103 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2105 PH/43 Yet another patch from the Sieve maintainer.
2107 PH/44 I found a way to check for a TCP/IP connection going away before sending
2108 the response to the final '.' that terminates a message, but only in the
2109 case where the client has not sent further data following the '.'
2110 (unfortunately, this is allowed). However, in many cases there won't be
2111 any further data because there won't be any more messages to send. A call
2112 to select() can be used: if it shows that the input is "ready", there is
2113 either input waiting, or the socket has been closed. An attempt to read
2114 the next input character can distinguish the two cases. Previously, Exim
2115 would have sent an OK response which the client would never have see.
2116 This could lead to message repetition. This fix should cure that, at
2117 least in a lot of common cases.
2119 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2120 advertised in response to EHLO.
2126 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2127 fixed by 4.65/MH/01 (is this a record?) are fixed:
2129 (i) An empty string was always treated as zero by the numeric comparison
2130 operators. This behaviour has been restored.
2132 (ii) It is documented that the numeric comparison operators always treat
2133 their arguments as decimal numbers. This was broken in that numbers
2134 starting with 0 were being interpreted as octal.
2136 While fixing these problems I realized that there was another issue that
2137 hadn't been noticed. Values of message_size_limit (both the global option
2138 and the transport option) were treated as octal if they started with 0.
2139 The documentation was vague. These values are now always treated as
2140 decimal, and I will make that clear in the documentation.
2146 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2147 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2150 MH/01 Don't check that the operands of numeric comparison operators are
2151 integers when their expansion is in "skipping" mode (fixes bug
2152 introduced by 4.64-PH/07).
2154 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2155 child addresses, Exim now panics and dies. Previously, because the count
2156 is held in a short int, deliveries were likely to be lost. As such a
2157 large number of recipients for a single message is ridiculous
2158 (performance will be very, very poor), I have chosen to impose a limit
2159 rather than extend the field.
2165 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2166 leftover -K file (the existence of which was triggered by #402).
2167 While we were at it, introduced process PID as part of the -K
2168 filename. This should rule out race conditions when creating
2171 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2172 processing considerably. Previous code took too long for large mails,
2173 triggering a timeout which in turn triggers #401.
2175 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2176 in the DK code in transports.c. sendfile() is not really portable,
2177 hence the _LINUX specificness.
2179 TF/01 In the add_headers option to the mail command in an Exim filter,
2180 there was a bug that Exim would claim a syntax error in any
2181 header after the first one which had an odd number of characters
2184 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2185 callout verification, Exim cached a "reject" for the entire domain. This
2186 is correct for most verifications, but it is not correct for a recipient
2187 verification with use_sender or use_postmaster set, because in that case
2188 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2189 case of MAIL FROM:<> rejection from other early rejections (e.g.
2190 rejection of HELO). When verifying a recipient using a non-null MAIL
2191 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2192 Whatever the result of the callout, the value of the domain cache is
2193 left unchanged (for any other kind of callout, getting as far as trying
2194 RCPT means that the domain itself is ok).
2196 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2199 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2200 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2201 session, the daemon ignores ECONNECT errors and logs others; it now
2202 ignores EPIPE as well.
2204 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2205 (quoted-printable decoding).
2207 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2208 later the small subsequent patch to fix an introduced bug.
2210 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2212 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2214 PH/08 An error is now given if message_size_limit is specified negative.
2216 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2217 to be given (somewhat) arbitrary names.
2219 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2222 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2223 miscellaneous code fixes
2225 PH/10 Added the log_reject_target ACL modifier to specify where to log
2228 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2229 hostname. This is wrong, because it relates to the incoming message (and
2230 probably the interface on which it is arriving) and not to the outgoing
2231 callout (which could be using a different interface). This has been
2232 changed to use the value of the helo_data option from the smtp transport
2233 instead - this is what is used when a message is actually being sent. If
2234 there is no remote transport (possible with a router that sets up host
2235 addresses), $smtp_active_hostname is used.
2237 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2238 tweaks were necessary in order to get it to work (see also 21 below):
2239 (a) The code assumed that strncpy() returns a negative number on buffer
2240 overflow, which isn't the case. Replaced with Exim's string_format()
2242 (b) There were several signed/unsigned issues. I just did the minimum
2243 hacking in of casts. There is scope for a larger refactoring.
2244 (c) The code used strcasecmp() which is not a standard C function.
2245 Replaced with Exim's strcmpic() function.
2246 (d) The code set only $1; it now sets $auth1 as well.
2247 (e) A simple test gave the error "authentication client didn't specify
2248 service in request". It would seem that Dovecot has changed its
2249 interface. Fortunately there's a specification; I followed it and
2250 changed what the client sends and it appears to be working now.
2252 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2255 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2256 address A is aliased to B and C, where B exists and C does not. Without
2257 -v the output is "A verified" because verification stops after a
2258 successful redirection if more than one address is generated. However,
2259 with -v the child addresses are also verified. Exim was outputting "A
2260 failed to verify" and then showing the successful verification for C,
2261 with its parentage. It now outputs "B failed to verify", showing B's
2262 parentage before showing the successful verification of C.
2264 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2265 look up a TXT record in a specific list after matching in a combined
2268 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2269 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2270 they consult the DNS. I had assumed they would set it the way they
2271 wanted; and indeed my experiments on Linux seem to show that in some
2272 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2273 To be on the safe side, however, I have now made the interface to
2274 host_find_byname() similar to host_find_bydns(), with an argument
2275 containing the DNS resolver options. The host_find_byname() function now
2276 sets these options at its start, just as host_find_bydns() does. The smtp
2277 transport options dns_qualify_single and dns_search_parents are passed to
2278 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2279 of host_find_byname() use the default settings of RES_DEFNAMES
2280 (qualify_single) but not RES_DNSRCH (search_parents).
2282 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2283 spool_read_header() do less string testing, by means of a preliminary
2284 switch on the second character of optional "-foo" lines. (This is
2285 overdue, caused by the large number of possibilities that now exist.
2286 Originally there were few.) While I was there, I also converted the
2287 str(n)cmp tests so they don't re-test the leading "-" and the first
2288 character, in the hope this might squeeze out yet more improvement.
2290 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2291 flag allowing group syntax was set by the header_syntax check but not
2292 turned off, possible causing trouble later; (2) The flag was not being
2293 set at all for the header_verify test, causing "group"-style headers to
2294 be rejected. I have now set it in this case, and also caused header_
2295 verify to ignore an empty address taken from a group. While doing this, I
2296 came across some other cases where the code for allowing group syntax
2297 while scanning a header line wasn't quite right (mostly, not resetting
2298 the flag correctly in the right place). These bugs could have caused
2299 trouble for malformed header lines. I hope it is now all correct.
2301 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2302 with the "reply" argument non-NULL. The code, however (which originally
2303 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2304 but it didn't always do it. This confused somebody who was copying the
2305 code for some other use. I have removed all the tests.
2307 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2308 feature that was used to support insecure browsers during the U.S. crypto
2309 embargo. It requires special client support, and Exim is probably the
2310 only MTA that supported it -- and would never use it because real RSA is
2311 always available. This code has been removed, because it had the bad
2312 effect of slowing Exim down by computing (never used) parameters for the
2313 RSA_EXPORT functionality.
2315 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2316 authenticator to fail if there's a tab character in the incoming data
2317 (there should never be unless someone is messing about, as it's supposed
2318 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2319 if the connection is using TLS or if the remote IP is the same as the
2320 local IP, and the "valid-client-cert option" if a client certificate has
2323 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2324 authenticators. This can be used for authorization after authentication
2325 succeeds. (In the case of plaintext, it servers for both authentication
2328 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2329 if any retry times were supplied.
2331 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2332 connection, where there is no client IP address to check. In this
2333 situation, the verify now always succeeds.
2335 PH/25 Applied John Jetmore's -Mset patch.
2337 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2339 PH/27 In a string expansion for a processed (not raw) header when multiple
2340 headers of the same name were present, leading whitespace was being
2341 removed from all of them, but trailing whitespace was being removed only
2342 from the last one. Now trailing whitespace is removed from each header
2343 before concatenation. Completely empty headers in a concatenation (as
2344 before) are ignored.
2346 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2347 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2349 PH/29 [Removed. This was a change that I later backed out, and forgot to
2350 correct the ChangeLog entry (that I had efficiently created) before
2351 committing the later change.]
2353 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2354 address errors (4xx response to RCPT) over the same connection as other
2355 messages routed to the same hosts. Such deliveries are always "forced",
2356 so retry times are not inspected. This resulted in far too many retries
2357 for the affected addresses. The effect occurred only when there were more
2358 hosts than the hosts_max_try setting in the smtp transport when it had
2359 the 4xx errors. Those hosts that it had tried were not added to the list
2360 of hosts for which the message was waiting, so if all were tried, there
2361 was no problem. Two fixes have been applied:
2363 (i) If there are any address or message errors in an SMTP delivery, none
2364 of the hosts (tried or untried) are now added to the list of hosts
2365 for which the message is waiting, so the message should not be a
2366 candidate for sending over the same connection that was used for a
2367 successful delivery of some other message. This seems entirely
2368 reasonable: after all the message is NOT "waiting for some host".
2369 This is so "obvious" that I'm not sure why it wasn't done
2370 previously. Hope I haven't missed anything, but it can't do any
2371 harm, as the worst effect is to miss an optimization.
2373 (ii) If, despite (i), such a delivery is accidentally attempted, the
2374 routing retry time is respected, so at least it doesn't keep
2375 hammering the server.
2377 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2378 in ${readsocket because some servers need this prod.
2380 PH/32 Added some extra debug output when updating a wait-xxx database.
2382 PH/33 The hint "could be header name not terminated by colon", which has been
2383 given for certain expansion errors for a long time, was not being given
2384 for the ${if def:h_colon_omitted{... case.
2386 PH/34 The spec says: "With one important exception, whenever a domain list is
2387 being scanned, $domain contains the subject domain." There was at least
2388 one case where this was not true.
2390 PH/35 The error "getsockname() failed: connection reset by peer" was being
2391 written to the panic log as well as the main log, but it isn't really
2392 panic-worthy as it just means the connection died rather early on. I have
2393 removed the panic log writing for the ECONNRESET error when getsockname()
2396 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2397 runs only) independently of the message's sender address. This meant
2398 that, if the 4xx error was in fact related to the sender, a different
2399 message to the same recipient with a different sender could confuse
2400 things. In particualar, this can happen when sending to a greylisting
2401 server, but other circumstances could also provoke similar problems.
2402 I have changed the default so that the retry time for these errors is now
2403 based a combination of the sender and recipient addresses. This change
2404 can be overridden by setting address_retry_include_sender=false in the
2407 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2408 remote server are returned as part of bounce messages. This was not
2409 happening for LMTP over a pipe (the lmtp transport), but now it is the
2410 same for both kinds of LMTP.
2412 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2413 in header lines that were in fact CNAMEs. This is no longer the case.
2415 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2416 and queue runs started by the daemon processed all messages. This has
2417 been fixed so that -R and -S can now usefully be given with -q<time>.
2419 PH/40 Import PCRE release 6.7 (fixes some bugs).
2421 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2423 PH/42 Give an error if -q is specified more than once.
2425 PH/43 Renamed the variables $interface_address and $interface_port as
2426 $received_ip_address and $received_port, to make it clear that these
2427 values apply to message reception, and not to the outgoing interface when
2428 a message is delivered. (The old names remain recognized, of course.)
2430 PH/44 There was no timeout on the connect() call when using a Unix domain
2431 socket in the ${readsocket expansion. There now is.
2433 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2434 be meaningful with "accept".
2436 SC/01 Eximstats V1.43
2437 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2439 SC/02 Eximstats V1.44
2440 Use a glob alias rather than an array ref in the generated
2441 parser. This improves both readability and performance.
2443 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2444 Collect SpamAssassin and rejection statistics.
2445 Don't display local sender or destination tables unless
2446 there is data to show.
2447 Added average volumes into the top table text output.
2449 SC/04 Eximstats V1.46
2450 Collect data on the number of addresses (recipients)
2451 as well as the number of messages.
2453 SC/05 Eximstats V1.47
2454 Added 'Message too big' to the list of mail rejection
2455 reasons (thanks to Marco Gaiarin).
2457 SC/06 Eximstats V1.48
2458 Mainlog lines which have GMT offsets and are too short to
2459 have a flag are now skipped.
2461 SC/07 Eximstats V1.49 (Alain Williams)
2462 Added the -emptyok flag.
2464 SC/08 Eximstats V1.50
2465 Fixes for obtaining the IP address from reject messages.
2467 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2468 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2469 whitesspace changes from 4.64-PH/27
2471 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2474 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2475 are found, allow negative numbers in numeric criteria)
2477 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2479 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2480 to match changes made in 4.64-PH/43
2482 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2484 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2485 "domain" is actually an IP address, and force a failure. This locks out
2486 those revolvers/nameservers that support "A-for-A" lookups, in
2487 contravention of the specifications.
2489 PH/48 When a host name was looked up from an IP address, and the subsequent
2490 forward lookup of the name timed out, the host name was left in
2491 $sender_host_name, contrary to the specification.
2493 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2494 restricted to single-key lookups, Exim was not diagnosing an error if
2495 * or *@ was used with a query-style lookup.
2497 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2499 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2500 long ago, but noone interested enough thought of it. Let's just say that
2501 the "1.1" means that there are some new functions that weren't there at
2502 some point in the past.
2504 PH/51 Error processing for expansion failure of helo_data from an smtp
2505 transport during callout processing was broken.
2507 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2508 tested/used via the -bh/-bhc/-bs options.
2510 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2511 bug, fixed in subsequent PCRE releases).
2513 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2514 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2516 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2522 SC/01 Use a glob alias rather than an array ref in eximstats generated
2523 parser. This improves both readability and performance.
2525 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2526 Don't display local sender or destination tables in eximstats unless
2527 there is data to show.
2528 Added average volumes into the eximstats top table text output.
2530 SC/03 Collect data on the number of addresses (recipients) as well
2531 as the number of messages in eximstats.
2533 TF/01 Correct an error in the documentation for the redirect router. Exim
2534 does (usually) call initgroups() when daemonizing.
2536 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2537 with consistent privilege compared to when running as a daemon.
2539 TF/03 Note in the spec that $authenticated_id is not set for local
2540 submissions from trusted users.
2542 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2543 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2545 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2546 by adding some example configuration directives to the default
2547 configuration file. A little bit of work is required to uncomment the
2548 directives and define how usernames and passwords are checked, but
2549 there is now a framework to start from.
2551 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2552 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2553 without this. I don't know how relevant this is to other LDAP libraries.
2555 PH/02 Add the verb name to the "unknown ACL verb" error.
2557 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2559 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2561 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2562 directory not expanded when it should be if an expanded home directory
2563 was set for the address (which is overridden by the transport).
2565 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2568 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2569 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2570 because it is too late at that time, and has no effect.
2572 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2573 security issue with \' (bugzilla #107). I could not use the
2574 PQescapeStringConn() function, because it needs a PGconn value as one of
2577 PH/08 When testing addresses using -bt, indicate those final addresses that
2578 are duplicates that would not cause an additional delivery. At least one
2579 person was confused, thinking that -bt output corresponded to deliveries.
2580 (Suppressing duplicates isn't a good idea as you lose the information
2581 about possibly different redirections that led to the duplicates.)
2583 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2584 systems where poll() doesn't work, in particular OS X.
2586 PH/10 Added more information to debugging output for retry time not reached.
2588 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2589 operations in malware.c.
2591 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2594 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2595 syslog_duplication set false, log lines that would normally be written
2596 both the the main log and to the reject log were not written to syslog at
2599 PH/14 In the default configuration, change the use of "message" in ACL warn
2600 statements to "add_header".
2602 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2603 not followed by a command (e.g. "seen endif").
2605 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2606 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2609 PH/17 Added extra conditions to the default value of delay_warning_condition
2613 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2614 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2615 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2618 The Auto-Submitted: and various List- headers are standardised, whereas I
2619 don't think Precedence: ever was.
2621 PH/18 Refactored debugging code in route_finduser() to show more information,
2622 in particular, the error code if getpwnam() issues one.
2624 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2625 This is apparently needed in addition to the PH/07 change above to avoid
2626 any possible encoding problems.
2628 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2629 but not after initializing Perl.
2631 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2632 output them only if debugging. By default they are written stderr,
2633 apparently, which is not desirable.
2635 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2638 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2641 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2643 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2644 authenticated or an ident call has been made. Suppress the default
2645 values for $authenticated_id and $authenticated_sender (but permit -oMai
2646 and -oMas) when testing with -bh.
2648 PH/24 Re-jigged the order of the tests in the default configuration so that the
2649 tests for valid domains and recipients precede the DNS black list and CSA
2650 tests, on the grounds that those ones are more expensive.
2652 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2653 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2654 This bug exists in every version of Exim that I still have, right back to
2657 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2658 However, an attempt to turn on case-sensitivity in a regex key by
2659 including (?-i) didn't work because the subject string was already
2660 lowercased, and the effects were non-intuitive. It turns out that a
2661 one-line patch can be used to allow (?-i) to work as expected.
2667 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2668 other effects) broke the use of negated acl sub-conditions.
2670 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2673 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2674 "Deny" causes Exim to reject the incoming connection with a 554 error.
2675 Unfortunately, if there is a major crisis, such as a disk failure,
2676 tcp-wrappers gives "deny", whereas what one would like would be some
2677 kind of temporary error. A kludge has been added to help with this.
2678 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2679 554 error is used if errno is still zero or contains ENOENT (which occurs
2680 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2683 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2685 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2686 errors. Otherwise a message that provokes a temporary error (when other
2687 messages do not) can cause a whole host to time out.
2689 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2690 addresses were routed directly to files or pipes from a redirect router.
2691 File deliveries just didn't batch; pipe deliveries might have suffered
2694 PH/06 A failure to get a lock for a hints database would erroneously always say
2695 "Failed to get write lock", even when it was really a read lock.
2697 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2698 of 0600. This has been changed to use the value of the lockfile_mode
2699 option (which defaults to 0600).
2701 PH/08 Applied small patch from the Sieve maintainer.
2703 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2704 folder from quota calculations, a direct delivery into this folder messed
2705 up the contents of the maildirsize file. This was because the regex was
2706 used only to exclude .Trash (or whatever) when the size of the mailbox
2707 was calculated. There was no check that a delivery was happening into an
2708 excluded directory. This bug has been fixed by ignoring all quota
2709 processing for deliveries into excluded directories.
2711 PH/10 Added the maildirfolder_create_regex option to appendfile.
2717 PH/01 The code for finding all the local interface addresses on a FreeBSD
2718 system running IPv6 was broken. This may well have applied to all BSD
2719 systems, as well as to others that have similar system calls. The broken
2720 code found IPv4 interfaces correctly, but gave incorrect values for the
2721 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2722 that it would not match correctly against @[] and not recognize the IPv6
2725 PH/02 The ipliteral router was not recognizing addresses of the form user@
2726 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2728 PH/03 Added disable_ipv6.
2730 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2731 decoded form, because it is most often used to construct To: headers
2732 lines in autoreplies, and the decoded form may well be syntactically
2733 invalid. However, $reply_address has leading white space removed, and all
2734 newlines turned into spaces so that the autoreply transport does not
2737 PH/05 If group was specified without a user on a router, and no group or user
2738 was specified on a transport, the group from the router was ignored.
2740 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2741 for visible compile-time settings that can be used to change these
2742 numbers, for those that want even more. Backwards compatibility with old
2743 spool files has been maintained. However, going back to a previous Exim
2744 release will lost any variables that are in spool files.
2746 PH/07 Two small changes when running in the test harness: increase delay when
2747 passing a TCP/IP connection to a new process, in case the original
2748 process has to generate a bounce, and remove special handling of
2749 127.0.0.2 (sic), which is no longer necessary.
2751 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2752 be the same on different OS.
2754 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2757 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2758 whether --show-vars was specified or not
2760 JJ/02 exipick: Added support for new ACL variable spool format introduced
2763 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2764 syntactically invalid From: or Reply-to: line, and a filter used this to
2765 generate an autoreply, and therefore failed to obtain an address for the
2766 autoreply, Exim could try to deliver to a non-existent relative file
2767 name, causing unrelated and misleading errors. What now happens is that
2768 it logs this as a hard delivery error, but does not attempt to create a
2771 PH/11 The exinext utility has a -C option for testing purposes, but although
2772 the given file was scanned by exinext itself; it wasn't being passed on
2773 when Exim was called.
2775 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2776 an end-of-file indication when reading a command response.
2778 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2779 compiled. In many other places in Exim, IPv6 addresses are always
2780 recognized, so I have changed this. It also means that IPv4 domain
2781 literals of the form [IPV4:n.n.n.n] are now always recognized.
2783 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2784 used if the router is not running as root, for example, when verifying at
2785 ACL time, or when using -bh. The debugging output from this situation was
2786 non-existent - all you got was a failure to exec. I have made two
2789 (a) Failures to set uid/gid, the current directory, or a process leader
2790 in a subprocess such as that created by queryprogram now generate
2791 suitable debugging ouput when -d is set.
2793 (b) The queryprogram router detects when it is not running as root,
2794 outputs suitable debugging information if -d is set, and then runs
2795 the subprocess without attempting to change uid/gid.
2797 PH/15 Minor change to Makefile for building test_host (undocumented testing
2800 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2801 additional section of a DNS packet that returns MX or SRV records.
2802 Instead, it always explicitly searches for A/AAAA records. This avoids
2803 major problems that occur when a DNS server includes only records of one
2804 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2805 fixed another bug: if SRV records were looked up and the corresponding
2806 address records were *not* found in the additional section, the port
2807 values from the SRV records were lost.
2809 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2810 using the correct key (the original address) when searching the retry
2811 rules in order to find which one to use for generating the retry hint.
2813 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2814 adding the default one. Similarly, if it contains a Reply-To: header, the
2815 errors_reply_to option, if set, is not used.
2817 PH/19 When calculating a retry time, Exim used to measure the "time since
2818 failure" by looking at the "first failed" field in the retry record. Now
2819 it does not use this if it is later than than the arrival time of the
2820 message. Instead it uses the arrival time. This makes for better
2821 behaviour in cases where some deliveries succeed, thus re-setting the
2822 "first failed" field. An example is a quota failure for a huge message
2823 when small messages continue to be delivered. Without this change, the
2824 "time since failure" will always be short, possible causing more frequent
2825 delivery attempts for the huge message than are intended.
2826 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2828 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2829 $1, $2, $3) because the numerical variables can be reset during some
2830 expansion items (e.g. "match"), thereby losing the authentication data.
2832 PH/21 Make -bV show the size of off_t variables so that the test suite can
2833 decide whether to run tests for quotas > 2G.
2835 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2836 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2837 filecount value is greater than 2G or if a quota value is greater than 2G
2838 on a system where the size of off_t is not greater than 4, a panic error
2841 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2842 never match. The debug and -bh output now contains an explicit error
2843 message indicating a malformed IPv4 address or mask.
2845 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2846 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2847 PH/23 above applies.
2849 PH/25 Do not write to syslog when running in the test harness. The only
2850 occasion when this arises is a failure to open the main or panic logs
2851 (for which there is an explicit test).
2853 PH/26 Added the /no_tell option to "control=freeze".
2855 PH/27 If a host name lookup failed very early in a connection, for example, if
2856 the IP address matched host_lookup and the reverse lookup yielded a name
2857 that did not have a forward lookup, an error message of the form "no IP
2858 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2859 could be logged. Now it outputs the IP address instead of "NULL".
2861 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2862 allows the sender and the authenticated sender to be set when
2863 submitting a message from within Exim. Since child_open_exim() is
2864 documented for local_scan(), the new function should be too.
2866 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2867 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2868 results in an empty string is now treated as unset.
2870 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2872 PH/31 Added sender_verify_fail logging option.
2874 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2875 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2876 tidied the source and removed it altogether.
2878 PH/33 When a queue run was abandoned because the load average was too high, a
2879 log line was always written; now it is written only if the queue_run log
2880 selector is set. In addition, the log line for abandonment now contains
2881 information about the queue run such as the pid. This is always present
2882 in "start" and "stop" lines but was omitted from the "abandon" line.
2884 PH/34 Omit spaces between a header name and the colon in the error message that
2885 is given when verify = headers_syntax fails (if there are lots of them,
2886 the message gets confusing).
2888 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2889 names, as there are now some PTR records that contain slashes. This check
2890 is only to protect against broken name servers that fall over on strange
2891 characters, so the fact that it applies to all lookups doesn't matter.
2893 PH/36 Now that the new test suite is complete, we can remove some of the
2894 special code in Exim that was needed for the old test suite. For example,
2895 sorting DNS records because real resolvers return them in an arbitrary
2896 order. The new test suite's fake resolver always returns records in the
2899 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2900 bounces) except when queue_only is set, to avoid logging races between
2901 the different processes.
2903 PH/38 Panic-die if .include specifies a non-absolute path.
2905 PH/39 A tweak to the "H" retry rule from its user.
2907 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2908 a label. They prevented compilation on older perls.
2910 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2911 a warning to be raised on newish perls.
2913 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2914 on queue. Changes to match documented behaviour of showing count of
2915 messages matching specified criteria.
2917 PH/40 Changed the default ident timeout from 30s to 5s.
2919 PH/41 Added support for the use of login_cap features, on those BSD systems
2920 that have them, for controlling the resources used by pipe deliveries.
2922 PH/42 The content-scanning code uses fopen() to create files in which to put
2923 message data. Previously it was not paying any attention to the mode of
2924 the files. Exim runs with umask(0) because the rest of the code creates
2925 files with open(), and sets the required mode explicitly. Thus, these
2926 files were ending up world-writeable. This was not a big issue, because,
2927 being within the spool directory, they were not world-accessible. I have
2928 created a function called modefopen, which takes an additional mode
2929 argument. It sets umask(777), creates the file, chmods it to the required
2930 mode, then resets the umask. All the relevant calls to fopen() in the
2931 content scanning code have been changed to use this function.
2933 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2934 to 24 hours. This avoids potential overflow problems when processing G
2935 and H retry rules. I suspect nobody ever tinkers with this value.
2937 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2939 PH/45 When the plaintext authenticator is running as a client, the server's
2940 challenges are checked to ensure they are valid base64 strings. By
2941 default, the authentication attempt is cancelled if an invalid string is
2942 received. Setting client_ignore_invalid_base64 true ignores these errors.
2943 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2944 they are received. Thus, the responses can be made to depend on the
2945 challenges. If an invalid string is ignored, an empty string is placed in
2948 PH/46 Messages that are created by the autoreply transport now contains a
2949 References: header, in accordance with RFCs 2822 and 3834.
2951 PH/47 Added authenticated_sender_force to the smtp transport.
2953 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2955 PH/49 Installed latest patch from the Sieve maintainer.
2957 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2958 was also set, the appendfile transport was unnecessarily scanning a
2959 directory of message files (e.g. for maildir delivery) to find the count
2960 of files (along with the size), even though it did not need this
2961 information. It now does the scan only if it needs to find either the
2962 size of the count of files.
2964 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2966 PH/52 Two bugs concerned with error handling when the smtp transport is
2969 (i) Exim was not creating retry information for temporary errors given
2970 for individual recipients after the DATA command when the smtp transport
2971 was used in LMTP mode. This meant that they could be retried too
2972 frequently, and not timed out correctly.
2974 (ii) Exim was setting the flag that allows error details to be returned
2975 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2976 recipients that were returned after the DATA command.
2978 PH/53 This is related to PH/52, but is more general: for any failing address,
2979 when detailed error information was permitted to be returned to the
2980 sender, but the error was temporary, then after the final timeout, only
2981 "retry timeout exceeded" was returned. Now it returns the full error as
2982 well as "retry timeout exceeded".
2984 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2985 do this, and (what is worse) MTAs that accept it.
2987 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2988 will now be deprecated.
2990 PH/56 New os.c-cygwin from the Cygwin maintainer.
2992 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2993 formats (previously only available in exim formats via -bpr, -bpru,
2994 and -bpra. Now also available in native and exiqgrep formats)
2996 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2997 with very large, slow to parse queues
2999 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3001 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3003 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3004 responses to authentication challenges, though it was showing the
3005 challenges; (ii) I've removed the CR characters from the debug output for
3008 PH/58 Allow for the insertion of a newline as well as a space when a string
3009 is turned into more than one encoded-word during RFC 2047 encoding. The
3010 Sieve code now uses this.
3012 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3013 data_4xx, lost_connection, tls_required.
3015 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3016 message was being sent as an SMTP response.
3018 PH/61 Add -l and -k options to exicyclog.
3020 PH/62 When verifying, if an address was redirected to one new address, so that
3021 verification continued, and the new address failed or deferred after
3022 having set something in $address_data, the value of $address_data was not
3023 passed back to the ACL. This was different to the case when no
3024 redirection occurred. The value is now passed back in both cases.
3026 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3027 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3028 use login_cap.h, so on its own it isn't the distinguishing feature. The
3029 new name refers directly to the setclassresources() function.
3031 PH/65 Added configuration files for NetBSD3.
3033 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3035 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3036 is preferred over IPv4.
3038 PH/68 The bounce_return_message and bounce_return_body options were not being
3039 honoured for bounces generated during the reception of non-SMTP messages.
3040 In particular, this applied to messages rejected by the ACL. This bug has
3041 been fixed. However, if bounce_return_message is true and bounce_return_
3042 body is false, the headers that are returned for a non-SMTP message
3043 include only those that have been read before the error was detected.
3044 (In the case of an ACL rejection, they have all been read.)
3046 PH/69 The HTML version of the specification is now built in a directory called
3047 spec_html instead of spec.html, because the latter looks like a path with
3048 a MIME-type, and this confuses some software.
3050 PH/70 Catch two compiler warnings in sieve.c.
3052 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3053 function verify_get_ident() calls ip_connect() to connect a socket, but
3054 if the "connect()" function timed out, ip_connect() used to close the
3055 socket. However, verify_get_ident() also closes the socket later, and in
3056 between Exim writes to the log, which may get opened at this point. When
3057 the socket was closed in ip_connect(), the log could get the same file
3058 descriptor number as the socket. This naturally causes chaos. The fix is
3059 not to close the socket in ip_connect(); the socket should be closed by
3060 the function that creates it. There was only one place in the code where
3061 this was missing, in the iplookup router, which I don't think anybody now
3062 uses, but I've fixed it anyway.
3064 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3065 well as to direct DNS lookups. Otherwise the handling of names in host
3066 lists is inconsistent and therefore confusing.
3072 PH/01 Two changes to the default runtime configuration:
3074 (1) Move the checks for relay_from_hosts and authenticated clients from
3075 after to before the (commented out) DNS black list checks.
3077 (2) Add control=submission to the relay_from_hosts and authenticated
3078 clients checks, on the grounds that messages accepted by these
3079 statements are most likely to be submissions.
3081 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3083 (1) Generate an error if the third argument for the ${prvs expansion is
3086 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3089 (3) Reset the variables that are obtained from the first argument of
3090 ${prvscheck and used in the second argument before leaving the code,
3091 because their memory is reclaimed, so using them afterwards may do
3094 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3095 one (it's much easier than Tom thought :-).
3097 (5) Because of (4), we can now allow for the use of $prvscheck_result
3098 inside the third argument.
3100 PH/03 For some reason, the default setting of PATH when running a command from
3101 a pipe transport was just "/usr/bin". I have changed it to
3104 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3105 anything to be listed in the output from -bV.
3107 PH/05 When a filter generated an autoreply, the entire To: header line was
3108 quoted in the delivery log line, like this:
3110 => >A.N.Other <ano@some.domain> <original@ddress> ...
3112 This has been changed so that it extracts the operative address. There
3113 may be more than one such address. If so, they are comma-separated, like
3116 => >ano@some.domain,ona@other.domain <original@ddress> ...
3118 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3119 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3120 not being looked up in the rDNS to get a host name, Exim was showing the
3121 IP address twice in Received: lines, even though the IP addresses were
3122 identical. For example:
3124 Received: from [1.2.3.4] (helo=[1.2.3.4])
3126 However, if the real host name was known, it was omitting the HELO data
3127 if it matched the actual IP address. This has been tidied up so that it
3128 doesn't show the same IP address twice.
3130 PH/07 When both +timestamp and +memory debugging was on, the value given by
3131 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3132 called by the debug printing, thereby overwriting the timestamp buffer.
3133 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3135 PH/08 When the original message was included in an autoreply transport, it
3136 always said "this is a copy of the message, including all the headers",
3137 even if body_only or headers_only was set. It now gives an appropriate
3140 PH/09 Applied a patch from the Sieve maintainer which:
3142 o fixes some comments
3143 o adds the (disabled) notify extension core
3144 o adds some debug output for the result of if/elsif tests
3145 o points to the current vacation draft in the documentation
3146 and documents the missing references header update
3150 o fixes a bug in processing the envelope test (when testing
3151 multiple envelope elements, the last element determinted the
3154 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3155 Electronic Mail") by including:
3157 Auto-submitted: auto-generated
3159 in the messages that it generates (bounce messages and others, such as
3160 warnings). In the case of bounce messages for non-SMTP mesages, there was
3161 also a typo: it was using "Auto_submitted" (underscore instead of
3162 hyphen). Since every message generated by Exim is necessarily in response
3163 to another message, thes have all been changed to:
3165 Auto-Submitted: auto-replied
3167 in accordance with these statements in the RFC:
3169 The auto-replied keyword:
3171 - SHOULD be used on messages sent in direct response to another
3172 message by an automatic process,
3174 - MUST NOT be used on manually-generated messages,
3176 - MAY be used on Delivery Status Notifications (DSNs) and Message
3177 Disposition Notifications (MDNs),
3179 - MUST NOT be used on messages generated by automatic or periodic
3180 processes, except for messages which are automatic responses to
3183 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3184 to the default Received: header definition.
3186 PH/12 Added log selector acl_warn_skipped (default on).
3188 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3189 variables because (a) they are in the wrong storage pool and (b) even if
3190 they were copied, it wouldn't work properly because of the caching.
3192 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3193 checking when decoding. Apparently there are clients that generate
3194 overlong encoded strings. Why am I not surprised?
3196 PH/15 If the first argument of "${if match_address" was not empty, but did not
3197 contain an "@" character, Exim crashed. Now it writes a panic log message
3198 and treats the condition as false.
3200 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3202 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3203 extension "envelope-auth". The code is finished and in agreement with
3204 other implementations, but there is no documentation so far and in fact,
3205 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3206 not changing the active code.
3208 Print executed "if" and "elsif" statements when debugging is used. This
3209 helps a great deal to understand what a filter does.
3211 Document more things not specified clearly in RFC3028. I had all this
3212 sorted out, when out of a sudden new issues came to my mind. Oops."
3214 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3217 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3218 canonical form (as documented). However, after a host name lookup from
3219 the IP address, check_host() was doing a simple string comparison with
3220 addresses acquired from the DNS when checking that the found name did
3221 have the original IP as one of its addresses. Since any found IPv6
3222 addresses are likely to be in abbreviated form, the comparison could
3223 fail. Luckily, there already exists a function for doing the comparison
3224 by converting both addresses to binary, so now that is used instead of
3225 the text comparison.
3227 PH/20 There was another similar case to PH/19, when a complete host name was
3228 given in a host list; looking up its IP address could give an abbreviated
3229 form, whereas the current host's name might or might not be abbreviated.
3230 The same fix has been applied.
3236 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3237 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3240 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3241 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3243 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3245 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3246 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3247 Also, for "Auto-Submitted", treat anything other than "no" as
3248 non-personal, in accordance with RFC 3834. (Previously it treated
3249 anything starting "auto-" as non-personal.)
3251 TF/01 The control=submission/name=... option had a problem with syntax
3252 errors if the name included a slash character. The /name= option
3253 now slurps the rest of the string, so it can include any characters
3254 but it must come last in the list of options (after /sender_retain
3257 PH/05 Some modifications to the interface to the fake nameserver for the new
3265 TK/01 Added the "success_on_redirect" address verification option. See
3266 NewStuff for rationale and an example.
3268 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3270 PH/02 Patch to exigrep to allow it to work on syslog lines.
3272 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3273 fread() to skip over the body file's header line, because in Cygwin the
3274 header line is locked and is inaccessible.
3276 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3277 co-exist for some time) to make it clear that it is the Exim ID that is
3278 referenced, not the Message-ID: header line.
3280 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3281 string_format() function, because snprintf() does not exist on all
3284 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3285 ${stat: expansion item.
3287 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3288 protocol synchronization error", to keep the pedants happy.
3290 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3291 well as for IRIX systems, when gcc is being used. See the host.c source
3294 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3296 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3299 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3300 transport and to the smtp transport in LMTP mode.
3302 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3304 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3305 run a filter in a subprocess. This could lead to confusion in subsequent
3306 lookups in the parent process. There should also be a search_tidyup() at
3307 the end of the subprocess.
3309 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3310 only if the host matched helo_try_verify_hosts, which caused the
3311 verification to occur when the EHLO/HELO command was issued. The ACL just
3312 tested the remembered result. Now, if a previous verification attempt has
3313 not happened, "verify = helo" does it there and then.
3315 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3317 TK/03 Fix log output including CR from clamd.
3319 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3320 exist provoked a memory error which could cause a segfault.
3322 PH/15 Installed PCRE 6.2
3324 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3326 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3327 of the problem. Specifically, suggested +O2 rather than +O1 for the
3330 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3332 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3333 record, the port was not correctly logged when the outgoing_port log
3334 selector was set (it logged the transort's default port).
3336 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3337 fallback_hosts, and "hosts" in the smtp transport.
3339 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3340 host errors such as "Connection refused".
3342 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3343 authentication with radiusclient 0.4.9:
3345 - Error returned from rc_read_config was caught wrongly
3346 - Username/password not passed on to radius server due to wrong length.
3348 The presumption is that some radiusclient API changes for 4.51/PH/17
3349 were not taken care of correctly. The code is still untested by me (my
3350 Linux distribution still has 0.3.2 of radiusclient), but it was
3351 contributed by a Radius user.
3353 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3354 expanding the "port" option of the smtp transport.
3356 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3357 while reading a MIME header. Thanks to Tom Hughes for a patch.
3359 PH/24 Include config.h inside local_scan.h so that configuration settings are
3362 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3363 that in an ACL for RCPT (for example), you can examine exactly what was
3366 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3367 commands, but it was not correctly comparing the address with the actual
3368 client host address. Thus, it would show the EHLO address in Received:
3369 header lines when this was not necessary.
3371 PH/27 Added the % operator to ${eval:}.
3373 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3374 it should be ignoring failures (because with -C, for example, it has lost
3375 privilege). It wasn't ignoring creation failures other than "already
3378 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3381 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3382 because an input line was too long, either on its own, or by virtue of
3383 too many continuations, the temporary file was not being removed, and the
3384 return code was incorrect.
3386 PH/31 Missing "BOOL" in function definition in filtertest.c.
3388 PH/32 Applied Sieve patches from the maintainer.
3390 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3392 PH/33 Added "verify = not_blind".
3394 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3395 Local/Makefile (with some defaults set). These are used in built scripts
3396 such as exicyclog, but they have never been used in the exim_install
3397 script (though there are many overriding facilities there). I have
3398 arranged that the exim_install script now takes note of these two
3401 PH/35 Installed configuration files for Dragonfly.
3403 PH/36 When a locally submitted message by a trusted user did not contain a
3404 From: header, and the sender address was obtained from -f or from an SMTP
3405 MAIL command, and the trusted user did not use -F to supply a sender
3406 name, $originator_name was incorrectly used when constructing a From:
3407 header. Furthermore, $originator_name was used for submission mode
3408 messages from external hosts without From: headers in a similar way,
3409 which is clearly wrong.
3411 PH/37 Added control=suppress_local_fixups.
3413 PH/38 When log_selector = +received_sender was set, and the addition of the
3414 sender made the log line's construction buffer exactly full, or one byte
3415 less than full, an overflow happened when the terminating "\n" was
3418 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3419 when the result of a list match is failure because a DNS lookup failed.
3421 PH/40 RM_COMMAND is now used in the building process.
3423 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3424 the "build-* directories that it finds.
3426 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3427 address in a domain literal was a prefix of an interface address.
3429 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3430 when verifying a sender address, unless rewrite_headers is false.
3432 PH/44 Wrote a long comment about why errors_to addresses are verified as
3433 recipients, not senders.
3435 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3436 the ratelimit ACL was added.
3438 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3440 PH/46 Added extra information about PostgreSQL errors to the error string.
3442 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3443 avoiding the need to install special zones in a real server. This is
3444 backwards compatible; if it can't find the fake resolver, it drops back.
3445 Thus, both old and new test suites can be run.
3447 TF/02 Added util/ratelimit.pl
3449 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3450 clock is set back in time.
3452 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3453 Candler <B.Candler@pobox.com>.
3455 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3456 OK for addresses that are the result of redirections.
3458 PH/48 A number of further additions for the benefit of the new test suite,
3459 including a fake gethostbyname() that interfaces to the fake DNS resolver
3462 TF/06 The fix for widen_domains has also been applied to qualify_single and
3463 search_parents which are the other dnslookup options that can cause
3466 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3469 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3471 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3472 service" if it encountered an MX record with an empty target hostname.
3473 The message is now "an MX or SRV record indicated no SMTP service".
3475 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3476 if the DNS of the sending site is misconfigured. This is quite a
3477 common situation. This change restores the behaviour of treating a
3478 helo verification defer as a failure.
3480 PH/51 If self=fail was set on a router, the bounce message did not include the
3481 actual error message.
3487 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3489 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3490 command itself ended in error, the underlying message about the transport
3491 filter timeout was being overwritten with the pipe command error. Now the
3492 underlying error message should be appended to the second error message.
3494 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3496 PH/02 Reduce the amount of output that "make" produces by default. Full output
3497 can still be requested.
3499 PH/03 The warning log line about a condition test deferring for a "warn" verb
3500 was being output only once per connection, rather than after each
3501 occurrence (because it was using the same function as for successful
3502 "warn" verbs). This seems wrong, so I have changed it.
3504 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3505 it should not have, which might have caused a crash in the right
3506 circumstances, but probably never did.
3508 PH/04 Installed a modified version of Tony Finch's patch to make submission
3509 mode fix the return path as well as the Sender: header line, and to
3510 add a /name= option so that you can make the user's friendly name appear
3513 TF/03 Added the control = fakedefer ACL modifier.
3515 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3516 Mark Lowes for thorough testing.
3518 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3520 TK/03 Merged latest SRS patch from Miles Wilton.
3522 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3523 with the definition in sysexits.h (which is #included earlier).
3524 Fortunately, Exim does not actually use EX_OK. The code used to try to
3525 preserve the sysexits.h value, by assumimg that macro definitions were
3526 scanned for macro replacements. I have been disabused of this notion,
3527 so now the code just undefines EX_OK before #including unistd.h.
3529 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3530 in the smtp transport. When a block could not be written in a single
3531 write() function, the timeout was being re-applied to each part-write.
3532 This seems wrong - if the receiver was accepting one byte at a time it
3533 would take for ever. The timeout is now adjusted when this happens. It
3534 doesn't have to be particularly precise.
3536 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3537 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3539 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3540 without a domain if the check to <postmaster@domain> fails.
3542 SC/01 Eximstats: added -xls and the ability to specify output files
3543 (patch written by Frank Heydlauf).
3545 SC/02 Eximstats: use FileHandles for outputing results.
3547 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3549 SC/04 Eximstats: fixed display of large numbers with -nvr option
3551 SC/05 Eximstats: fixed merging of reports with empty tables.
3553 SC/06 Eximstats: added the -include_original_destination flag
3555 SC/07 Eximstats: removed tabs and trailing whitespace.
3557 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3559 TK/06 MBOX spool code: Add real "From " MBOX separator line
3560 so the .eml file is really in mbox format (even though
3561 most programs do not really care). Patch from Alex Miller.
3563 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3564 The latter is generated from $received_to and is only set if the
3565 message has one envelope recipient. SA can use these headers,
3566 obviously out-of-the-box. Patch from Alex Miller.
3568 PH/08 The ${def test on a variable was returning false if the variable's
3569 value was "0", contrary to what the specification has always said!
3570 The result should be true unless the variable is empty.
3572 PH/09 The syntax error of a character other than { following "${if
3573 def:variable_name" (after optional whitespace) was not being diagnosed.
3574 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3575 accidental colon was present, for example, could give incorrect results.
3577 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3578 result is used (not including appendfile, where other changes are about
3581 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3582 This involved changing a lot of size variables from int to off_t. It
3583 should work with maildirs and everything.
3585 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3586 spamd dying while we are connected to it.
3588 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3591 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3592 function for BDB 4.3.
3594 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3596 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3597 the sources for PCRE have been re-arranged and more files are now
3600 PH/15 The code I had for printing potentially long long variables in PH/11
3601 above was not the best (it lost precision). The length of off_t variables
3602 is now inspected at build time, and an appropriate printing format (%ld
3603 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3604 to be "long long int" or "long int". This is needed for the internal
3605 formatting function string_vformat().
3607 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3608 the configuration file to be ":syslog", then the script "guesses" where
3609 the logs files are, rather than using the compiled in default. In our
3610 case the guess is not the same as the compiled default, so the script
3611 suddenly stopped working when I started to use syslog. The patch checks
3612 to see if log_file_path is "". If so, it attempts to read it from exim
3613 with no configuration file to get the compiled in version, before it
3614 falls back to the previous guessing code."
3616 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3617 implementing BATV in an Exim configuration. See NewStuff for the gory
3620 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3621 Makefile that are specific to HP-UX.
3623 PH/18 If the "use_postmaster" option was set for a recipient callout together
3624 with the "random" option, the postmaster address was used as the MAIL
3625 FROM address for the random test, but not for the subsequent recipient
3626 test. It is now used for both.
3628 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3629 patch removes a few documentation additions to RFC 3028, because the
3630 latest draft now contains them. It adds the new en;ascii-case comparator
3631 and a new error check for 8bit text in MIME parts. Comparator and
3632 require names are now matched exactly. I enabled the subaddress
3633 extension, but it is not well tested yet (read: it works for me)."
3635 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3636 rework some of the code of TK/09 above to avoid the hardwired use of
3637 "%lld" and "long long". Replaced the call to snprintf() with a call to
3640 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3641 records point to non-existent hosts", "retry timeout exceeded", and
3642 "retry time not reached for any host after a long failure period".
3644 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3645 experimental DomainKeys support:
3647 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3648 (2) On an error such as an illegally used "control", the wrong name for
3649 the control was given.
3651 These problems did NOT occur unless DomainKeys support was compiled.
3653 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3655 PH/24 Added ${if match_ip condition.
3657 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3658 databases so that it will be absolutely obvious if a crash occurs in the
3659 DB library. This is a regular occurrence (often caused by mis-matched
3662 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3663 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3664 picked up on a user's system that detects such things. There doesn't seem
3665 to be a gcc warning option for this - only an attribute that has to be
3666 put on the function's prototype. It seems that in Fedora Core 4 they have
3667 set this on a number of new functions. No doubt there will be more in due
3670 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3671 specify a transport. However, if an address that was verified by such a
3672 router was the subject of a callout, Exim crashed because it tried to
3673 read the rcpt_include_affixes from the non-existent transport. Now it
3674 just assumes that the setting of that option is false. This bug was
3675 introduced by 4.51/PH/31.
3677 PH/28 Changed -d+all to exclude +memory, because that information is very
3678 rarely of interest, but it makes the output a lot bigger. People tend to
3679 do -d+all out of habit.
3681 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3682 code in os-type was giving problems when libc.so lives in lib64, like on
3685 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3686 aren't the modern standard, and it seems that some systems' include files
3687 don't always have them. Exim was already checking for some of the newer
3688 ones like T_AAAA, and defining it itself. I've added checks for all the
3689 record types that Exim uses.
3691 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3692 not automatically generating a new one, as it is supposed to. This
3693 prevented TLS from working. If the file did exist, but contained invalid
3694 data, a new version was generated, as expected. It was only the case of a
3695 non-existent file that was broken.
3697 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3698 with a change in libdomainkeys > 0.64.
3700 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3701 from DNS. If the selector record carries the flag, it now has
3702 precedence over the domain-wide flag.
3704 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3706 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3707 the use of a port name that isn't defined in /etc/services) occurred, the
3708 message was deferred as in a normal delivery, and thus remained on the
3709 spool, instead of being failed because of the mua_wrapper setting. This
3710 is now fixed, and I tidied up some of the mua_wrapper messages at the
3713 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3714 the messages in a hash of arrays rather than using individual hashes.
3715 This is a bit cleaner and results in dramatic memory savings, albeit
3716 at a slight CPU cost.
3718 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3719 as requested by Marc Sherman.
3721 SC/10 Eximstats: added histograms for user specified patterns as requested
3724 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3726 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3727 fopen() in the content-scanning modules that did not already have it.
3733 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3734 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3736 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3738 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3740 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3741 to test Sieve filters that use "vacation".
3743 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3744 that changes the way the GnuTLS parameters are stored in the cache file.
3745 The new format can be generated externally. For backward compatibility,
3746 if the data in the cache doesn't make sense, Exim assumes it has read an
3747 old-format file, and it generates new data and writes a new file. This
3748 means that you can't go back to an older release without removing the
3751 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3752 work if there are any delivery delays because "one_time" forces the
3753 parent to be marked "delivered", so its unseen clone is never tried
3754 again. For this reason, Exim now forbids the simultaneous setting of
3757 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3758 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3759 entry if you want to know the details. The fix, however, neglected to
3760 consider the case when local delivery batching is involved. The test for
3761 "previously delivered" was not happening when checking to see if an
3762 address could be batched with a previous (undelivered) one; under
3763 certain circumstances this could lead to multiple deliveries to the same
3766 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3767 in its include files, and this causes problems building Exim.
3769 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3770 header_syntax) but Exim was just ignoring anything given after a slash.
3771 In particular, this caused confusion with an attempt to use "verify =
3772 reverse_host_lookup/defer_ok". An error is now given when options are
3773 supplied for verify items that do not have them. (Maybe reverse_host_
3774 lookup should have a defer_ok option, but that's a different point.)
3776 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3777 defined by RFC 821) to 2048, because there were problems with some AUTH
3778 commands, and RFC 1869 says the size should be increased for extended
3779 SMTP commands that take arguments.
3781 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3784 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3785 "unknown" error; now it says that the functionality isn't in the binary.
3787 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3788 an address' error message when a string expansion fails (syntax or
3789 whatever). Otherwise the password may appear in the log. Following change
3790 PH/42 below, there is no longer a chance of it appearing in a bounce
3793 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3795 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3796 of its addresses was ever tried. (Bugzilla bug #2.)
3798 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3799 the result incorrectly in the debug output. (It correctly added a newline
3800 to what was transported.)
3802 TF/01 Added $received_time.
3804 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3805 commented out examples of how to interface to a virus scanner and to
3806 SpamAssassin. Also added commented examples of av_scanner and
3807 spamd_address settings.
3809 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3810 and controls are allowed in which ACLs. There were a couple of minor
3811 errors. Some of the entries in the conditions table (which is a table of
3812 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3813 negation of where the condition IS allowed.
3815 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3817 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3818 header file does not have a version number, so I've had to invent a new
3819 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3820 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3821 radiusclient), but it was contributed by a Radius user.
3823 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3824 files or directories, for OpenSSL.
3826 PH/19 When an Exim process that is running as root has to create an Exim log
3827 file, it does so in a subprocess that runs as exim:exim so as to get the
3828 ownership right at creation (otherwise, other Exim processes might see
3829 the file with the wrong ownership). There was no test for failure of this
3830 fork() call, which would lead to the process getting stuck as it waited
3831 for a non-existent subprocess. Forks do occasionally fail when resources
3832 run out. I reviewed all the other calls to fork(); they all seem to check
3835 PH/20 When checking for unexpected SMTP input at connect time (before writing
3836 the banner), Exim was not dealing correctly with a non-positive return
3837 from the read() function. If the client had disconnected by this time,
3838 the result was a log entry for a synchronization error with an empty
3839 string after "input=" when read() returned zero. If read() returned -1
3840 (an event I could not check), uninitialized data bytes were printed.
3841 There were reports of junk text (parts of files, etc) appearing after
3844 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3846 PH/22 Added support for macro redefinition, and (re)definition in between
3847 driver and ACL definitions.
3849 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3850 forgetting to use the resulting value; it was using the unexpanded value.
3852 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3853 hadn't been configured. The fix is from Juergen Kreileder, who
3854 understands it better than I do:
3856 "Here's what I see happening with three configured cyrus_sasl
3857 authenticators configured (plain, login, cram-md5):
3859 On startup auth_cyrus_sasl_init() gets called for each of these.
3860 This means three calls to sasl_listmech() without a specified mech_list.
3861 => SASL tests which mechs of all available mechs actually work
3862 => three warnings about OTP not working
3863 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3865 With the patch, sasl_listmech() also gets called three times. But now
3866 SASL's mech_list option is set to the server_mech specified in the the
3867 authenticator. Or in other words, the answer from sasl_listmech()
3868 gets limited to just the mech you're testing for (which is different
3870 => the return list contains just 'plain' or 'login', 'cram-md5' or
3871 nothing depending on the value of ob->server_mech.
3873 I've just tested the patch: Authentication still works fine,
3874 unavailable mechs specified in the exim configuration are still
3875 caught, and the auth.log warnings about OTP are gone."
3877 PH/25 When debugging is enabled, the contents of the command line are added
3878 to the debugging output, even when log_selector=+arguments is not
3881 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3882 answer is "GNU", and only if the return is "GNU/something" is the answer
3885 PH/27 $acl_verify_message is now set immediately after the failure of a
3886 verification in an ACL, and so is available in subsequent modifiers. In
3887 particular, the message can be preserved by coding like this:
3889 warn !verify = sender
3890 set acl_m0 = $acl_verify_message
3892 Previously, $acl_verify_message was set only while expanding "message"
3893 and "log_message" when a very denied access.
3895 PH/28 Modified OS/os.c-Linux with
3897 -#ifndef OS_LOAD_AVERAGE
3898 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3900 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3901 nomenclature these days.)
3903 PH/29 Installed patch from the Sieve maintainer that adds the options
3904 sieve_useraddress and sieve_subaddress to the redirect router.
3906 PH/30 In these circumstances:
3907 . Two addresses routed to the same list of hosts;
3908 . First host does not offer TLS;
3909 . First host accepts first address;
3910 . First host gives temporary error to second address;
3911 . Second host offers TLS and a TLS session is established;
3912 . Second host accepts second address.
3913 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3914 and peerdn, if requested) that were in fact used only for the second
3917 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3918 attention to any local part prefix or suffix that was matched by the
3919 router that accepted the address. It now behaves in the same way as it
3920 does for delivery: the affixes are removed from the local part unless
3921 rcpt_include_affixes is set on the transport.
3923 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3924 timeout during the DATA phase of an incoming message.
3926 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3927 applied a patch sanctioned by the Sieve maintainer.
3929 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3930 the uid or gid is negative. A case of a negative gid caused this to be
3931 noticed. The fix allows for either to be negative.
3933 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3934 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3937 PH/36 The change PH/12 above was broken. Fixed it.
3939 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3940 the grounds that routing the same address twice would always produce the
3941 same answer. This might have been true once, but it is certainly no
3942 longer true now. Routing a child address may depend on the previous
3943 routing that produced that child. Some complicated redirection strategies
3944 went wrong when messages had multiple recipients, and made Exim's
3945 behaviour dependent on the order in which the addresses were given.
3947 I have moved the duplicate checking until after the routing is complete.
3948 Exim scans the addresses that are assigned to local and remote
3949 transports, and removes any duplicates. This means that more work will be
3950 done, as duplicates will always all be routed, but duplicates are
3951 presumably rare, so I don't expect this is of any significance.
3953 For deliveries to pipes, files, and autoreplies, the duplicate checking
3954 still happens during the routing process, since they are not going to be
3957 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3958 It corrects a timeout issue with spamd. This is Ian's comment: "The
3959 background is that sometimes spamd either never reads data from a
3960 connection it has accepted, or it never writes response data. The exiscan
3961 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3962 blindly assumes that writes won't block so it may never time out."
3964 PH/39 Allow G after quota size as well as K and M.
3966 PH/40 The value set for $authenticated_id in an authenticator may not contain
3967 binary zeroes or newlines because the value is written to log lines and
3968 to spool files. There was no check on this. Now the value is run through
3969 the string_printing() function so that such characters are converted to
3970 printable escape sequences.
3972 PH/41 $message_linecount is a new variable that contains the total number of
3973 lines in the message. Compare $body_linecount, which is the count for the
3976 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3977 bounce and delay warning messages, except in certain special cases, which
3980 (a) An SMTP error message from a remote host;
3981 (b) A message specified in a :fail: redirection;
3982 (c) A message specified in a "fail" command in a system filter;
3983 (d) A message specified in a FAIL return from the queryprogram router;
3984 (e) A message specified by the cannot_route_message router option.
3986 In these cases only, Exim does include the error details in bounce and
3987 warning messages. There are also a few cases where bland messages such
3988 as "unrouteable address" or "local delivery error" are given.
3990 PH/43 $value is now also set for the "else" part of a ${run expansion.
3992 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3993 being worked on, but at least Exim now implements the latest version to
3996 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3997 process to complete was treated as a delivery failure, a timeout while
3998 writing the message to the pipe was logged, but erroneously treated as a
3999 successful delivery. Such timeouts include transport filter timeouts. For
4000 consistency with the overall process timeout, these timeouts are now
4001 treated as errors, giving rise to delivery failures by default. However,
4002 there is now a new Boolean option for the pipe transport called
4003 timeout_defer, which, if set TRUE, converts the failures into defers for
4004 both kinds of timeout. A transport filter timeout is now identified in
4007 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4008 systems where "make" and "gmake" are different, calling "gmake" at top
4009 level broke things. I've arranged for the value of $(MAKE) to be passed
4010 from the Makefile to this script so that it can call the same version of
4014 A note about Exim versions 4.44 and 4.50
4015 ----------------------------------------
4017 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4018 changes of various kinds. As a consequence, a big documentation update was
4019 needed. This delayed the release for rather longer than seemed good, especially
4020 in the light of a couple of (minor) security issues. Therefore, the changes
4021 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4022 release. So 4.44 and 4.50 are in effect two different branches that both start
4025 I have left the 4.50 change log unchanged; it contains all the changes since
4026 4.43. The change log for 4.44 is below; many of its items are identical to
4027 those for 4.50. This seems to be the most sensible way to preserve the
4028 historical information.
4034 1. Minor wording change to the doc/README.SIEVE file.
4036 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4037 computation of the current number of files was incorrect.
4039 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4040 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4043 4. Give more explanation in the error message when the command for a transport
4044 filter fails to execute.
4046 5. There are several places where Exim runs a non-Exim command in a
4047 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4048 was being done only for the command run by the queryprogram router. It is
4049 now done for all such subprocesses. The other cases are: ${run, transport
4050 filters, and the commands run by the lmtp and pipe transports.
4052 6. Added CONFIGURE_GROUP build-time option.
4054 7. Some older OS have a limit of 256 on the maximum number of file
4055 descriptors. Exim was using setrlimit() to set 1000 as a large value
4056 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4057 systems. I've change it so that if it can't get 1000, it tries for 256.
4059 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4060 was an oversight, and furthermore, ever since the addition of extra
4061 controls (e.g. 4.43/32), the checks on when to allow different forms of
4062 "control" were broken. There should now be diagnostics for all cases when a
4063 control that does not make sense is encountered.
4065 9. Added the /retain_sender option to "control=submission".
4067 10. $recipients is now available in the predata ACL (oversight).
4069 11. Tidy the search cache before the fork to do a delivery from a message
4070 received from the command line. Otherwise the child will trigger a lookup
4071 failure and thereby defer the delivery if it tries to use (for example) a
4072 cached ldap connection that the parent has called unbind on.
4074 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4075 of $address_data from the recipient verification was clobbered by the
4076 sender verification.
4078 13. The value of address_data from a sender verification is now available in
4079 $sender_address_data in subsequent conditions in the ACL statement.
4081 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4083 15. Added a new option "connect=<time>" to callout options, to set a different
4086 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4087 was its contents. (It was OK if the option was not defined at all.)
4089 17. A "Completed" log line is now written for messages that are removed from
4090 the spool by the -Mrm option.
4092 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4093 information about exactly what failed.
4095 19. Added -dd to debug only the daemon process.
4097 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4098 handles timeouts, both on the server side and network timeouts. Renamed the
4099 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4101 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4102 It is now set to "smtps".
4104 22. $host_address is now set to the target address during the checking of
4105 ignore_target_hosts.
4107 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4108 being passed; this would have caused $sender_host_name to have been used if
4109 matching the list had actually called for a host name (not very likely,
4110 since this list is usually IP addresses). A host name is now passed as
4113 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4114 code that specifies a non-restarting handler (typically sigaction() in
4115 modern systems) in an attempt to fix a rare and obscure crash bug.
4117 25. Narrowed the window for a race in the daemon that could cause it to ignore
4118 SIGCHLD signals. This is not a major problem, because they are used only to
4119 wake it up if nothing else does.
4121 26. A malformed maildirsize file could cause Exim to calculate negative values
4122 for the mailbox size or file count. Odd effects could occur as a result.
4123 The maildirsize information is now recalculated if the size or filecount
4126 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4127 support for a long time. Removed HAVE_SYS_VFS_H.
4129 28. Installed the latest version of exipick from John Jetmore.
4131 29. In an address list, if the pattern was not a regular expression, an empty
4132 subject address (from a bounce message) matched only if the pattern was an
4133 empty string. Non-empty patterns were not even tested. This was the wrong
4134 because it is perfectly reasonable to use an empty address as part of a
4135 database query. An empty address is now tested by patterns that are
4136 lookups. However, all the other forms of pattern expect the subject to
4137 contain a local part and a domain, and therefore, for them, an empty
4138 address still always fails if the pattern is not itself empty.
4140 30. Exim went into a mad DNS loop when attempting to do a callout where the
4141 host was specified on an smtp transport, and looking it up yielded more
4142 than one IP address.
4144 31. Re-factored the code for checking spool and log partition space into a
4145 function that finds that data and another that does the check. The former
4146 is then used to implement four new variables: $spool_space, $log_space,
4147 $spool_inodes, and $log_inodes.
4149 32. The RFC2047 encoding function was originally intended for short strings
4150 such as real names; it was not keeping to the 75-character limit for
4151 encoded words that the RFC imposes. It now respects the limit, and
4152 generates multiple encoded words if necessary. To be on the safe side, I
4153 have increased the buffer size for the ${rfc2047: expansion operator from
4156 33. It is now permitted to omit both strings after an "if" condition; if the
4157 condition is true, the result is "true". As before, when the second string
4158 is omitted, a false condition yields an empty string. This makes it less
4159 cumbersome to write custom ACL and router conditions.
4161 34. Failure to deliver a bounce message always caused it to be frozen, even if
4162 there was an errors_to setting on the router. The errors_to setting is now
4165 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4166 canonical form (fully expanded) before being placed in
4167 $sender_host_address.
4169 36. The table in the code that translates DNS record types into text (T_A to
4170 "A" for instance) was missing entries for NS and CNAME. It is just possible
4171 that this could have caused confusion if both these types were looked up
4172 for the same domain, because the text type is used as part of Exim's
4173 per-process caching. But the chance of anyone hitting this buglet seems
4176 37. The dnsdb lookup has been extended in a number of ways.
4178 (1) There is a new type, "zns", which walks up the domain tree until it
4179 finds some nameserver records. It should be used with care.
4181 (2) There is a new type, "mxh", which is like "mx" except that it returns
4182 just the host names, not the priorities.
4184 (3) It is now possible to give a list of domains (or IP addresses) to be
4185 looked up. The behaviour when one of the lookups defers can be
4186 controlled by a keyword.
4188 (4) It is now possible to specify the separator character for use when
4189 multiple records are returned.
4191 38. The dnslists ACL condition has been extended: it is now possible to supply
4192 a list of IP addresses and/or domains to be looked up in a particular DNS
4195 39. Added log_selector=+queue_time_overall.
4197 40. When running the queue in the test harness, wait just a tad after forking a
4198 delivery process, to get repeatability of debugging output.
4200 41. Include certificate and key file names in error message when GnuTLS fails
4201 to set them up, because the GnuTLS error message doesn't include the name
4202 of the failing file when there is a problem reading it.
4204 42. Allow both -bf and -bF in the same test run.
4206 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4208 44. The "Exiscan patch" is now merged into the mainline Exim source.
4210 45. Sometimes the final signoff response after QUIT could fail to get
4211 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4212 < 0 before doing a fflush(). This bug looks as though it goes back to the
4213 introduction of TLS in release 3.20, but "sometimes" must have been rare
4214 because the tests only now provoked it.
4216 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4217 (this can affect the format of dates).
4219 47. exim_tidydb, when checking for the continued existence of a message for
4220 which it has found a message-specific retry record, was not finding
4221 messages that were in split spool directories. Consequently, it was
4222 deleting retry records that should have stayed in existence.
4224 48. Steve fixed some bugs in eximstats.
4226 49. The SPA authentication driver was not abandoning authentication and moving
4227 on to the next authenticator when an expansion was forced to fail,
4228 contradicting the general specification for all authenticators. Instead it
4229 was generating a temporary error. It now behaves as specified.
4231 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4232 (the order specifies the preference for clients). The order is now AES256,
4233 AES128, 3DES, ARCFOUR128.
4235 51. Small patch to Sieve code - explicitly set From: when generating an
4238 52. Exim crashed if a remote delivery caused a very long error message to be
4239 recorded - for instance if somebody sent an entire SpamAssassin report back
4240 as a large number of 550 error lines. This bug was coincidentally fixed by
4241 increasing the size of one of Exim's internal buffers (big_buffer) that
4242 happened as part of the Exiscan merge. However, to be on the safe side, I
4243 have made the code more robust (and fixed the comments that describe what
4246 53. Now that there can be additional text after "Completed" in log lines (if
4247 the queue_time_overall log selector is set), a one-byte patch to exigrep
4248 was needed to allow it to recognize "Completed" as not the last thing in
4251 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4252 patch that reportedly fixes this has been added. I am not expert enough to
4253 create a test for it. This is what the patch creator wrote:
4255 "I found a little strange behaviour of ldap code when working with
4256 Windows 2003 AD Domain, where users was placed in more than one
4257 Organization Units. When I tried to give exim partial DN, the exit code
4258 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4259 But simultaneously result of request was absolutely normal ldap result,
4260 so I produce this patch..."
4262 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4263 REFERENCE, so I have modified the code to exclude the patch when that macro
4266 55. Some experimental protocols are using DNS PTR records for new purposes. The
4267 keys for these records are domain names, not reversed IP addresses. The
4268 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4269 leaves it alone. Component reversal etc. now happens only for IP addresses.
4272 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4274 57. Double the size of the debug message buffer (to 2048) so that more of very
4275 long debug lines gets shown.
4277 58. The exicyclog utility now does better if the number of log files to keep
4278 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4280 59. Two changes related to the smtp_active_hostname option:
4282 (1) $smtp_active_hostname is now available as a variable.
4283 (2) The default for smtp_banner uses $smtp_active_hostname instead
4284 of $primary_hostname.
4286 60. The host_aton() function is supposed to be passed a string that is known
4287 to be a valid IP address. However, in the case of IPv6 addresses, it was
4288 not checking this. This is a hostage to fortune. Exim now panics and dies
4289 if the condition is not met. A case was found where this could be provoked
4290 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4291 components; fortuitously, this particular loophole had already been fixed
4292 by change 4.50/55 above.
4294 If there are any other similar loopholes, the new check in host_aton()
4295 itself should stop them being exploited. The report I received stated that
4296 data on the command line could provoke the exploit when Exim was running as
4297 exim, but did not say which command line option was involved. All I could
4298 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4299 running as the user.
4302 61. There was a buffer overflow vulnerability in the SPA authentication code
4303 (which came originally from the Samba project). I have added a test to the
4304 spa_base64_to_bits() function which I hope fixes it.
4307 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4308 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4310 63. The daemon start-up calls getloadavg() while still root for those OS that
4311 need the first call to be done as root, but it missed one case: when
4312 deliver_queue_load_max is set with deliver_drop_privilege. This is
4313 necessary for the benefit of the queue runner, because there is no re-exec
4314 when deliver_drop_privilege is set.
4316 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4317 This has been fixed.
4319 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4320 was in use, was not putting the data itself into the right store pool;
4321 consequently, it could be overwritten for a subsequent message in the same
4322 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4325 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4327 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4328 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4329 the function were treating the return as a boolean value, which happened to
4330 work because 0=false and not-0=true, but is not correct code.
4332 68. The host_aton() function was not handling scoped IPv6 addresses (those
4333 with, for example, "%eth0" on the end) correctly.
4335 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4336 negated items (that is, ~something) in unsigned ints. Some compilers
4337 apparently mutter when there is no cast.
4339 70. If an address verification called from an ACL failed, and did not produce a
4340 user-specific message (i.e. there was only a "system" message), nothing was
4341 put in $acl_verify_message. In this situation, it now puts the system
4344 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4345 session; change 4.31/43 added the unwanted input to the log line - except
4346 that it did not do this in the start of session case. It now does.
4348 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4349 This is wrong and can cause the other end to generate a synchronization
4350 error if it is another Exim or anything else that does the synchronization
4351 check. A QUIT command is no longer sent after a timeout.
4353 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4354 during host lookups.
4356 74. The defer_ok option of callout verification was not working if it was used
4357 when verifying addresses in header lines, that is, for this case:
4359 verify = header_sender/callout=defer_ok
4361 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4362 those file descriptors could be used for SMTP connections. If anything
4363 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4364 could be sent to the SMTP client, causing chaos. The daemon now opens
4365 stdin, stdout, and stderr to /dev/null when it puts itself into the
4368 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4369 log by default. The user can override this with suitable Perl magic.
4371 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4372 the log message when discard triggers, was not working for the DATA ACL or
4373 for the non-SMTP ACL.
4375 78. Error message wording change in sieve.c.
4377 79. If smtp_accept_max_per_host was set, the number of connections could be
4378 restricted to fewer than expected, because the daemon was trying to set up
4379 a new connection before checking whether the processes handling previous
4380 connections had finished. The check for completed processes is now done
4381 earlier. On busy systems, this bug wouldn't be noticed because something
4382 else would have woken the daemon, and it would have reaped the completed
4385 80. If a message was submitted locally by a user whose login name contained one
4386 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4387 It caused a spool format error. I have fixed the spool reading code. A
4388 related problem was that the "from" clause in the Received: line became
4389 illegal because of the space(s). It is now covered by ${quote_local_part.
4391 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4394 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4396 83. Patch from Sieve maintainer to fix unterminated string problem in
4397 "vacation" handling.
4399 84. Some minor changes to the Linux configuration files to help with other
4400 OS variants using glibc.
4402 85. One more patch for Sieve to update vacation handling to latest spec.
4405 ----------------------------------------------------
4406 See the note above about the 4.44 and 4.50 releases.
4407 ----------------------------------------------------
4413 1. Change 4.43/35 introduced a bug that caused file counts to be
4414 incorrectly computed when quota_filecount was set in an appendfile
4417 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4418 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4421 3. Give more explanation in the error message when the command for a transport
4422 filter fails to execute.
4424 4. There are several places where Exim runs a non-Exim command in a
4425 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4426 was being done only for the command run by the queryprogram router. It is
4427 now done for all such subprocesses. The other cases are: ${run, transport
4428 filters, and the commands run by the lmtp and pipe transports.
4430 5. Some older OS have a limit of 256 on the maximum number of file
4431 descriptors. Exim was using setrlimit() to set 1000 as a large value
4432 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4433 systems. I've change it so that if it can't get 1000, it tries for 256.
4435 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4436 was an oversight, and furthermore, ever since the addition of extra
4437 controls (e.g. 4.43/32), the checks on when to allow different forms of
4438 "control" were broken. There should now be diagnostics for all cases when a
4439 control that does not make sense is encountered.
4441 7. $recipients is now available in the predata ACL (oversight).
4443 8. Tidy the search cache before the fork to do a delivery from a message
4444 received from the command line. Otherwise the child will trigger a lookup
4445 failure and thereby defer the delivery if it tries to use (for example) a
4446 cached ldap connection that the parent has called unbind on.
4448 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4449 of $address_data from the recipient verification was clobbered by the
4450 sender verification.
4452 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4453 was its contents. (It was OK if the option was not defined at all.)
4455 11. A "Completed" log line is now written for messages that are removed from
4456 the spool by the -Mrm option.
4458 12. $host_address is now set to the target address during the checking of
4459 ignore_target_hosts.
4461 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4462 being passed; this would have caused $sender_host_name to have been used if
4463 matching the list had actually called for a host name (not very likely,
4464 since this list is usually IP addresses). A host name is now passed as
4467 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4468 code that specifies a non-restarting handler (typically sigaction() in
4469 modern systems) in an attempt to fix a rare and obscure crash bug.
4471 15. Narrowed the window for a race in the daemon that could cause it to ignore
4472 SIGCHLD signals. This is not a major problem, because they are used only to
4473 wake it up if nothing else does.
4475 16. A malformed maildirsize file could cause Exim to calculate negative values
4476 for the mailbox size or file count. Odd effects could occur as a result.
4477 The maildirsize information is now recalculated if the size or filecount
4480 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4481 support for a long time. Removed HAVE_SYS_VFS_H.
4483 18. Updated exipick to current release from John Jetmore.
4485 19. Allow an empty sender to be matched against a lookup in an address list.
4486 Previously the only cases considered were a regular expression, or an
4489 20. Exim went into a mad DNS lookup loop when doing a callout where the
4490 host was specified on the transport, if the DNS lookup yielded more than
4493 21. The RFC2047 encoding function was originally intended for short strings
4494 such as real names; it was not keeping to the 75-character limit for
4495 encoded words that the RFC imposes. It now respects the limit, and
4496 generates multiple encoded words if necessary. To be on the safe side, I
4497 have increased the buffer size for the ${rfc2047: expansion operator from
4500 22. Failure to deliver a bounce message always caused it to be frozen, even if
4501 there was an errors_to setting on the router. The errors_to setting is now
4504 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4505 canonical form (fully expanded) before being placed in
4506 $sender_host_address.
4508 24. Updated eximstats to version 1.33
4510 25. Include certificate and key file names in error message when GnuTLS fails
4511 to set them up, because the GnuTLS error message doesn't include the name
4512 of the failing file when there is a problem reading it.
4514 26. Expand error message when OpenSSL has problems setting up cert/key files.
4517 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4518 (this can affect the format of dates).
4520 28. exim_tidydb, when checking for the continued existence of a message for
4521 which it has found a message-specific retry record, was not finding
4522 messages that were in split spool directories. Consequently, it was
4523 deleting retry records that should have stayed in existence.
4525 29. eximstats updated to version 1.35
4526 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4527 1.35 - bugfix such that pie charts by volume are generated correctly
4529 30. The SPA authentication driver was not abandoning authentication and moving
4530 on to the next authenticator when an expansion was forced to fail,
4531 contradicting the general specification for all authenticators. Instead it
4532 was generating a temporary error. It now behaves as specified.
4534 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4535 (the order specifies the preference for clients). The order is now AES256,
4536 AES128, 3DES, ARCFOUR128.
4538 31. Small patch to Sieve code - explicitly set From: when generating an
4541 32. Exim crashed if a remote delivery caused a very long error message to be
4542 recorded - for instance if somebody sent an entire SpamAssassin report back
4543 as a large number of 550 error lines. This bug was coincidentally fixed by
4544 increasing the size of one of Exim's internal buffers (big_buffer) that
4545 happened as part of the Exiscan merge. However, to be on the safe side, I
4546 have made the code more robust (and fixed the comments that describe what
4549 33. Some experimental protocols are using DNS PTR records for new purposes. The
4550 keys for these records are domain names, not reversed IP addresses. The
4551 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4552 leaves it alone. Component reversal etc. now happens only for IP addresses.
4555 34. The host_aton() function is supposed to be passed a string that is known
4556 to be a valid IP address. However, in the case of IPv6 addresses, it was
4557 not checking this. This is a hostage to fortune. Exim now panics and dies
4558 if the condition is not met. A case was found where this could be provoked
4559 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4560 components; fortuitously, this particular loophole had already been fixed
4561 by change 4.50/55 or 4.44/33 above.
4563 If there are any other similar loopholes, the new check in host_aton()
4564 itself should stop them being exploited. The report I received stated that
4565 data on the command line could provoke the exploit when Exim was running as
4566 exim, but did not say which command line option was involved. All I could
4567 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4568 running as the user.
4571 35. There was a buffer overflow vulnerability in the SPA authentication code
4572 (which came originally from the Samba project). I have added a test to the
4573 spa_base64_to_bits() function which I hope fixes it.
4576 36. The daemon start-up calls getloadavg() while still root for those OS that
4577 need the first call to be done as root, but it missed one case: when
4578 deliver_queue_load_max is set with deliver_drop_privilege. This is
4579 necessary for the benefit of the queue runner, because there is no re-exec
4580 when deliver_drop_privilege is set.
4582 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4583 was in use, was not putting the data itself into the right store pool;
4584 consequently, it could be overwritten for a subsequent message in the same
4585 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4588 38. Sometimes the final signoff response after QUIT could fail to get
4589 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4590 < 0 before doing a fflush(). This bug looks as though it goes back to the
4591 introduction of TLS in release 3.20, but "sometimes" must have been rare
4592 because the tests only now provoked it.
4598 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4599 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4600 Now it returns an integer. A number of calls were still expecting a T/F
4601 return. Fortuitously, in all cases, the tests worked in OK situations,
4602 which is the norm. However, things would have gone wrong on any write
4603 failures on the smtp file descriptor. This function is used when sending
4604 messages over SMTP and also when doing verify callouts.
4606 2. When Exim is called to do synchronous delivery of a locally submitted
4607 message (the -odf or -odi options), it no longer closes stderr before doing
4610 3. Implemented the mua_wrapper option.
4612 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4614 5. Implemented the functions header_remove(), header_testname(),
4615 header_add_at_position(), and receive_remove_recipient(), and exported them
4618 6. If an ACL "warn" statement specified the addition of headers, Exim already
4619 inserted X-ACL-Warn: at the start if there was no header name. However, it
4620 was not making this test for the second and subsequent header lines if
4621 there were newlines in the string. This meant that an invalid header could
4622 be inserted if Exim was badly configured.
4624 7. Allow an ACL "warn" statement to add header lines at the start or after all
4625 the Received: headers, as well as at the end.
4627 8. Added the rcpt_4xx retry error code.
4629 9. Added postmaster_mailfrom=xxx to callout verification option.
4631 10. Added mailfrom=xxxx to the callout verification option, for verify=
4634 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4635 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4637 12. Inserted some casts to stop certain compilers warning when using pointer
4638 differences as field lengths or precisions in printf-type calls (mostly
4639 affecting debugging statements).
4641 13. Added optional readline() support for -be (dynamically loaded).
4643 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4644 same clock tick as a message's arrival, so that its received time was the
4645 same as the "first fail" time on the retry record, and that message
4646 remained on the queue past the ultimate address timeout, every queue runner
4647 would try a delivery (because it was past the ultimate address timeout) but
4648 after another failure, the ultimate address timeout, which should have then
4649 bounced the address, did not kick in. This was a "< instead of <=" error;
4650 in most cases the first failure would have been in the next clock tick
4651 after the received time, and all would be well.
4653 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4654 being recognized when the domain list was tested by the match_domain
4655 condition in an expansion string.
4657 16. Added the ${str2b64: operator.
4659 17. Exim was always calling setrlimit() to set a large limit for the number of
4660 processes, without checking whether the existing limit was already
4661 adequate. (It did check for the limit on file descriptors.) Furthermore,
4662 errors from getrlimit() and setrlimit() were being ignored. Now they are
4663 logged to the main and panic logs, but Exim does carry on, to try to do its
4664 job under whatever limits there are.
4666 18. Imported PCRE 5.0.
4668 19. Trivial typo in log message " temporarily refused connection" (the leading
4671 20. If the log selector return_path_on_delivery was set and an address was
4672 redirected to /dev/null, the delivery process crashed because it assumed
4673 that a return path would always be set for a "successful" delivery. In this
4674 case, the whole delivery is bypassed as an optimization, and therefore no
4677 21. Internal re-arrangement: the function for sending a challenge and reading
4678 a response while authentication was assuming a zero-terminated challenge
4679 string. It's now changed to take a pointer and a length, to allow for
4680 binary data in such strings.
4682 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4684 23. Exim was not respecting finduser_retries when seeking the login of the
4685 uid under which it was called; it was always trying 10 times. (The default
4686 setting of finduser_retries is zero.) Also, it was sleeping after the final
4687 failure, which is pointless.
4689 24. Implemented tls_on_connect_ports.
4691 25. Implemented acl_smtp_predata.
4693 26. If the domain in control=submission is set empty, Exim assumes that the
4694 authenticated id is a complete email address when it generates From: or
4695 Sender: header lines.
4697 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4698 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4699 chown and chgrp in /bin and hostname in /usr/bin.
4701 28. Exim was keeping the "process log" file open after each use, just as it
4702 does for the main log. This opens the possibility of it remaining open for
4703 long periods when the USR1 signal hits a daemon. Occasional processlog
4704 errors were reported, that could have been caused by this. Anyway, it seems
4705 much more sensible not to leave this file open at all, so that is what now
4708 29. The long-running daemon process does not normally write to the log once it
4709 has entered its main loop, and it closes the log before doing so. This is
4710 so that log files can straightforwardly be renamed and moved. However,
4711 there are a couple of unusual error situations where the daemon does write
4712 log entries, and I had neglected to close the log afterwards.
4714 30. The text of an SMTP error response that was received during a remote
4715 delivery was being truncated at 512 bytes. This is too short for some of
4716 the long messages that one sometimes sees. I've increased the limit to
4719 31. It is now possible to make retry rules that apply only when a message has a
4720 specific sender, in particular, an empty sender.
4722 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4723 it possible to be selective about when SMTP synchronization is enforced.
4725 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4727 32. Implemented hosts_connection_nolog.
4729 33. Added an ACL for QUIT.
4731 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4734 35. Added mailbox_size and mailbox_filecount to appendfile.
4736 36. Added control = no_multiline_responses to ACLs.
4738 37. There was a bug in the logic of the code that waits for the clock to tick
4739 in the case where the clock went backwards by a substantial amount such
4740 that the microsecond fraction of "now" was more than the microsecond
4741 fraction of "then" (but the whole seconds number was less).
4743 38. Added support for the libradius Radius client library this is found on
4744 FreeBSD (previously only the radiusclient library was supported).
4750 1. When certain lookups returned multiple values in the form name=value, the
4751 quoting of the values was not always being done properly. Specifically:
4752 (a) If the value started with a double quote, but contained no whitespace,
4754 (b) If the value contained whitespace other than a space character (i.e.
4755 tabs or newlines or carriage returns) it was not quoted.
4756 This fix has been applied to the mysql and pgsql lookups by writing a
4757 separate quoting function and calling it from the lookup code. The fix
4758 should probably also be applied to nisplus, ibase and oracle lookups, but
4759 since I cannot test any of those, I have not disturbed their existing code.
4761 2. A hit in the callout cache for a specific address caused a log line with no
4762 reason for rejecting RCPT. Now it says "Previous (cached) callout
4763 verification failure".
4765 3. There was an off-by-one bug in the queryprogram router. An over-long
4766 return line was truncated at 256 instead of 255 characters, thereby
4767 overflowing its buffer with the terminating zero. As well as fixing this, I
4768 have increased the buffer size to 1024 (and made a note to document this).
4770 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4771 when Exim is waiting for an SMTP response from a remote server, Exim
4772 restarts its select() call on the socket, thereby resetting its timeout.
4773 This is not a problem when such interrupts are rare. Somebody set up a cron
4774 job to run exiwhat every 2 minutes, which is less than the normal select()
4775 timeout (5 or 10 minutes). This meant that the select() timeout never
4776 kicked in because it was always reset. I have fixed this by comparing the
4777 time when an interrupt arrives with the time at the start of the first call
4778 to select(). If more time than the timeout has elapsed, the interrupt is
4779 treated as a timeout.
4781 5. Some internal re-factoring in preparation for the addition of Sieve
4782 extensions (by MH). In particular, the "personal" test is moved to a
4783 separate function, and given an option for scanning Cc: and Bcc: (which is
4784 not set for Exim filters).
4786 6. When Exim created an email address using the login of the caller as the
4787 local part (e.g. when creating a From: or Sender: header line), it was not
4788 quoting the local part when it contained special characters such as @.
4790 7. Installed new OpenBSD configuration files.
4792 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4793 try to make them clearer.
4795 9. Callout options, other than the timeout value, were being ignored when
4796 verifying sender addresses in header lines. For example, when using
4798 verify = header_sender/callout=no_cache
4800 the cache was (incorrectly) being used.
4802 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4803 only the Cygwin environment.
4805 11. When return_path_on_delivery was set as a log selector, if different remote
4806 addresses in the same message used different return paths and parallel
4807 remote delivery occurred, the wrong values would sometimes be logged.
4808 (Whenever a remote delivery process finished, the return path value from
4809 the most recently started remote delivery process was logged.)
4811 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4812 lines when AUTH and/or TLS are in use. This is the "received protocol"
4813 field. Exim used to use "asmtp" for authenticated SMTP, without any
4814 indication (in the protocol name) for TLS use. Now it follows the RFC and
4815 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4816 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4817 names appear in log lines as well as in Received: header lines.
4819 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4820 extensions, and comparison tests, and to fix some bugs.
4822 14. Changes to the "personal" filter test:
4824 (1) The test was buggy in that it was just doing the equivalent of
4825 "contains" tests on header lines. For example, if a user's address was
4826 anne@some.where, the "personal" test would incorrectly be true for
4828 To: susanne@some.where
4830 This test is now done by extracting each address from the header in turn,
4831 and checking the entire address. Other tests that are part of "personal"
4832 are now done using regular expressions (for example, to check local parts
4833 of addresses in From: header lines).
4835 (2) The list of non-personal local parts in From: addresses has been
4836 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4837 taken from the Sieve specification recommendations.
4839 (3) If the message contains any header line starting with "List-" it is
4840 treated as non-personal.
4842 (4) The test for "circular" in the Subject: header line has been removed
4843 because it now seems ill-conceived.
4845 15. Minor typos in src/EDITME comments corrected.
4847 16. Installed latest exipick from John Jetmore.
4849 17. If headers_add on a router specified a text string that was too long for
4850 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4851 of string_sprintf() is now avoided.
4853 18. $message_body_size was not set (it was always zero) when running the DATA
4854 ACL and the local_scan() function.
4856 19. For the "mail" command in an Exim filter, no default was being set for
4857 the once_repeat time, causing a random time value to be used if "once" was
4858 specified. (If the value happened to be <= 0, no repeat happened.) The
4859 default is now 0s, meaning "never repeat". The "vacation" command was OK
4860 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4861 (I found it when inspecting the code).
4863 20. There is now an overall timeout for performing a callout verification. It
4864 defaults to 4 times the callout timeout, which applies to individual SMTP
4865 commands during the callout. The overall timeout applies when there is more
4866 than one host that can be tried. The timeout is checked before trying the
4867 next host. This prevents very long delays if there are a large number of
4868 hosts and all are timing out (e.g. when the network connections are timing
4869 out). The value of the overall timeout can be changed by specifying an
4870 additional sub-option for "callout", called "maxwait". For example:
4872 verify = sender/callout=5s,maxwait=20s
4874 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4875 to the end before writing, but this should make it even safer).
4877 22. Exim was forgetting that it had advertised PIPELINING for the second and
4878 subsequent messages on an SMTP connection. It was also not resetting its
4879 memory on STARTTLS and an internal HELO.
4881 23. When Exim logs an SMTP synchronization error within a session, it now
4882 records whether PIPELINING has been advertised or not.
4884 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4885 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4886 rather than long int.
4888 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4890 26. Added the never_mail option to autoreply.
4896 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4897 crash if the getsockname() call failed; this can happen if a connection is
4898 closed very soon after it is established. The problem was simply in the
4899 order in which certain operations were done, causing Exim to try to write
4900 to the SMTP stream before it had set up the file descriptor. The bug has
4901 been fixed by making things happen in the correct order.
4907 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4908 before the connection was closed, thus losing the rejection response.
4910 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4911 some early Solaris releases, but causes trouble in current releases where
4912 socklen_t is defined.
4914 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4917 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4920 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4921 that releasing the top of it at the end releases what was used for sub-
4922 expansions (unless the block got too big). However, discard this block if
4923 the first thing is a variable or header, so that we can use its block when
4924 it is dynamic (useful for very large $message_headers, for example).
4926 6. Lookups now cache *every* query, not just the most recent. A new, separate
4927 store pool is used for this. It can be recovered when all lookup caches are
4928 flushed. Lookups now release memory at the end of their result strings.
4929 This has involved some general refactoring of the lookup sources.
4931 7. Some code has been added to the store_xxx() functions to reduce the amount
4932 of flapping under certain conditions.
4934 8. log_incoming_interface used to affect only the <= reception log lines. Now
4935 it causes the local interface and port to be added to several more SMTP log
4936 lines, for example "SMTP connection from", and rejection lines.
4938 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4940 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4942 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4943 could be overwritten at the end of the current message (or the start of a
4944 new message if it was set in a HELO ACL). The value is now preserved for
4945 the duration of the SMTP connection.
4947 12. If a transport had a headers_rewrite setting, and a matching header line
4948 contained an unqualified address, that address was qualified, even if it
4949 did not match any rewriting rules. The underlying bug was that the values
4950 of the flags that permit the existence of unqualified sender and recipient
4951 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4952 non-local messages, and by -bnq for local messages) were not being
4953 preserved with the message after it was received.
4955 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4956 "next input=" as part of the text comprising the host identity instead of
4957 the correct text. The code was using the same buffer for two different
4958 strings. However, depending on which order the printing function evaluated
4959 its arguments, the bug did not always show up. Under Linux, for example, my
4960 test suite worked just fine.
4962 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4963 doesn't work with some older versions of Perl. It has been changed to "my",
4964 which in any case is probably the better facility to use.
4966 15. A really picky compiler found some instances of statements for creating
4967 error messages that either had too many or two few arguments for the format
4970 16. The size of the buffer for calls to the DNS resolver has been increased
4971 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4972 for addresses that have a lot of PTR records. This alleviates a problem; it
4973 does not fully solve it.
4975 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4976 buffer now truncates the list and logs the incident, which is the same
4977 action as happens when Exim is looking up a host name and its aliases.
4978 Previously in this situation something unpredictable would happen;
4979 sometimes it was "internal error: store_reset failed".
4981 18. If a server dropped the connection unexpectedly when an Exim client was
4982 using GnuTLS and trying to read a response, the client delivery process
4983 crashed while trying to generate an error log message.
4985 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4986 string, for example:
4988 warn message = H1: something\nH2: something
4990 the text was added as a single header line from Exim's point of view
4991 though it ended up OK in the delivered message. However, searching for the
4992 second and subsequent header lines using $h_h2: did not work. This has been
4993 fixed. Similarly, if a system filter added multiple headers in this way,
4994 the routers could not see them.
4996 20. Expanded the error message when iplsearch is called with an invalid key to
4997 suggest using net-iplsearch in a host list.
4999 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5000 ACLs are no longer actually imposed (and a message to that effect is
5003 22. If a "gecos" field in a passwd entry contained escaped characters, in
5004 particular, if it contained a \" sequence, Exim got it wrong when building
5005 a From: or a Sender: header from that name. A second bug also caused
5006 incorrect handling when an unquoted " was present following a character
5007 that needed quoting.
5009 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5010 was not being matched caselessly.
5012 24. Arranged for all hyphens in the exim.8 source to be escaped with
5015 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5016 properly. Recipient callout cache records were still being keyed to include
5017 the sender, even when use_sender was set false. This led to far more
5018 callouts that were necessary. The sender is no longer included in the key
5019 when use_sender is false.
5021 26. Added "control = submission" modifier to ACLs.
5023 27. Added the ${base62d: operator to decode base 62 numbers.
5025 28. dnsdb lookups can now access SRV records.
5027 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5028 the configuration file.
5030 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5031 (-v) mode. This makes the output for a verbose queue run more intelligible.
5033 31. Added a use_postmaster feature to recipient callouts.
5035 32. Added the $body_zerocount variable, containing the number of binary zero
5036 bytes in the message body.
5038 33. The time of last modification of the "new" subdirectory is now used as the
5039 "mailbox time last read" when there is a quota error for a maildir
5042 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5044 35. Added +ignore_unknown as a special item in host lists.
5046 36. Code for decoding IPv6 addresses in host lists is now included, even if
5047 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5048 address was recognized as an IP address, but was then not correctly decoded
5049 into binary, causing unexpected and incorrect effects when compared with
5056 1. Very minor rewording of debugging text in manualroute to say "list of
5057 hosts" instead of "hostlist".
5059 2. If verify=header_syntax was set, and a header line with an unqualified
5060 address (no domain) and a large number of spaces between the end of the
5061 name and the colon was received, the reception process suffered a buffer
5062 overflow, and (when I tested it) crashed. This was caused by some obsolete
5063 code that should have been removed. The fix is to remove it!
5065 3. When running in the test harness, delay a bit after writing a bounce
5066 message to get a bit more predictability in the log output.
5068 4. Added a call to search_tidyup() just before forking a reception process. In
5069 theory, someone could use a lookup in the expansion of smtp_accept_max_
5070 per_host which, without the tidyup, could leave open a database connection.
5072 5. Added the variables $recipient_data and $sender_data which get set from a
5073 lookup success in an ACL "recipients" or "senders" condition, or a router
5074 "senders" option, similar to $domain_data and $local_part_data.
5076 6. Moved the writing of debug_print from before to after the "senders" test
5079 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5080 problems for message scanning, either using a data ACL, or using
5081 local_scan() because the Received: header was not generated till after they
5082 were called (in order to set the time as the time of reception completion).
5083 I have revised the way this works. The header is now generated after the
5084 body is received, but before the ACL or local_scan() are called. After they
5085 are run, the timestamp in the header is updated.
5091 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5092 before starting a queue runner without re-exec. This happened only when
5093 deliver_drop_privilege was set or when the Exim user was set to root. The
5094 effect of the bug was that timeouts during subsequent deliveries caused
5095 crashes instead of being properly handled. The handler is now left at its
5096 default (and expected) setting.
5098 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5099 message, again when deliver_drop_privilege is set or Exim is run as root.
5100 The bug described in (1) was not present in this case, but the tidying up
5101 of the other signals was missing. I have made the two cases consistent.
5103 3. The ignore_target_hosts setting on a manualroute router was being ignored
5104 for hosts that were looked up using the /MX notation.
5106 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5109 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5110 operated on the sender address. After changing the $sender_address to <>
5111 for the sender address verify, Exim was re-instated it as the original
5112 (before rewriting) address, but remembering that it had rewritten it, so it
5113 wasn't rewriting it again. This bug also had the effect of breaking the
5114 sender address verification caching when the sender address was rewritten.
5116 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5117 This has been changed so that if the ip literal address matches
5118 ignore_target_hosts, the router declines.
5120 7. Added expansion conditions match_domain, match_address, and match_local_
5121 part (NOT match_host).
5123 8. The placeholder for the Received: header didn't have a length field set.
5125 9. Added code to Exim itself and to exim_lock to test for a specific race
5126 condition that could lead to file corruption when using MBX delivery. The
5127 issue is with the lockfile that is created in /tmp. If this file is removed
5128 after a process has opened it but before that process has acquired a lock,
5129 there is the potential for a second process to recreate the file and also
5130 acquire a lock. This could lead to two Exim processes writing to the file
5131 at the same time. The added code performs the same test as UW imapd; it
5132 checks after acquiring the lock that its file descriptor still refers to
5133 the same named file.
5135 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5136 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5139 11. Added the smtp_active_hostname option. If used, this will typically be made
5140 to depend on the incoming interface address. Because $interface_address is
5141 not set up until the daemon has forked a reception process, error responses
5142 that can happen earlier (such as "too many connections") no longer contain
5145 12. If an expansion in a condition on a "warn" statement fails because a lookup
5146 defers, the "warn" statement is abandoned, and the next ACL statement is
5147 processed. Previously this caused the whole ACL to be aborted.
5149 13. Added the iplsearch lookup type.
5151 14. Added ident_timeout as a log selector.
5153 15. Added tls_certificate_verified as a log selector.
5155 16. Added a global option tls_require_ciphers (compare the smtp transport
5156 option of the same name). This controls incoming TLS connections.
5158 17. I finally figured out how to make tls_require_ciphers do a similar thing
5159 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5160 before starting the TLS session.
5162 18. Tabs are now shown as \t in -bP output.
5164 19. If the log selector return_path_on_delivery was set, Exim crashed when
5165 bouncing a message because it had too many Received: header lines.
5167 20. If two routers both had headers_remove settings, and the first one included
5168 a superfluous trailing colon, the final name in the first list and the
5169 first name in the second list were incorrectly joined into one item (with a
5170 colon in the middle).
5176 1. Added -C and -D options to the exinext utility, mainly to make it easier
5177 to include in the automated testing, but these could be helpful when
5178 multiple configurations are in use.
5180 2. The exinext utility was not formatting the output nicely when there was
5181 an alternate port involved in the retry record key, nor when there was a
5182 message id as well (for retries that were specific to a specific message
5183 and a specific host). It was also confused by IPv6 addresses, because of
5184 the additional colons they contain. I have fixed the IPv4 problem, and
5185 patched it up to do a reasonable job for IPv6.
5187 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5188 delivery, the log line now contains "pipelined" if PIPELINING was used.
5190 4. An SMTP transport process used to panic and die if the bind() call to set
5191 an explicit outgoing interface failed. This has been changed; it is now
5192 treated in the same way as a connect() failure.
5194 5. A reference to $sender_host_name in the part of a conditional expansion
5195 that was being skipped was still causing a DNS lookup. This no longer
5198 6. The def: expansion condition was not recognizing references to header lines
5199 that used bh_ and bheader_.
5201 7. Added the _cache feature to named lists.
5203 8. The code for checking quota_filecount in the appendfile transport was
5204 allowing one more file than it should have been.
5206 9. For compatibility with Sendmail, the command line option
5214 and sets the incoming protocol and host name (for trusted callers). The
5215 host name and its colon can be omitted when only the protocol is to be set.
5216 Note the Exim already has two private options, -pd and -ps, that refer to
5217 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5218 "s", but I don't think that's a major issue.
5220 10. A number of refactoring changes to the code, none of which should affect
5223 (a) The number of logging options was getting close to filling up the
5224 32-bit word that was used as a bit map. I have split them into two classes:
5225 those that are passed in the argument to log_write(), and those that are
5226 only ever tested independently outside of that function. These are now in
5227 separate 32-bit words, so there is plenty of room for expansion again.
5228 There is no change in the user interface or the logging behaviour.
5230 (b) When building, for example, log lines, the code previously used a
5231 macro that called string_cat() twice, in order to add two strings. This is
5232 not really sufficiently general. Furthermore, there was one instance where
5233 it was actually wrong because one of the argument was used twice, and in
5234 one call a function was used. (As it happened, calling the function twice
5235 did not affect the overall behaviour.) The macro has been replaced by a
5236 function that can join an arbitrary number of extra strings onto a growing
5239 (c) The code for expansion conditions now uses a table and a binary chop
5240 instead of a serial search (which was left over from when there were very
5241 few conditions). Also, it now recognizes conditions like "pam" even when
5242 the relevant support is not compiled in: a suitably worded error message is
5243 given if an attempt is made to use such a condition.
5245 11. Added ${time_interval:xxxxx}.
5247 12. A bug was causing one of the ddress fields not to be passed back correctly
5248 from remote delivery subprocesses. The field in question was not being
5249 subsequently used, so this caused to problems in practice.
5251 13. Added new log selectors queue_time and deliver_time.
5253 14. Might have fixed a bug in maildirsizefile handling that threw up
5254 "unexpected character" debug warnings, and recalculated the data
5255 unnecessarily. In any case, I expanded the warning message to give more
5258 15. Added the message "Restricted characters in address" to the statements in
5259 the default ACL that block characters like @ and % in local parts.
5261 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5262 Three changes have been made:
5264 (a) There was a serious bug; a negative response to MAIL caused the whole
5265 recipient domain to be cached as invalid, thereby blocking all messages
5266 to all local parts at the same domain, from all senders. This bug has
5267 been fixed. The domain is no longer cached after a negative response to
5268 MAIL if the sender used is not empty.
5270 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5273 (c) A new callout option, "use_sender" has been added for people who want
5274 the modified behaviour.
5280 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5283 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5284 indeed breaks things for older releases.
5286 3. Added additional logging to the case where there is a problem reading data
5287 from a filter that is running in a subprocess using a pipe, in order to
5288 try to track down a specific problem.
5290 4. Testing facility fudge: when running in the test harness and attempting
5291 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5292 getting "No route to host". Convert this to a timeout.
5294 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5297 6. Some OS don't have socklen_t but use size_t instead. This affects the
5298 fifth argument of getsockopt() amongst other things. This is now
5299 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5300 can be set for individual OS. I have set it for SunOS5, OSF1, and
5301 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5302 some earlier ones do not.
5304 7. Change 4.30/15 was not doing the test caselessly.
5306 8. The standard form for an IPv6 address literal was being rejected by address
5307 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5308 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5309 this, as well as the form without the "IPv6" on the front (but only when
5310 address literals are enabled, of course).
5312 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5314 10. Exim crashed if a message with an empty sender address specified by -f
5315 encountered a router with an errors_to setting. This could be provoked only
5316 by a command such as
5320 where an empty string was supplied; "<>" did not hit this bug.
5322 11. Installed PCRE release 4.5.
5324 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5325 remained set. It is now erased.
5327 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5328 times from message ids (which are base 36 rather than the normal 62).
5330 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5331 were being counted as actual protocol errors, and logged if the log
5332 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5333 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5334 and DATA following a set of rejected RCPTs do not count as protocol errors.
5335 In other words, Exim assumes they were pipelined, though this may not
5336 actually be the case. Of course, in all cases the client gets an
5337 appropriate error code.
5339 15. If a lookup fails in an ACL condition, a message about the failure may
5340 be available; it is used if testing the ACL cannot continue, because most
5341 such messages specify what the cause of the deferral is. However, some
5342 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5343 that caused an old message to be retained and used if a later statement
5344 caused a defer, replacing the real cause of the deferral.
5346 16. If an IP address had so many PTR records that the DNS lookup buffer
5347 was not large enough to hold them, Exim could crash while trying to process
5348 the truncated data. It now detects and logs this case.
5350 17. Further to 4.21/58, another change has been made: if (and only if) the
5351 first line of a message (the first header line) ends with CRLF, a bare LF
5352 in a subsequent header line has a space inserted after it, so as not to
5353 terminate the header.
5355 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5356 unnecessarily, used atoi() instead of strtol(), and didn't check the
5357 termination when getting file sizes from file names by regex.
5359 19. Completely re-implemented the support for maildirsize files, in the light
5360 of a number of problems with the previous contributed implementation
5361 (4.30/29). In particular:
5363 . If the quota is zero, the maildirsize file is maintained, but no quota is
5366 . If the maildir directory does not exist, it is created before any attempt
5367 to write a maildirsize file.
5369 . The quota value in the file is just a cache; if the quota is changed in
5370 the transport, the new value overrides.
5372 . A regular expression is available for excluding directories from the
5375 20. The autoreply transport checks the characters in options that define the
5376 message's headers; it allows continued headers, but it was checking with
5377 isspace() after an embedded newline instead of explicitly looking for a
5380 21. If all the "regular" hosts to which an address was routed had passed their
5381 expiry times, and had not reached their retry times, the address was
5382 bounced, even if fallback hosts were defined. Now Exim should go on to try
5385 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5386 equivalent code in the SMTP transport. Some hosts send humungous responses
5387 to HELO/EHLO, more than 1024 it seems.
5389 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5390 gives compiler warnings in some environments. I've now done it "properly",
5393 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5394 (because of problems with the built-in one) was declared to return uschar *
5395 instead of char *, causing compiler failure.
5397 25. Fixed a file descriptor leak when processing alias/forward files.
5399 26. Fixed a minor format string issue in dbfn.c.
5401 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5403 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5404 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5407 29. When a sender address is verified, it is cached, to save repeating the test
5408 when there is more than one recipient in a message. However, when the
5409 verification involves a callout, it is possible for different callout
5410 options to be set for different recipients. It is too complicated to keep
5411 track of this in the cache, so now Exim always runs a verification when a
5412 callout is required, relying on the callout cache for the optimization.
5413 The overhead is duplication of the address routing, but this should not be
5416 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5417 to be verified with callout=postmaster, and the main callout worked but the
5418 postmaster check failed, the verification correctly failed. However, if a
5419 subsequent RCPT command asked for sender verification *without* the
5420 postmaster check, incorrect caching caused this verification also to fail,
5423 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5424 it was not caching the DNS options (qualify_single, search_parents) that
5425 were used when the lookup failed. A subsequent lookup with different
5426 options therefore always gave the same answer, though there were cases
5427 where it should not have. (Example: a "domains = !$mx_any" option on a
5428 dnslookup router: the "domains" option is always processed without any
5429 widening, but the router might have qualify_single set.) Now Exim uses the
5430 cached value only when the same options are set.
5432 32. Added John Jetmore's "exipick" utility to the distribution.
5434 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5435 than a timeout (e.g. a certificate is required, and is not provided), an
5436 Exim server now closes the connection immediately. Previously it waited for
5437 the client to close - but if the client is SSL, it seems that they each
5438 wait for each other, leading to a delay before one of them times out.
5440 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5441 maintained 0.8.x compatibility because I don't think many are using it, and
5442 it is clearly obsolete.
5444 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5447 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5448 client certificate was expired. A simple patch fixes this, though I don't
5449 understand the full logic of why the verify callback is called multiple
5452 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5453 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5454 which causes problems with some clients (such as the Certicom SSL Plus
5455 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5456 disables the coutermeasure allowing Eudora to connect."
5458 38. Exim was not checking that a write() to a log file succeeded. This could
5459 lead to Bad Things if a log got too big, in particular if it hit a file
5460 size limit. Exim now panics and dies if it cannot write to a log file, just
5461 as it does if it cannot open a log file.
5463 39. Modified OS/Makefile-Linux so that it now contains
5465 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5467 The two -D definitions ensure that Exim is compiled with large file
5468 support, which makes it possible to handle log files that are bigger than
5471 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5472 instance) a domain was checked against a named list that involved a lookup,
5473 causing $domain_data to be set, then another domain was checked against the
5474 same list, then the first domain was re-checked, the value of $domain_data
5475 after the final check could be wrong. In particular, if the second check
5476 failed, it could be set empty. This bug probably also applied to
5479 41. The strip_trailing_dot option was not being applied to the address given
5480 with the -f command-line option.
5482 42. The code for reading a message's header from the spool was incrementing
5483 $received_count, but never initializing it. This meant that the value was
5484 incorrect (doubled) while delivering a message in the same process in which
5485 it was received. In the most common configuration of Exim, this never
5486 happens - a fresh exec is done - but it can happen when
5487 deliver_drop_privilege is set.
5489 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5490 it now includes up to 150 characters of the unexpected data in the log
5493 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5494 and building data strings. The size of both of these buffers was 10 000
5495 bytes - far larger than anybody would *ever* want, thought I. Needless to
5496 say, somebody hit the limit. I have increased the maximum line length to
5497 20 000 and the maximum data length of concatenated lines to 100 000. I have
5498 also fixed two bugs, because there was no checking on these buffers. Tsk,
5499 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5500 buffer is too small.
5502 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5503 lsearch lookups. Now it does.
5505 46. When parsing a route_list item in a manualroute router, a fixed-length
5506 buffer was used for the list of hosts. I made this 1024 bytes long,
5507 thinking that nobody would ever have a list of hosts that long. Wrong.
5508 Somebody had a whole pile of complicated expansion conditions, and the
5509 string was silently truncated, leading to an expansion error. It turns out
5510 that it is easier to change to an unlimited length (owing to other changes
5511 that have happened since this code was originally written) than to build
5512 structure for giving a limitation error. The length of the item that
5513 expands into the list of hosts is now unlimited.
5515 47. The lsearch lookup could not handle data where the length of text line was
5516 more than 4095 characters. Such lines were truncated, leading to shortened
5517 data being returned. It should now handle lines of any length.
5519 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5520 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5523 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5524 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5525 no longer happens in comment lines. A list of replacements is now placed
5526 at the head of all of the source files, except those whose only change is
5527 to replace PERL_COMMAND in the very first #! line.
5529 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5530 messages on the queue, with a bottom-up merge sort, using code contributed
5531 by Michael Haardt. This should make operations like -bp somewhat faster on
5532 large queues. It won't affect queue runners, except when queue_run_in_order
5535 51. Installed eximstats 1.31 in the distribution.
5537 52. Added support for SRV lookups to the dnslookup router.
5539 53. If an ACL referred to $message_body or $message_body_end, the value was not
5540 reset for any messages that followed in the same SMTP session.
5542 54. The store-handling optimization for building very long strings was not
5543 differentiating between the different store pools. I don't think this
5544 actually made any difference in practice, but I've tidied it.
5546 55. While running the routers to verify a sender address, $sender_address
5547 was still set to the sender address. This is wrong, because when routing to
5548 send a bounce to the sender, it would be empty. Therefore, I have changed
5549 it so that, while verifying a sender address, $sender_address is set to <>.
5550 (There is no change to what happens when verifying a recipient address.)
5552 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5553 target A or AAAA records (if not already returned) without resetting the
5554 qualify_single or search_parents options of the DNS resolver. These are
5555 inappropriate in this case because the targets of MX and SRV records must
5556 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5557 target that, when qualified, matched something in the local domain. These
5558 two options are now turned off when doing these lookups.
5560 57. It seems that at least some releases of Reiserfs (which does not have the
5561 concept of a fixed number of inodes) returns zero and not -1 for the
5562 number of available inodes. This interacted badly with check_spool_inodes,
5563 which assumed that -1 was the "no such thing" setting. What I have done is
5564 to check that the total number of inodes is greater than zero before doing
5565 the test of how many are available.
5567 58. When a "warn" ACL statement has a log_message modifier, the message is
5568 remembered, and not repeated. This is to avoid a lot of repetition when a
5569 message has many recipients that cause the same warning to be written.
5570 Howewer, Exim was preserving the list of already written lines for an
5571 entire SMTP session, which doesn't seem right. The memory is now reset if a
5572 new message is started.
5574 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5575 debugging output unless log_rewrite was also set.
5577 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5578 of (int)(handle) when we know that handle contains (void *)(-1).
5580 61. The Exim daemon panic-logs an error return when it closes the incoming
5581 connection. However "connection reset by peer" seems to be common, and
5582 isn't really an error worthy of noting specially, so that particular error
5585 62. When Exim is trying to find all the local interfaces, it used to panic and
5586 die if the ioctl to get the interface flags failed. However, it seems that
5587 on at least one OS (Solaris 9) it is possible to have an interface that is
5588 included in the list of interfaces, but for which you get a failure error
5589 for this call. This happens when the interface is not "plumbed" into a
5590 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5591 failure of the "get flags" call assumes that the interface is down.
5593 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5594 makes life easier for people who are doing arithmetic on fields extracted
5595 from dates, where you often get leading zeros that should not be
5596 interpreted as octal.
5598 64. Added qualify_domain to the redirect router, to override the global
5601 65. If a pathologically long header line contained very many addresses (the
5602 report of this problem mentioned 10 000) and each of them was rewritten,
5603 Exim could use up a very large amount of memory. (It kept on making new
5604 copies of the header line as it rewrote, and never released the old ones.)
5605 At the expense of a bit more processing, the header rewriting function has
5606 been changed so that it no longer eats memory in this way.
5608 66. The generation of the Received: header has been moved from the time that a
5609 message starts to be received, to the time that it finishes. The timestamp
5610 in the Received: header should now be very close to that of the <= log
5611 line. There are two side-effects of this change:
5613 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5614 logged header lines no longer include the local Received: line, because
5615 it has not yet been created. The same applies to a copy of the message
5616 that is returned to a non-SMTP sender when a message is rejected.
5618 (b) When a filter file is tested using -bf, no additional Received: header
5619 is added to the test message. After some thought, I decided that this
5622 This change does not affect the value of $received_for. It is still set
5623 after address rewriting, but before local_scan() is called.
5625 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5627 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5628 gave an unhelpful panic error message, and a defer error. I have managed to
5629 change this behaviour so that it now rejects any supplied certificate,
5630 which seems right, as the list of acceptable certificates is empty.
5632 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5633 gave an unhelpful defer error. I have not managed to make this reject any
5634 supplied certificates, but the error message it gives is "no certificate
5635 supplied", which is not helpful.
5637 70. exigrep's output now also includes lines that are not associated with any
5638 message, but which match the given pattern. Implemented by a patch from
5639 Martin Sluka, which also tidied up the Perl a bit.
5641 71. Recipient callout verification, like sender verification, was using <> in
5642 the MAIL FROM command. This isn't really the right thing, since the actual
5643 sender may affect whether the remote host accepts the recipient or not. I
5644 have changed it to use the actual sender in the callout; this means that
5645 the cache record is now keyed on a recipient/sender pair, not just the
5646 recipient address. There doesn't seem to be a real danger of callout loops,
5647 since a callout by the remote host to check the sender would use <>.
5648 [SEE ABOVE: changed after hitting problems.]
5650 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5651 temporary errors. However, in the case of such a code being given after
5652 the end of a data transmission (i.e. after ".") Exim was failing to write
5653 a retry record for the message. (Yes, there was some broken host that was
5654 actually sending 8xx at this point.)
5656 73. An unknown lookup type in a host list could cause Exim to panic-die when
5657 the list was checked. (An example that provoked this was putting <; in the
5658 middle of a list instead of at the start.) If this happened during a DATA
5659 ACL check, a -D file could be left lying around. This kind of configuration
5660 error no longer causes Exim to die; instead it causes a defer errror. The
5661 incident is still logged to the main and panic logs.
5663 74. Buglet left over from Exim 3 conversion. The message "too many messages
5664 in one connection" was written to the rejectlog but not the mainlog, except
5665 when address rewriting (yes!) was being logged.
5667 75. Added write_rejectlog option.
5669 76. When a system filter was run not as root (that is, when system_filter_user
5670 was set), the values of the $n variables were not being returned to the
5671 main process; thus, they were not subsequently available in the $sn
5674 77. Added +return_path_on_delivery log selector.
5676 78. A connection timeout was being treated differently from recipients deferred
5677 when testing hosts_max_try with a message that was older than the host's
5678 retry timeout. (The host should not be counted, thus allowing all hosts to
5679 be tried at least once before bouncing.) This may have been the cause of an
5680 occasionally reported bug whereby a message would remain on the queue
5681 longer than the retry timeout, but would be bounced if a delivery was
5682 forced. I say "may" because I never totally pinned down the problem;
5683 setting up timeout/retry tests is difficult. See also the next item.
5685 79. The ultimate address timeout was not being applied to errors that involved
5686 a combination of host plus message (for example, a timeout on a MAIL
5687 command). When an address resolved to a number of possible hosts, and they
5688 were not all tried for each delivery (e.g. because of hosts_max_try), a
5689 message could remain on the queue longer than the retry timeout.
5691 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5694 81. Fixed an obscure SMTP outgoing bug which required at least the following
5695 conditions: (a) there was another message waiting for the same server;
5696 (b) the server returned 5xx to all RCPT commands in the first message so
5697 that the message was not completed; (c) the server dropped the connection
5698 or gave a negative response to the RSET that Exim sends to abort the
5699 transaction. The observed case was a dropped connection after DATA that had
5700 been sent in pipelining mode. That is, the server had advertised PIPELINING
5701 but was not implementing it correctly. The effect of the bug was incorrect
5702 behaviour, such as trying another host, and this could lead to a crash.
5708 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5709 and daemon.c were passed as pointers to ints; they should have been
5710 pointers to socklen_t variables (which are typically unsigned ints).
5712 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5715 3. Fixed a really odd bug that affected only the testing scheme; patching a
5716 certain fixed string in the binary changed the value of another string that
5717 happened to be identical to the end of the original first string.
5719 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5720 name", it returns that address as the IP address. On some operating
5721 systems (e.g. Solaris), it also passes back the IP address string as the
5722 "host name". However, on others (e.g. Linux), it passes back an empty
5723 string. Exim wasn't checking for this, and was changing the host name to an
5724 empty string, assuming it had been canonicized.
5726 5. Although rare, it is permitted to have more than one PTR record for a given
5727 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5728 all the names associated with an address, because they do in Solaris.
5729 However, it seems that they do not in Linux for data that comes from the
5730 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5731 I found this out when I moved to a new Linux workstation and tried to run
5732 the Exim test suite.
5734 To get round this problem I have changed the code so that it now does its
5735 own call to the DNS to look up PTR records when searching for a host name.
5736 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5737 addresses that are only in /etc/hosts are still found.
5739 This behaviour is, however, controlled by an option called host_lookup_
5740 order, which defaults to "bydns:byaddr". If people want to use the other
5741 order, or indeed, just use one or the other means of lookup, they can
5742 specify it in this variable.
5744 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5745 some operating systems, this comes back from gethostbyaddr() as an empty
5746 string, and this is what Exim used to test for. However, it seems that in
5747 other systems, "." is yielded. Exim now tests for this case too.
5749 7. The values of check_spool_space and check_log_space are now held internally
5750 as a number of kilobytes instead of an absolute number of bytes. If a
5751 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5752 kilobyte. This means that much larger values can be stored.
5754 8. Exim monitor: an attempt to get the action menu when not actually pointing
5755 at a message produces an empty menu entitled "No message selected". This
5756 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5757 no entries in it ("Shell widget menu has zero width and/or height"). So I
5758 have added a single, blank menu entry in this case.
5760 9. Added ${quote_local_part.
5762 10. MIME decoding is now applied to the contents of Subject: header lines when
5765 11. Now that a reference to $sender_host_address automatically causes a reverse
5766 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5767 host lookup before query-style lookups in lists that might use this
5768 variable. This has therefore been abolished, and the "net-" prefix is no
5769 longer necessary for query-style lookups.
5771 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5772 have been a typo for LFLAGS, so it has been changed.
5774 13. The install script calls Exim with "-C /dev/null" in order to find the
5775 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5776 to be output. Howeve, since Exim outputs its version number before the
5777 error, it didn't break the script. It just looked ugly. I fixed this by
5778 always allowing "-C /dev/null" if the caller is root.
5780 14. Ignore overlarge ACL variable number when reading spool file - insurance
5781 against a later release with more variables having written the file.
5783 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5784 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5785 this, as well as the form without the "IPv6" on the front.
5787 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5788 OS/Makefile-Darwin file.
5790 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5791 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5793 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5794 variable. However, some OS use ints for inodes. I've added cast to long int
5795 to get rid of the compiler warning.
5797 19. I had forgotten to lock out "/../" in configuration file names when
5798 ALT_CONFIG_PREFIX was set.
5800 20. Routers used for verification do not need to specify transports. However,
5801 if such a router generated a host list, and callout was configured, Exim
5802 crashed, because it could not find a port number from the (non-existent)
5803 transport. It now assumes port 25 in this circumstance.
5805 21. Added the -t option to exigrep.
5807 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5808 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5809 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5810 them is set, LOOKUP_LSEARCH is forced.
5812 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5814 24. Added sender and host information to the "rejected by local_scan()" log
5815 line; previously there was no indication of these.
5817 25. Added .include_if_exists.
5819 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5820 Linux. It turns out that not all file systems support this. Apparently some
5821 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5822 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5823 not supported on the file descriptor, is now ignored when Exim is trying to
5824 sync a directory. This applies only to Linux.
5826 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5828 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5829 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5832 29. Added support for maildirsize files from supplied patch (modified a bit).
5834 30. The use of :fail: followed by an empty string could lead Exim to respond to
5835 sender verification failures with (e.g.):
5837 550 Verification failed for <xxx>
5838 550 Sender verify failed
5840 where the first response line was missing the '-' that indicates it is not
5841 the final line of the response.
5843 31. The loop for finding the name of the user that called Exim had a hardwired
5844 limit of 10; it now uses the value of finduser_retries, which is used for
5845 all other user lookups.
5847 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5850 33. Exim was neglecting to zero errno before one call of strtol() when
5851 expanding a string and expecting an integer value. On some systems this
5852 resulted in spurious "integer overflow" errors. Also, it was casting the
5853 result into an int without checking.
5855 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5856 did not work. The code looks as if it has *never* worked, though it appears
5857 to have been documented since at least releast 1.62. I have made it work.
5859 35. The "timeout_DNS" error in retry rules, also documented since at least
5860 1.62, also never worked. As it isn't clear exactly what this means, and
5861 clearly it isn't a major issue, I have abolished the feature by treating it
5862 as "timeout", and writing a warning to the main and panic logs.
5864 36. The display of retry rules for -brt wasn't always showing the error code
5867 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5868 timeout_connect_A, timeout_connect_MX.
5870 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5871 to the empty sender.
5873 39. The daemon was not analysing the content of -oX till after it had closed
5874 stderr and disconnected from the controlling terminal. This meant that any
5875 syntax errors were only noted on the panic log, and the return code from
5876 the command was 0. By re-arranging the code a little, I've made the
5877 decoding happen first, so such errors now appear on stderr, and the return
5878 code is 1. However, the actual setting up of the sockets still happens in
5879 the disconnected process, so errors there are still only recorded on the
5882 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5883 connections (as happens on some IP stacks) was logged at start up time as
5884 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5885 it from "IPv6 and IPv4", which means that two separate sockets are being
5888 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5889 says whether AF_INET or AF_INET6 was passed as an argument.
5891 42. Exiwhat output was messed up when time zones were included in log
5894 43. Exiwhat now gives more information about the daemon's listening ports,
5895 and whether -tls-on-connect was used.
5897 44. The "port" option of the smtp transport is now expanded.
5899 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5900 silently ignored. Now an error message is written to the main and panic
5903 46. There's a new ACL modifier called "logwrite" which writes to a log file
5904 as soon as it is encountered.
5906 47. Added $local_user_uid and $local_user_gid at routing time.
5908 48. Exim crashed when trying to verify a sender address that was being
5911 49. Exim was recognizing only a space character after ".include". It now also
5912 recognizes a tab character.
5914 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5915 extracting the relevant information from the specification. The man page no
5916 longer contains scrambled data for the -d option, and I've added a section
5917 at the front about calling Exim under different names.
5919 51. Added "extra_headers" argument to the "mail" command in filter files.
5921 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5924 53. Installed eximstats 1.29.
5926 54. Added transport_filter_timeout as a generic transport option.
5928 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5929 Cc: header lines. This was required by RFC 822, but it not required by RFC
5932 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5933 incoming messages that did not have them. Now it does so only if the
5934 message originates locally, that is, if there is no associated remote host
5935 address. When Resent- header lines are present, this applies to the Resent-
5936 lines rather than the non-Resent- lines.
5938 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5939 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5941 58. Messages for configuration errors now include the name of the main
5942 configuration file - useful now that there may be more than one file in a
5943 list (.included file names were always shown).
5945 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5946 for those rare installations that do not start the daemon as root or run it
5947 setuid root. I've cut out the call to initgroups() if the daemon is not
5950 60. The Exim user and group can now be bound into the binary as text strings
5951 that are looked up at the start of Exim's processing.
5953 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5955 62. Added $mailstore_basename variable.
5957 63. Installed patch to sieve.c from Michael Haardt.
5959 64. When Exim failed to open the panic log after failing to open the main log,
5960 the original message it was trying to log was written to stderr and debug
5961 output, but if they were not available (the usual case in production), it
5962 was lost. Now it is written to syslog before the two lines that record the
5963 failures to open the logs.
5965 65. Users' Exim filters run in subprocesses under the user's uid. It is
5966 possible for a "deliver" command or an alias in a "personal" command to
5967 provoke an address rewrite. If logging of address rewriting is configured,
5968 this fails because the process is not running as root or exim. There may be
5969 a better way of dealing with this, but for the moment (because 4.30 needs
5970 to be released), I have disabled address rewrite logging when running a
5971 filter in a non-root, non-exim process.
5977 1. The buildconfig auxiliary program wasn't quoting the value set for
5978 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5979 not defined. This bug was masked in 4.22 by the effect that was fixed in
5982 2. Some messages that were rejected after a message id was allocated were
5983 shown as "incomplete" by exigrep. It no longer does this for messages that
5984 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5986 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5987 have allow_domain_literals set, the ID did not get logged in the <= line.
5988 Domain literals are now always recognized in Message-ID: header lines.
5990 4. The first argument for a ${extract expansion item is the key name or field
5991 number. Leading and trailing spaces in this item were not being ignored,
5992 causing some misleading effects.
5994 5. When deliver_drop_privilege was set, single queue runner processes started
5995 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5996 same command in the process it spins off) were not dropping privilege.
5998 6. When the daemon running as "exim" started a queue runner, it always
5999 re-executed Exim in the spun-off process. This is a waste of effort when
6000 deliver_drop_privilege is set. The new process now just calls the
6001 queue-runner function directly.
6007 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6010 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6011 ignored. Though the use of -oP was forcing the writing of a pid file, it
6012 was always written to the default place.
6014 3. If the message "no IP address found for host xxxx" is generated during
6015 incoming verification, it is now followed by identification of the incoming
6016 connection (so you can more easily find what provoked it).
6018 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6020 5. Added some features to "harden" Exim a bit more against certain attacks:
6022 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6023 be put in Local/Makefile. This is like the never_users runtime option,
6024 but it cannot be overridden. The default setting is "root".
6026 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6027 prefix string with which any file named in a -C command line option
6030 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6031 is retained for -C and -D only if the caller of Exim is root. Without
6032 it, the exim user may also use -C and -D and retain privilege.
6034 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6035 command line option is disabled.
6037 6. Macro names set by the -D option must start with an upper case letter, just
6038 like macro names defined in the configuration file.
6040 7. Added "dereference=" facility to LDAP.
6042 8. Two instances of the typo "uknown" in the source files are fixed.
6044 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6045 the Configure-Makefile script screwed up while processing it.
6047 10. Incorporated PCRE 4.4.
6049 11. The SMTP synchronization check was not operating right at the start of an
6050 SMTP session. For example, it could not catch a HELO sent before the client
6051 waited for the greeting. There is now a check for outstanding input at the
6052 point when the greeting is written. Because of the duplex, asynchronous
6053 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6054 way, but not yet received, when the check is performed.
6056 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6057 on TCP/IP sockets, because this apparently causes some broken clients to
6060 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6061 unchanged) from the Cygwin maintainer.
6063 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6064 was supported instead of testing for mbx. Effectively a typo.
6066 15. The spa authenticator server code was not checking that the input it
6067 received was valid base64.
6069 16. The debug output line for the "set" modifier in ACLs was not showing the
6070 name of the variable that was being set.
6072 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6074 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6075 lookup on its own. Something else was needed to trigger the lookup. For
6076 example, a match in host_lookup or the need for a host name in a host list.
6077 Now, if $sender_host_name is referenced and the host name has not yet been
6078 looked up, a lookup is performed. If the lookup fails, the variable remains
6079 empty, and $host_lookup_failed is set to "1".
6081 19. Added "eqi" as a case-independent comparison operator.
6083 20. The saslauthd authentication condition could segfault if neither service
6084 nor realm was specified.
6086 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6087 error message that was logged was misleading, and incoming SMTP
6088 connections were dropped. The message is now more accurate, and temporary
6089 errors are given to SMTP connections.
6091 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6092 (except RSET) with a 421 error, until QUIT is received. However, it was
6093 failing to send a response to QUIT.
6095 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6096 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6097 ACL is now run after the helo_try_verify_hosts code.
6099 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6100 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6101 case-independent, so other case variants are also recognized). Apparently
6102 some systems use these upper case variants.
6104 25. If more than two messages were waiting for the same host, and a transport
6105 filter was specified for the transport, Exim sent two messages over the
6106 same TCP/IP connection, and then failed with "socket operation on non-
6107 socket" when it tried to send the third.
6109 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6111 27. The extern definition of crypt16() in expand.c was not being excluded when
6112 the OS had its own crypt16() function.
6114 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6115 as a preferred synonym for return_size_limit, both as an option and as an
6118 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6120 30. Changed the default configuration ACL to relax the local part checking rule
6121 for addresses that are not in any local domains. For these addresses,
6122 slashes and pipe symbols are allowed within local parts, but the sequence
6123 /../ is explicitly forbidden.
6125 31. SPA server authentication was not clearing the challenge buffer before
6128 32. log_message in a "warn" ACL statement was writing to the reject log as
6129 well as to the main log, which contradicts the documentation and doesn't
6130 seem right (because no rejection is happening). So I have stopped it.
6132 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6133 However, I am unable to do any testing of this.
6135 34. Fixed an infelicity in the appendfile transport. When checking directories
6136 for a mailbox, to see if any needed to be created, it was accidentally
6137 using path names with one or more superfluous leading slashes; tracing
6138 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6140 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6141 contents are added to the log line that is written for every discarded
6142 recipient. (Previously a log_message setting was ignored.)
6144 36. The ${quote: operator now quotes the string if it is empty.
6146 37. The install script runs exim in order to find its version number. If for
6147 some reason other than non-existence or emptiness, which it checks, it
6148 could not run './exim', it was installing it with an empty version number,
6149 i.e. as "exim-". This error state is now caught, and the installation is
6152 38. An argument was missing from the function that creates an error message
6153 when Exim fails to connect to the socket for saslauthd authentication.
6154 This could cause Exim to crash, or give a corrupted message.
6156 39. Added isip, isip4, and isip6 to ${if conditions.
6158 40. The ACL variables $acl_xx are now saved with the message, and can be
6159 accessed later in routers, transports, and filters.
6161 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6162 strings in the file are not string-expanded.
6164 42. If a MAIL command specified a SIZE value that was too large to fit into an
6165 int variable, the check against message_size_limit failed. Such values are
6166 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6167 day this will have to be increased, but I don't think I want to be around
6168 when emails are that large.
6175 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6176 iconv() is not standard in FreeBSD.
6178 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6179 IPv6 enabled. The observed symptom was a segmentation fault on return from
6180 the function os_common_find_running_interfaces() in src/os.c.
6182 3. In the check_special_case() function in daemon.c I had used "errno" as an
6183 argument name, which causes warnings on some systems. This was basically a
6184 typo, since it was named "eno" in the comments!
6186 4. The code that waits for the clock to tick (at a resolution of some fraction
6187 of a second) so as to ensure message-id uniqueness was always waiting for
6188 at least one whole tick, when it could have waited for less. [This is
6189 almost certainly not relevant at current processor speeds, where it is
6190 unlikely to ever wait at all. But we try to future-proof.]
6192 5. The function that sleeps for a time interval that includes fractions of a
6193 second contained a race. It did not block SIGALRM between setting the
6194 timer, and suspending (a couple of lines later). If the interval was short
6195 and the sigsuspend() was delayed until after it had expired, the suspension
6196 never ended. On busy systems this could lead to processes getting stuck for
6199 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6200 process, before it forks any delivery processes. The open lookup caching
6201 mechanism meant that the open file or database connection was passed into
6202 the delivery process. The problem was that delivery processes always tidy
6203 up cached lookup data. This could cause a problem for the next delivery
6204 process started by the queue runner, because the external queue runner
6205 process does not know about the closure. So the next delivery process
6206 still has data in the lookup cache. In the case of a file lookup, there was
6207 no problem because closing a file descriptor in a subprocess doesn't affect
6208 the parent. However, if the lookup was caching a connection to a database,
6209 the connection was closed, and the second delivery process was likely to
6210 see errors such as "PGSQL: query failed: server closed the connection
6211 unexpectedly". The problem has been fixed by closing all cached lookups
6212 in a queue runner before running a delivery process.
6214 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6215 seem to have the "const" qualifier which it has on other OS. I've
6218 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6219 *of the same type* (client or server) with the same public name that an
6220 error should be diagnosed.
6222 9. When Exim looked up a host name for an IP address, but failed to find the
6223 original IP address when looking up the host name (a safety check), it
6224 output the message "<ip address> does not match any IP for NULL", which was
6225 confusing, to say the least. The bug was that the host name should have
6226 appeared instead of "NULL".
6228 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6229 user that is built into the binary, and the -C or -D options is used, root
6230 privilege is dropped before the configuration file is read. In addition,
6231 logging is switched to stderr instead of the normal log files. If the
6232 configuration then re-defines the Exim user, the unprivileged environment
6233 is probably not what is expected, so Exim logs a panic warning message (but
6236 However, if deliver_drop_privilege is set, the unprivileged state may well
6237 be exactly what is intended, so the warning has been cut out in that case,
6238 and Exim is allowed to try to write to its normal log files.
6244 1. smtp_return_error_details was not giving details for temporary sender
6245 or receiver verification errors.
6247 2. Diagnose a configuration error if two authenticators have the same public
6250 3. Exim used not to create the message log file for a message until the first
6251 delivery attempt. This could be confusing when incoming messages were held
6252 for policy or load reasons. The message log file is now created at the time
6253 the message is received, and an initial "Received" line is written to it.
6255 4. The automatically generated man page for command line options had a minor
6256 bug that caused no ill effects; however, a more serious problem was that
6257 the procedure for building the man page automatically didn't always
6258 operate. Consequently, release 4.20 contains an out-of-date version. This
6259 shouldn't happen again.
6261 5. When building Exim with embedded Perl support, the script that builds the
6262 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6263 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6265 6. The freeze_tell option was not being used for messages that were frozen on
6266 arrival, either by an ACL or by local_scan().
6268 7. Added the smtp_incomplete_transaction log selector.
6270 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6271 was accepting AUTH without a new EHLO.
6273 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6274 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6277 10. Logging of TCP/IP connections (when configured) now happens in the main
6278 daemon process instead of the child process, so that the TCP/IP connection
6279 count is more accurate (but it can never be perfect).
6281 11. The use of "drop" in a nested ACL was not being handled correctly in the
6282 outer ACL. Now, if condition failure induced by the nested "drop" causes
6283 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6284 or "require"), the connection is dropped.
6286 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6287 that yield "discard" can now be used with an "accept" or a "discard" verb,
6288 but an error is generated for any others (because I can't see a useful way
6289 to define what should happen).
6291 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6292 are now processed in the same way as lines in the Exim configuration file.
6293 In particular, continuation lines are supported.
6295 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6297 15. Added -ti meaning -t -i.
6299 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6300 domains, and warn by logging if others are found.
6302 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6303 returned from GIFCONF when Exim is trying to find the list of interfaces on
6304 a host. The code in os.c has been modified to copy each ifreq to an aligned
6305 structure in all cases.
6307 Also, in some cases, the returned ifreq's were being copied to a 'struct
6308 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6309 means the last couple of bytes of an IPv6 address could be chopped if the
6310 ifreq contained only a normal sockaddr (14 bytes storage).
6312 18. Named domain lists were not supported in the hosts_treat_as_local option.
6313 An entry such as +xxxx was not recognized, and was treated as a literal
6316 19. Ensure that header lines added by a DATA ACL are included in the reject log
6317 if the ACL subsequently rejects the message.
6319 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6320 MD5 (which is deprecated).
6322 21. When testing a filter file using -bf, Exim was writing a message when it
6323 took the sender from a "From " line in the message, but it was not doing so
6324 when it took $return_path from a Return-Path: header line. It now does.
6326 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6327 with a valid header line field name (a series of printing characters
6328 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6330 23. Changed "disc" in the source to "disk" to conform to the documentation and
6331 the book and for uniformity.
6333 24. Ignore Sendmail's -Ooption=value command line item.
6335 25. When execve() failed while trying to run a command in a pipe transport,
6336 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6337 could be confused with a return value of 69 from the command itself. This
6338 has been changed to 127, the value the shell returns if it is asked to run
6339 a non-existent command. The wording for the related log line suggests a
6340 non-existent command as the problem.
6342 26. If received_header_text expands to an empty string, do not add a Received:
6343 header line to the message. (Well, it adds a token one on the spool, but
6344 marks it "old" so that it doesn't get used or transmitted.)
6346 27. Installed eximstats 1.28 (addition of -nt option).
6348 28. There was no check for failure on the call to getsockname() in the daemon
6349 code. This can fail if there is a shortage of resources on the system, with
6350 ENOMEM, for example. A temporary error is now given on failure.
6352 29. Contrary to the C standard, it seems that in some environments, the
6353 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6354 program. Exim now does this explicitly; it affects the formatting of
6355 timestamps using strftime().
6357 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6358 complaints. I've now initialized all the variables, to avoid this.
6360 32. Header lines added by a system filter were not being "seen" during
6361 transport-time rewrites.
6363 33. The info_callback() function passed to OpenSSL is set up with type void
6364 (*)(SSL *, int, int), as described somewhere. However, when calling the
6365 function (actually a macro) that sets it up, the type void(*)() is
6366 expected. I've put in a cast to prevent warnings from picky compilers.
6368 34. If a DNS black list lookup found a CNAME record, but there were no A
6369 records associated with the domain it pointed at, Exim crashed.
6371 35. If a DNS black list lookup returned more than one A record, Exim ignored
6372 all but the first. It now scans all returned addresses if a particular IP
6373 value is being sought. In this situation, the contents of the
6374 $dnslist_value variable are a list of all the addresses, separated by a
6377 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6378 to accept a host name and all its aliases if the forward lookup for any of
6379 them yielded the IP address of the incoming connection. Now it accepts only
6380 those names whose forward lookup yields the correct IP address. Any other
6381 names are discarded. This closes a loophole whereby a rogue DNS
6382 administrator could create reverse DNS records to break through a
6383 wildcarded host restriction in an ACL.
6385 37. If a user filter or a system filter that ran in a subprocess used any of
6386 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6387 the wrong values were passed to the pipe command ($thisaddress had the
6388 value of $0, $0 had the value of $1, etc). This bug was introduced by
6389 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6391 38. Improved the line breaking for long SMTP error messages from ACLs.
6392 Previously, if there was no break point between 40 and 75 characters, Exim
6393 left the rest of the message alone. Two changes have been made: (a) I've
6394 reduced the minimum length to 35 characters; (b) if it can't find a break
6395 point between 35 and 75 characters, it looks ahead and uses the first one
6396 that it finds. This may give the occasional overlong line, but at least the
6397 remaining text gets split now.
6399 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6400 file descriptors might be low, and that setting 1000 would always raise it.
6401 It turns out that in some environments, the limit is already over 1000 and
6402 that lowering it causes trouble. So now Exim takes care not to decrease it.
6404 40. When delivering a message, the value of $return_path is set to $sender_
6405 address at the start of routing (routers may change the value). By an
6406 oversight, this default was not being set up when an address was tested by
6407 -bt or -bv, which affected the outcome if any router or filter referred to
6410 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6411 log only when "message" or "log_message" are set. However, if one of the
6412 conditions was an address verification, or a call to a nested ACL, the
6413 messages generated by the underlying test were being passed through. This
6414 no longer happens. The underlying message is available in $acl_verify_
6415 message for both "message" and "log_message" expansions, so it can be
6416 passed through if needed.
6418 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6419 new expansion $bh_ to give the encoded byte string without charset
6420 translation. Translation happens only if iconv() is available; HAVE_ICONV
6421 indicates this at build time. HEADERS_CHARSET gives the charset to
6422 translate to; headers_charset can change it in the configuration, and
6423 "headers charset" can change it in an individual filter file.
6425 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6426 that creates RFC 2047 encoded "words" labels them as that charset instead
6427 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6428 expansion operator; (ii) when Exim creates a From: line for a local
6429 message; (iii) when a header line is rewritten to include a "phrase" part.
6431 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6432 buggy, causing it to skip the first lines of messages whose message ID
6433 ended in 'D'. This would not have bitten before Exim release 4.14, because
6434 message IDs were unlikely to end in 'D' before then. The effect was to have
6435 incorrect size information for certain domains.
6437 45. #include "config.h" was missing at the start of the crypt16.c module. This
6438 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6441 46. If there was a timeout during a "random" callout check, Exim treated it as
6442 a failure of the random address, and carried on sending RSET and the real
6443 address. If the delay was just some slowness somewhere, the response to the
6444 original RCPT would be taken as a response to RSET and so on, causing
6445 mayhem of various kinds.
6447 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6448 when I implemented it. It didn't allow for the fact that some option values
6449 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6450 the right test for positive values.
6452 48. Domain names in DNS records are case-independent. Exim always looks them up
6453 in lower case. Some resolvers return domain names in exactly the case they
6454 appear in the zone file, that is, they may contain uppercase letters. Not
6455 all resolvers do this - some return always lower case. Exim was treating a
6456 change of case by a resolver as a change of domain, similar to a widening
6457 of a domain abbreviation. This triggered its re-routing code and so it was
6458 trying to route what was effectively the same domain again. This normally
6459 caused routing to fail (because the router wouldn't handle the domain
6460 twice). Now Exim checks for this case specially, and just changes the
6461 casing of the domain that it ultimately uses when it transmits the message
6464 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6467 50. If a filter generated a file delivery with a non-absolute name (possible if
6468 no home directory exists for the router), the forbid_file option was not
6471 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6472 the existing equality matching.
6474 52. Exim was using ints instead of ino_t variables in some places where it was
6475 dealing with inode numbers.
6477 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6478 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6479 called TMPDIR, and if it finds it is different, it changes its value.
6481 54. The smtp_printf() function is now made available to local_scan() so
6482 additional output lines can be written before returning. There is also an
6483 smtp_fflush() function to enable the detection of a dropped connection.
6484 The variables smtp_input and smtp_batched_input are exported to
6487 55. Changed the default runtime configuration: the message "Unknown user"
6488 has been removed from the ACL, and instead placed on the localuser router,
6489 using the cannot_route_message feature. This means that any verification
6490 failures that generate their own messages won't get overridden. Similarly,
6491 the "Unrouteable address" message that was in the ACL for unverifiable
6492 relay addresses has also been removed.
6494 56. Added hosts_avoid_esmtp to the smtp transport.
6496 57. The exicyclog script was not checking for the esoteric option
6497 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6498 will work only if exicyclog is run under the appropriate euid.
6500 58. Following a discussion on the list, the rules by which Exim recognises line
6501 endings on incoming messages have been changed. The -dropcr and drop_cr
6502 options are now no-ops, retained only for backwards compatibility. The
6503 following line terminators are recognized: LF CRLF CR. However, special
6504 processing applies to CR:
6506 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6507 nor a local message in the state where . is a terminator.
6509 (ii) If a bare CR is encountered in a header line, an extra space is added
6510 after the line terminator so as not to end the header. The reasoning
6511 behind this is that bare CRs in header lines are most likely either
6512 to be mistakes, or people trying to play silly games.
6514 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6515 was being incorrectly given as 18 bytes larger than it should have been.
6516 This is a VOB (very old bug).
6518 60. This may never have affected anything current, but just in case it has:
6519 When the local host is found other than at the start of a list of hosts,
6520 the local host, those with the same MX, and any that follow, are discarded.
6521 When the list in question was part of a longer list of hosts, the following
6522 hosts (not currently being processed) were also being discarded. This no
6523 longer happens. I'm not sure if this situation could ever has previously
6526 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6529 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6530 header. This is something that is recommended in a new Internet Draft, and
6531 is something that is documented as being done by Sendmail. There are two
6532 possible values. For messages generated by the autoreply transport, Exim
6535 Auto-Submitted: auto-replied
6537 whereas for all other generated messages (e.g. bounces) it adds
6539 Auto-Submitted: auto-generated
6541 63. The "personal" condition in filters now includes a test for the
6542 Auto-Submitted: header. If it contains the string "auto-" the message it
6543 not considered personal.
6545 64. Added rcpt_include_affixes as a generic transport option.
6547 65. Added queue_only_override (default true).
6549 66. Added the syslog_duplication option.
6551 67. If what should have been the first header line of a message consisted of
6552 a space followed by a colon, Exim was mis-interpreting it as a header line.
6553 It isn't of course - it is syntactically invalid and should therefore be
6554 treated as the start of the message body. The misbehaviour could have
6555 caused a number of strange effects, including loss of data in subsequent
6556 header lines, and spool format errors.
6558 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6559 client host had authenticated. This control can now be exercised by an ACL
6560 for more flexibility.
6562 69. By default, callouts do not happen when testing with -bh. There is now a
6563 variant, -bhc, which does actually run the callout code, including
6564 consulting and updating the callout cache.
6566 70. Added support for saslauthd authentication, courtesy of Alexander
6569 71. If statvfs() failed on the spool or log directories while checking their
6570 size for availability, Exim confusingly gave the error "space shortage".
6571 Furthermore, in debugging mode it crashed with a floating point exception.
6572 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6573 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6574 serious problem, Exim now writes to the main and panic logs when this
6575 happens, with details of the failure. It then refuses to accept the
6576 incoming message, giving the message "spool directory problem" or "log
6577 directory problem" with a 421 code for SMTP messages.
6579 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6580 0, 1, and 2 exist, because some OS complain for execs without them (see
6581 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6582 descriptors. However, the code omitted to check that the open succeeded,
6583 causing mysterious errors if for some reason the permissions on /dev/null
6584 got screwed. Now Exim writes a message to the main and panic logs, and
6585 bombs out if it can't open /dev/null.
6587 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6588 interact so that it is all more flexible. It is supposed to remain
6589 backwards compatible. Also added extra_local_interfaces.
6591 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6592 to bomb out with an assertion failure - to the client this appears as a
6593 connection drop. This problem occurs in the part of the code that was taken
6594 from the Samba project. Fortunately, the assertion is in a very simple
6595 function, so I have fixed this by reproducing the function inline in the
6596 one place where it is called, and arranging for authentication to fail
6597 instead of killing the process with assert().
6599 75. The SPA client code was not working when the server requested OEM rather
6600 than Unicode encoding.
6602 76. Added code to make require_files with a specific uid setting more usable in
6603 the case where statting the file as root fails - usually a non-root-mounted
6604 NFS file system. When this happens and the failure is EACCES, Exim now
6605 forks a subprocess and does the per-uid checking as the relevant uid.
6607 77. Added process_log_path.
6609 78. If log_file_path was not explicitly set, a setting of check_log_space or
6610 check_log_inodes was ignored.
6612 79. If a space check for the spool or log partitions fails, the incident is now
6613 logged. Of course, in the latter case the data may get lost...
6615 80. Added the %p formatting code to string_format() so that it can be used to
6616 print addresses in debug_print(). Adjusted all the address printing in the
6617 debugging in store.c to use %p rather than %d.
6619 81. There was a concern that a line of code in smtp_in.c could overflow a
6620 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6621 initially expressed, the concern was not well-founded, because trailing
6622 spaces are removed early. However, if the trailing spaces were followed by
6623 a NULL, they did not get removed, so the overflow was possible. Two fixes
6626 (a) I re-wrote the offending code in a cleaner fashion.
6627 (b) If an incoming SMTP command contains a NULL character, it is rejected
6630 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6631 runs initgroups(), so that if the Exim user is in any additional groups,
6632 they will be used during message reception.
6638 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob