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 selfsigned 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 sourcecode
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
118 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
119 and 3.4.4 - once the server is enabled to respond to an OCSP request
120 it does even when not requested, resulting in a stapling non-aware
121 client dropping the TLS connection.
123 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
124 support variable-length bit vectors. No functional change.
126 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
127 The I= interface field on outgoing lines is now after the H= remote
128 host field, same as incoming lines. There is a separate
129 outgoing_interface log selector which allows you to disable the
132 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
133 If not running log_selector +smtp_connection the mainlog would be held
134 open indefinitely after a "too many connections" event, including to a
135 deleted file after a log rotate. Leave the per net connection logging
136 leaving it open for efficiency as that will be quickly detected by the
137 check on the next write.
139 HS/01 Bug 1671: Fix post transport crash.
140 Processing the wait-<transport> messages could crash the delivery
141 process if the message IDs didn't exist for some reason. When
142 using 'split_spool_directory=yes' the construction of the spool
143 file name failed already, exposing the same netto behaviour.
145 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
146 mime_regex ACL conditions.
148 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
149 to DSN fail messages (bounces): remote IP, remote greeting, remote response
150 to HELO, local diagnostic string.
152 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
153 log line to debug. Even when configured with a tls authenticator many
154 client connections are expected to not authenticate in this way, so
155 an authenticate fail is not an error.
157 HS/02 Add the Exim version string to the process info. This way exiwhat
158 gives some more detail about the running daemon.
160 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
161 matter for fast-change records such as DNSBLs.
163 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
164 retry record, even if constant. There may be multiple transports with
165 different interface settings and the retry behaviour needs to be kept
168 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
170 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
172 JH/10 Bug 840: fix log_defer_output option of pipe transport
174 JH/11 Bug 830: use same host for all RCPTS of a message, even under
175 hosts_randomize. This matters a lot when combined with mua_wrapper.
177 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
178 ${quote_pgsql:<string>} operator.
180 JH/13 Bug 1708: avoid misaligned access in cached lookup.
182 JH/14 Change header file name for freeradius-client. Relevant if compiling
183 with Radius support; from the Gentoo tree and checked under Fedora.
185 JH/15 Bug 1712: Introduce $prdr_requested flag variable
187 JH/16 Bug 1714: Permit an empty string as expansion result for transport
188 option transport_filter, meaning no filtering.
190 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
192 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
193 defaults to "*" (all hosts). The variable is now available when not built
194 with TLS, default unset, mainly to enable keeping the testuite sane.
195 If a server certificate is not supplied (via tls_certificate) an error is
196 logged, and clients will find TLS connections fail on startup. Presumably
197 they will retry in-clear.
198 Packagers of Exim are strongly encouraged to create a server certificate
199 at installation time.
201 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
202 with the $config_file variable.
204 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
205 in transport context, after the attempt, and per-recipient. The latter type
206 is per host attempted. The event data is the error message, and the errno
207 information encodes the lookup type (A vs. MX) used for the (first) host,
208 and the trailing two digits of the smtp 4xx reponse.
210 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
211 to write to mainlog (or rejectlog, paniclog) in the window between file
212 creation and permissions/ownership being changed. Particularly affects
213 installations where exicyclog is run as root, rather than exim user;
214 result is that the running daemon panics and dies.
216 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
218 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
219 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
220 "pri" and "weight". Note that the previous implicit priority given by the
221 list order is no longer honoured.
223 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
226 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
227 by defining SUPPORT_SOCKS.
229 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
230 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
231 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
232 variables are renamed to proxy_{local,external}_{address,port}.
234 JH/27 Move Internationalisation support from Experimental to mainline, enabled
235 for a build by defining SUPPORT_I18N
237 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
238 of the query string, and make ${quote_redis:} do that quoting.
240 JH/29 Move Events support from Experimental to mainline, enabled by default
241 and removable for a build by defining DISABLE_EVENT.
243 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
245 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
246 cached by the daemon.
248 JH/32 Move Redis support from Experimental to mainline, enabled for a build
249 by defining LOOKUP_REDIS. The libhiredis library is required.
251 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
252 keys are given for lookup.
254 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
255 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
256 only supported when built with TLS support. The PolarSSL SHA routines
257 are still used when the TLS library is too old for convenient support.
259 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
260 openssl_options), for security. OpenSSL forces this from version 1.1.0
261 server-side so match that on older versions.
263 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
264 allocation for $value could be released as the expansion processing
265 concluded, but leaving the global pointer active for it.
267 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
268 and to use the domains and local_parts ACL conditions.
270 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
271 incorrectly not doubled on cutthrough transmission, hence seen as a
272 body-termination at the receiving system - resulting in truncated mails.
273 Commonly the sender saw a TCP-level error, and retransmitted the message
274 via the normal store-and-forward channel. This could result in duplicates
275 received - but deduplicating mailstores were liable to retain only the
276 initial truncated version.
278 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
280 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
282 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
283 we're in there, support oversigning also; bug 1309.
285 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
287 HS/04 Add support for keep_environment and add_environment options.
289 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
290 either intentional arithmetic overflow during PRNG, or testing config-
293 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
294 delivery resulted in actual delivery. Cancel cutthrough before DATA
297 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
298 rejects a recipient: pass the reject to the originator.
300 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
301 Many were false-positives and ignorable, but it's worth fixing the
304 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
305 for the new environment-manipulation done at startup. Move the routines
306 from being local to tls.c to being global via the os.c file.
308 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
309 an extract embedded as result-arg for a map, the first arg for extract
310 is unavailable so we cannot tell if this is a numbered or keyed
311 extraction. Accept either.
317 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
320 JH/02 The smtp transport option "multi_domain" is now expanded.
322 JH/03 The smtp transport now requests PRDR by default, if the server offers
325 JH/04 Certificate name checking on server certificates, when exim is a client,
326 is now done by default. The transport option tls_verify_cert_hostnames
327 can be used to disable this per-host. The build option
328 EXPERIMENTAL_CERTNAMES is withdrawn.
330 JH/05 The value of the tls_verify_certificates smtp transport and main options
331 default to the word "system" to access the system default CA bundle.
332 For GnuTLS, only version 3.0.20 or later.
334 JH/06 Verification of the server certificate for a TLS connection is now tried
335 (but not required) by default. The verification status is now logged by
336 default, for both outbound TLS and client-certificate supplying inbound
339 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
342 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
343 Status Notification (bounce) messages are now MIME format per RFC 3464.
344 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
345 under the control of the dsn_advertise_hosts option, and routers may
346 have a dsn_lasthop option.
348 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
349 default, modifiable by a malware= option. The list separator for
350 the options can now be changed in the usual way. Bug 68.
352 JH/10 The smtp_receive_timeout main option is now expanded before use.
354 JH/11 The incoming_interface log option now also enables logging of the
355 local interface on delivery outgoing connections.
357 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
358 if the interface and destination host and port all match.
360 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
363 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
364 Patch from Andrew Lewis.
366 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
367 now supports optional time-restrictions, weighting, and priority
368 modifiers per server. Patch originally by <rommer@active.by>.
370 JH/16 The spamd_address main option now supports a mixed list of local
371 and remote servers. Remote servers can be IPv6 addresses, and
372 specify a port-range.
374 JH/17 Bug 68: The spamd_address main option now supports an optional
375 timeout value per server.
377 JH/18 Bug 1581: Router and transport options headers_add/remove can
378 now have the list separator specified.
380 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
383 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
386 JH/21 Support for the A6 type of dns record is withdrawn.
388 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
389 rather than the verbs used.
391 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
392 from 255 to 1024 chars.
394 JH/24 Verification callouts now attempt to use TLS by default.
396 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
397 are generic router options now. The defaults didn't change.
399 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
400 Original patch from Alexander Shikoff, worked over by JH.
402 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
403 files if ZCAT_COMMAND is not executable.
405 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
407 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
409 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
410 Normally benign, it bites when the pair was led to by a CNAME;
411 modern usage is to not canoicalize the domain to a CNAME target
412 (and we were inconsistent anyway for A-only vs AAAA+A).
414 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
416 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
417 when evaluating $sender_host_dnssec.
419 JH/31 Check the HELO verification lookup for DNSSEC, adding new
420 $sender_helo_dnssec variable.
422 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
424 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
426 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
428 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
429 documented as working, but never had. Support all but $spam_report.
431 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
432 added for tls authenticator.
434 HS/03 Add perl_taintmode main config option
439 TL/01 When running the test suite, the README says that variables such as
440 no_msglog_check are global and can be placed anywhere in a specific
441 test's script, however it was observed that placement needed to be near
442 the beginning for it to behave that way. Changed the runtest perl
443 script to read through the entire script once to detect and set these
444 variables, reset to the beginning of the script, and then run through
445 the script parsing/test process like normal.
447 TL/02 The BSD's have an arc4random API. One of the functions to induce
448 adding randomness was arc4random_stir(), but it has been removed in
449 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
450 function when detected.
452 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
453 cause callback expansion.
455 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
456 syntax errors in an expansion can be treated as a string instead of
457 logging or causing an error, due to the internal use of bool_lax
458 instead of bool when processing it.
460 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
461 server certificates when making smtp deliveries.
463 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
465 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
467 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
469 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
470 Merged patch from Sebastian Wiedenroth.
472 JH/05 Fix results-pipe from transport process. Several recipients, combined
473 with certificate use, exposed issues where response data items split
474 over buffer boundaries were not parsed properly. This eventually
475 resulted in duplicates being sent. This issue only became common enough
476 to notice due to the introduction of conection certificate information,
477 the item size being so much larger. Found and fixed by Wolfgang Breyha.
479 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
480 size buffer was used, resulting in syntax errors when an expansion
483 JH/07 Add support for directories of certificates when compiled with a GnuTLS
484 version 3.3.6 or later.
486 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
487 is EXPERIMENTAL_EVENT, the main-configuration and transport options
488 both become "event_action", the variables become $event_name, $event_data
489 and $event_defer_errno. There is a new variable $verify_mode, usable in
490 routers, transports and related events. The tls:cert event is now also
491 raised for inbound connections, if the main configuration event_action
494 TL/06 In test suite, disable OCSP for old versions of openssl which contained
495 early OCSP support, but no stapling (appears to be less than 1.0.0).
497 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
498 server certificate names available under the smtp transport option
499 "tls_verify_cert_hostname" now do not permit multi-component wildcard
502 JH/10 Time-related extraction expansions from certificates now use the main
503 option "timezone" setting for output formatting, and are consistent
504 between OpenSSL and GnuTLS compilations. Bug 1541.
506 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
507 encoded parameter in the incoming message. Bug 1558.
509 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
510 include certificate info, eximon was claiming there were spoolfile
513 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
515 JH/14 Log delivery-related information more consistently, using the sequence
516 "H=<name> [<ip>]" wherever possible.
518 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
519 are problematic for Debian distribution, omit them from the release
522 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
524 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
527 JH/17 Fix a null-indirection in certextract expansions when a nondefault
528 output list separator was used.
533 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
534 checkers that were complaining about end of non-void function with no
537 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
538 This was a regression intruduced in 4.83 by another bugfix.
540 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
542 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
543 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
549 TF/01 Correctly close the server side of TLS when forking for delivery.
551 When a message was received over SMTP with TLS, Exim failed to clear up
552 the incoming connection properly after forking off the child process to
553 deliver the message. In some situations the subsequent outgoing
554 delivery connection happened to have the same fd number as the incoming
555 connection previously had. Exim would try to use TLS and fail, logging
556 a "Bad file descriptor" error.
558 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
559 utilities have not been installed.
561 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
562 temporary space as the ACL may create new global variables.
564 TL/01 LDAP support uses per connection or global context settings, depending
565 upon the detected version of the libraries at build time.
567 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
568 to extract and use the src ip:port in logging and expansions as if it
569 were a direct connection from the outside internet. PPv2 support was
570 updated based on HAProxy spec change in May 2014.
572 JH/02 Add ${listextract {number}{list}{success}{fail}}.
574 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
575 Properly escape header and check for NULL return.
577 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
580 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
582 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
583 characters in header names, implemented as a verify condition.
584 Contributed by Michael Fischer v. Mollard.
586 TL/05 Rename SPF condition results err_perm and err_temp to standardized
587 results permerror and temperror. Previous values are deprecated but
588 still accepted. In a future release, err_perm and err_temp will be
589 completely removed, which will be a backward incompatibility if the
590 ACL tests for either of these two old results. Patch contributed by
591 user bes-internal on the mailing list.
593 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
595 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
596 selectors, in both main and reject logs.
598 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
601 JH/07 Add malware type "sock" for talking to simple daemon.
603 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
605 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
606 routers/transports under cutthrough routing.
608 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
609 numbers. Touch up "bool" conditional to keep the same definition.
611 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
613 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
615 JH/12 Expand items in router/transport headers_add or headers_remove lists
616 individually rather than the list as a whole. Bug 1452.
618 Required for reasonable handling of multiple headers_ options when
619 they may be empty; requires that headers_remove items with embedded
620 colons must have them doubled (or the list-separator changed).
622 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
623 view the policy declared in the DMARC record. Currently, $dmarc_status
624 is a combined value of both the record presence and the result of the
627 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
629 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
630 dnslookup router and the smtp transport (applying to the forward
633 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
634 of ldap servers used for a specific lookup. Patch provided by Heiko
637 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
638 New variable $lookup_dnssec_authenticated for observability.
640 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
641 Patch submitted by Lars Timman.
643 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
645 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
646 Requires trusted mode and valid format message id, aborts otherwise.
647 Patch contributed by Heiko Schlichting.
649 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
650 certextract with support for various fields. Bug 1358.
652 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
653 is requested by default, modifiable by smtp transport option
656 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
657 operate on certificate variables to give certificate fingerprints
658 Also new ${sha256:cert_variable}.
660 JH/23 The PRDR feature is moved from being Experimental into the mainline.
662 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
663 Christian Aistleitner.
665 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
667 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
668 file. Patch from Wolfgang Breyha.
670 JH/25 Expand the coverage of the delivery $host and $host_address to
671 client authenticators run in verify callout. Bug 1476.
673 JH/26 Port service names are now accepted for tls_on_connect_ports, to
674 align with daemon_smtp_ports. Bug 72.
676 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
677 support and error reporting did not work properly.
679 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
680 and is readable. Patch from Andrew Colin Kissa.
682 TL/14 Enhance documentation of ${run expansion and how it parses the
683 commandline after expansion, particularly in the case when an
684 unquoted variable expansion results in an empty value.
686 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
688 PP/02 Fix internal collision of T_APL on systems which support RFC3123
689 by renaming away from it. Addresses GH issue 15, reported by
692 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
694 TL/15 SECURITY: prevent double expansion in math comparison functions
695 (can expand unsanitized data). Not remotely exploitable.
702 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
704 PP/02 Make -n do something, by making it not do something.
705 When combined with -bP, the name of an option is not output.
707 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
710 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
711 $sender_host_name and config options to manage this, and basic check
714 PP/05 DSCP support for outbound connections and control modifier for inbound.
716 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
717 (Only plugin which currently uses this is kerberos4, which nobody should
718 be using, but we should make it available and other future plugins might
719 conceivably use it, even though it would break NAT; stuff *should* be
720 using channel bindings instead).
722 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
723 name; added for Sendmail compatibility; requires admin caller.
724 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
725 just ignore it); requires trusted caller.
726 Also parse but ignore: -Ac -Am -X<logfile>
729 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
731 TL/02 Add +smtp_confirmation as a default logging option.
733 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
734 Patch by Magnus Holmgren from 2007-02-20.
736 TL/04 Bugzilla 1281 - Spec typo.
737 Bugzilla 1283 - Spec typo.
738 Bugzilla 1290 - Spec grammar fixes.
740 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
742 TL/06 Add Experimental DMARC support using libopendmarc libraries.
744 TL/07 Fix an out of order global option causing a segfault. Reported to dev
745 mailing list by by Dmitry Isaikin.
747 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
749 JH/02 Support "G" suffix to numbers in ${if comparisons.
751 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
753 NM/01 Bugzilla 1197 - Spec typo
754 Bugzilla 1196 - Spec examples corrections
756 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
758 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
759 gnutls_enable_pkcs11, but renamed to more accurately indicate its
762 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
763 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
765 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
766 "acl {{name}{arg}...}", and optional args on acl condition
769 JH/05 Permit multiple router/transport headers_add/remove lines.
771 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
773 JH/07 Avoid using a waiting database for a single-message-only transport.
774 Performance patch from Paul Fisher. Bugzilla 1262.
776 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
779 JH/09 Add $headers_added variable, with content from use of ACL modifier
780 add_header (but not yet added to the message). Bugzilla 199.
782 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
783 Pulled from Bugzilla 817 by Wolfgang Breyha.
785 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
787 (nb: this is the same fix as in Exim 4.80.1)
789 JH/11 Add A= logging on delivery lines, and a client_set_id option on
792 JH/12 Add optional authenticated_sender logging to A= and a log_selector
795 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
797 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
798 advertise SMTP AUTH mechanism to us, instead of a generic
799 protocol violation error. Also, make Exim more robust to bad
800 data from the Dovecot auth socket.
802 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
804 When a queue runner is handling a message, Exim first routes the
805 recipient addresses, during which it prunes them based on the retry
806 hints database. After that it attempts to deliver the message to
807 any remaining recipients. It then updates the hints database using
810 So if a recipient address works intermittently, it can get repeatedly
811 deferred at routing time. The retry hints record remains fresh so the
812 address never reaches the final cutoff time.
814 This is a fairly common occurrence when a user is bumping up against
815 their storage quota. Exim had some logic in its local delivery code
816 to deal with this. However it did not apply to per-recipient defers
817 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
819 This change adds a proper retry rule check during routing so that the
820 final cutoff time is checked against the message's age. We only do
821 this check if there is an address retry record and there is not a
822 domain retry record; this implies that previous attempts to handle
823 the address had the retry_use_local_parts option turned on. We use
824 this as an approximation for the destination being like a local
825 delivery, as in LMTP.
827 I suspect this new check makes the old local delivery cutoff check
828 redundant, but I have not verified this so I left the code in place.
830 TF/02 Correct gecos expansion when From: is a prefix of the username.
832 Test 0254 submits a message to Exim with the header
836 When I ran the test suite under the user fanf2, Exim expanded
837 the header to contain my full name, whereas it should have added
838 a Resent-Sender: header. It erroneously treats any prefix of the
839 username as equal to the username.
841 This change corrects that bug.
843 GF/01 DCC debug and logging tidyup
844 Error conditions log to paniclog rather than rejectlog.
845 Debug lines prefixed by "DCC: " to remove any ambiguity.
847 TF/03 Avoid unnecessary rebuilds of lookup-related code.
849 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
850 Bug spotted by Jeremy Harris; was flawed since initial commit.
851 Would have resulted in OCSP responses post-SNI triggering an Exim
852 NULL dereference and crash.
854 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
856 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
857 Bug detection, analysis and fix by Samuel Thibault.
858 Bugzilla 1331, Debian bug #698092.
860 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
862 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
863 Server implementation by Todd Lyons, client by JH.
864 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
865 config variable "prdr_enable" controls whether the server
866 advertises the facility. If the client requests PRDR a new
867 acl_data_smtp_prdr ACL is called once for each recipient, after
868 the body content is received and before the acl_smtp_data ACL.
869 The client is controlled by bolth of: a hosts_try_prdr option
870 on the smtp transport, and the server advertisement.
871 Default client logging of deliveries and rejections involving
872 PRDR are flagged with the string "PRDR".
874 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
875 fclose(). Diagnosis by Todd Lyons.
877 PP/17 Update configure.default to handle IPv6 localhost better.
878 Patch by Alain Williams (plus minor tweaks).
881 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
882 This is now consistent with GnuTLS, and is now documented: the
883 previous undocumented portable approach to treating the option as
884 unset was to force an expansion failure. That still works, and
885 an empty string is now equivalent.
887 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
888 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
889 not performing validation itself.
891 PP/20 Added force_command boolean option to pipe transport.
892 Patch from Nick Koston, of cPanel Inc.
894 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
897 TF/04 Added udpsend ACL modifer and hexquote expansion operator
899 PP/21 Fix eximon continuous updating with timestamped log-files.
900 Broken in a format-string cleanup in 4.80, missed when I repaired the
901 other false fix of the same issue.
902 Report and fix from Heiko Schlichting.
905 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
906 Report from Prashanth Katuri.
908 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
909 It's SecureTransport, so affects any MacOS clients which use the
910 system-integrated TLS libraries, including email clients.
912 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
913 using a MIME ACL for non-SMTP local injection.
914 Report and assistance in diagnosis by Warren Baker.
916 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
918 JH/16 Fix comparisons for 64b. Bugzilla 1385.
920 TL/09 Add expansion variable $authenticated_fail_id to keep track of
921 last id that failed so it may be referenced in subsequent ACL's.
923 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
926 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
927 ldap library initialization, allowing self-signed CA's to be
928 used. Also properly sets require_cert option later in code by
929 using NULL (global ldap config) instead of ldap handle (per
930 session). Bug diagnosis and testing by alxgomz.
932 TL/12 Enhanced documentation in the ratelimit.pl script provided in
933 the src/util/ subdirectory.
935 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
936 renamed to Transport Post Delivery Action by Jeremy Harris, as
939 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
940 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
941 redis_servers = needs to be configured which will be used by the redis
942 lookup. Patch from Warren Baker, of The Packet Hub.
944 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
946 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
947 hostname or reverse DNS when processing a host list. Used suggestions
948 from multiple comments on this bug.
950 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
952 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
953 Missed a few lines, added it to make the runtest require no keyboard
956 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
957 contains upper case chars. Make router use caseful_local_part.
959 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
960 support when GnuTLS has been built with p11-kit.
966 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
968 This, or similar/improved, will also be change PP/11 of 4.82.
974 PP/01 Handle short writes when writing local log-files.
975 In practice, only affects FreeBSD (8 onwards).
976 Bugzilla 1053, with thanks to Dmitry Isaikin.
978 NM/01 Bugzilla 949 - Documentation tweak
980 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
983 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
985 PP/02 Implemented gsasl authenticator.
987 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
989 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
990 `pkg-config foo` for cflags/libs.
992 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
993 with rest of GSASL and with heimdal_gssapi.
995 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
996 `pkg-config foo` for cflags/libs for the TLS implementation.
998 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
999 properties get this fed in as external SSF. A number of robustness
1000 and debugging improvements to the cyrus_sasl authenticator.
1002 PP/08 cyrus_sasl server now expands the server_realm option.
1004 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1005 Patch by Jeremy Harris.
1007 PP/10 Added dbmjz lookup type.
1009 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1011 PP/12 MAIL args handles TAB as well as SP, for better interop with
1012 non-compliant senders.
1013 Analysis and variant patch by Todd Lyons.
1015 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1016 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1017 Patch from Dirk Mueller <dmueller@suse.com>
1019 PP/13 tls_peerdn now print-escaped for spool files.
1020 Observed some $tls_peerdn in wild which contained \n, which resulted
1021 in spool file corruption.
1023 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1024 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1025 or write after TLS renegotiation, which otherwise led to messages
1028 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1029 as a tracking header (ie: a signed header comes before the signature).
1030 Patch from Wolfgang Breyha.
1032 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1033 comma-sep list; embedded commas doubled.
1035 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1037 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1039 Report and patch from Dmitry Banschikov.
1041 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1042 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1043 failures. We appear to now support TLS1.1+ with Exim.
1045 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1046 lets Exim select keys and certificates based upon TLS SNI from client.
1047 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1048 before an outbound SMTP session. New log_selector, +tls_sni.
1050 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1051 NULL dereference. Report and patch from Alun Jones.
1053 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1054 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1055 Not seeing resolver debug output on NetBSD, but suspect this is a
1056 resolver implementation change.
1058 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1059 Left warnings. Added "eximon gdb" invocation mode.
1061 PP/21 Defaulting "accept_8bitmime" to true, not false.
1063 PP/22 Added -bw for inetd wait mode support.
1065 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1066 locate the relevant includes and libraries. Made this the default.
1068 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1069 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1071 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1072 This may cause build issues on older platforms.
1074 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1075 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1076 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1077 Added SNI support via GnuTLS too.
1078 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1080 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1082 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1083 Applied second patch from Janne, implementing suggestion to default
1084 multiple-strings-in-record handling to match SPF spec.
1086 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1088 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1089 read-only, out of scope).
1090 Patch from Wolfgang Breyha, report from Stuart Northfield.
1092 PP/29 Fix three issues highlighted by clang analyser static analysis.
1093 Only crash-plausible issue would require the Cambridge-specific
1094 iplookup router and a misconfiguration.
1095 Report from Marcin Mirosław.
1097 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1099 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1100 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1101 As part of this, removing so much warning spew let me fix some minor
1102 real issues in debug logging.
1104 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1105 assignment on my part. Fixed.
1107 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1108 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1109 Janne Snabb (who went above and beyond: thank you).
1111 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1112 string otherwise requires a connection and a bunch more work and it's
1113 relatively easy to get wrong. Should also expose TLS library linkage
1116 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1117 64-bit ${eval} (JH/03).
1119 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1120 GNU libc to support some of the 64-bit stuff, should not lead to
1121 conflicts. Defined before os.h is pulled in, so if a given platform
1122 needs to override this, it can.
1124 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1125 protection layer was required, which is not implemented.
1126 Bugzilla 1254, patch from Wolfgang Breyha.
1128 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1129 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1130 tls_dhparam take prime identifiers. Also unbreak combination of
1131 OpenSSL+DH_params+TLSSNI.
1133 PP/39 Disable SSLv2 by default in OpenSSL support.
1139 PP/01 Solaris build fix for Oracle's LDAP libraries.
1140 Bugzilla 1109, patch from Stephen Usher.
1142 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1144 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1147 TF/02 Fix a couple more cases where we did not log the error message
1148 when unlink() failed. See also change 4.74-TF/03.
1150 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1151 lock up or crash if it happened to be inside a call to libc when it
1152 got a SIGUSR1 from exiwhat.
1154 The SIGUSR1 handler appends the current process status to the process
1155 log which is later printed by exiwhat. It used to use the general
1156 purpose logging code to do this, but several functions it calls are
1157 not safe for signals.
1159 The new output code in the SIGUSR1 handler is specific to the process
1160 log, and simple enough that it's easy to inspect for signal safety.
1161 Removing some special cases also simplifies the general logging code.
1162 Removing the spurious timestamps from the process log simplifies
1165 TF/04 Improved ratelimit ACL condition.
1167 The /noupdate option has been deprecated in favour of /readonly which
1168 has clearer semantics. The /leaky, /strict, and /readonly update modes
1169 are mutually exclusive. The update mode is no longer included in the
1170 database key; it just determines when the database is updated. (This
1171 means that when you upgrde Exim will forget old rate measurements.)
1173 Exim now checks that the per_* options are used with an update mode that
1174 makes sense for the current ACL. For example, when Exim is processing a
1175 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1176 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1177 must specify per_mail/readonly. If you omit the update mode it defaults to
1178 /leaky where that makes sense (as before) or /readonly where required.
1180 The /noupdate option is now undocumented but still supported for
1181 backwards compatibility. It is equivalent to /readonly except that in
1182 ACLs where /readonly is required you may specify /leaky/noupdate or
1183 /strict/noupdate which are treated the same as /readonly.
1185 A useful new feature is the /count= option. This is a generalization
1186 of the per_byte option, so that you can measure the throughput of other
1187 aggregate values. For example, the per_byte option is now equivalent
1188 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1190 The per_rcpt option has been generalized using the /count= mechanism
1191 (though it's more complicated than the per_byte equivalence). When it is
1192 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1193 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1194 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1195 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1196 Note that using per_rcpt with a non-readonly update mode in more than
1197 one ACL will cause the recipients to be double-counted. (The per_mail
1198 and per_byte options don't have this problem.)
1200 The handling of very low rates has changed slightly. If the computed rate
1201 is less than the event's count (usually one) then this event is the first
1202 after a long gap. In this case the rate is set to the same as this event's
1203 count, so that the first message of a spam run is counted properly.
1205 The major new feature is a mechanism for counting the rate of unique
1206 events. The new per_addr option counts the number of different
1207 recipients that someone has sent messages to in the last time period. It
1208 behaves like per_rcpt if all the recipient addresses are different, but
1209 duplicate recipient addresses do not increase the measured rate. Like
1210 the /count= option this is a general mechanism, so the per_addr option
1211 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1212 example, measure the rate that a client uses different sender addresses
1213 with the options per_mail/unique=$sender_address. There are further
1214 details in the main documentation.
1216 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1218 TF/06 Removed a few PCRE remnants.
1220 TF/07 Automatically extract Exim's version number from tags in the git
1221 repository when doing development or release builds.
1223 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1224 Bugzilla 879. Patch from Paul Fisher.
1226 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1227 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1230 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1232 PP/05 Variable $av_failed, true if the AV scanner deferred.
1233 Bugzilla 1078. Patch from John Horne.
1235 PP/06 Stop make process more reliably on build failure.
1236 Bugzilla 1087. Patch from Heiko Schlittermann.
1238 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1239 Bugzilla 1089. Patch from Heiko Schlittermann.
1241 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1242 Bugzilla 1131. Patch from Holger Weiß.
1244 PP/09 Handle IPv6 addresses with SPF.
1245 Bugzilla 860. Patch from Wolfgang Breyha.
1247 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1249 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1252 PP/11 match_* no longer expand right-hand-side by default.
1253 New compile-time build option, EXPAND_LISTMATCH_RHS.
1254 New expansion conditions, "inlist", "inlisti".
1256 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1258 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1260 PP/14 fix log_write() format string regression from TF/03.
1261 Bugzilla 1152. Patch from Dmitry Isaikin.
1267 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1269 PP/02 Harmonised TLS library version reporting; only show if debugging.
1270 Layout now matches that introduced for other libraries in 4.74 PP/03.
1272 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1274 PP/04 New "dns_use_edns0" global option.
1276 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1279 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1280 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1282 TK/01 Updated PolarSSL code to 0.14.2.
1283 Bugzilla 1097. Patch from Andreas Metzler.
1285 PP/07 Catch divide-by-zero in ${eval:...}.
1286 Fixes bugzilla 1102.
1288 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1291 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1292 format-string attack -- SECURITY: remote arbitrary code execution.
1294 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1295 time unintentionally subject to list matching rules, letting the header
1296 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1297 arbitrary string expansion). This allowed for information disclosure.
1299 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1300 INT_MIN/-1 -- value coerced to INT_MAX.
1306 NM/01 Workround for PCRE version dependency in version reporting
1309 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1310 This fixes portability to compilers other than gcc, notably
1311 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1313 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1314 makefiles for portability to HP-UX and POSIX correctness.
1316 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1317 Also via indented variable definition in the Makefile.
1318 (Debugging by Oliver Heesakkers).
1320 PP/02 Restore caching of spamd results with expanded spamd_address.
1321 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1323 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1324 Improves build reliability. Fix from: Frank Elsner
1326 NM/02 Fix wide character breakage in the rfc2047 coding
1327 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1329 NM/03 Allow underscore in dnslist lookups
1330 Fixes bug 1026. Patch from Graeme Fowler
1332 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1333 Code patches from Adam Ciarcinski of NetBSD.
1335 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1338 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1339 is logged, to avoid truncation. Patch from John Horne.
1341 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1342 Patch from Jakob Hirsch.
1344 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1345 SQL string expansion failure details.
1346 Patch from Andrey Oktyabrski.
1348 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1349 Patch from Simon Arlott.
1351 PP/09 New lookups functionality failed to compile on old gcc which rejects
1352 extern declarations in function scope.
1353 Patch from Oliver Fleischmann
1355 PP/10 Use sig_atomic_t for flags set from signal handlers.
1356 Check getgroups() return and improve debugging.
1357 Fixed developed for diagnosis in bug 927 (which turned out to be
1360 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1361 Patch from Mark Zealey.
1363 PP/12 Bugzilla 1056: Improved spamd server selection.
1364 Patch from Mark Zealey.
1366 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1367 Based on patch from Heiko Schlittermann.
1369 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1370 Patch from Uwe Doering, sign-off by Michael Haardt.
1372 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1373 variable declaration deep within a block. Bug and patch from
1376 PP/15 lookups-Makefile IRIX compatibilty coercion.
1378 PP/16 Make DISABLE_DKIM build knob functional.
1380 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1381 Patch by Simon Arlott
1383 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1384 variable argument macros. Our copy now differs from upstream.
1390 TF/01 Failure to get a lock on a hints database can have serious
1391 consequences so log it to the panic log.
1393 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1394 controlled using the smtp_confirmation log selector.
1396 TF/03 Include the error message when we fail to unlink a spool file.
1398 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1399 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1400 for maintaining out-of-tree patches for some time.
1402 PP/01 Bugzilla 139: Documentation and portability issues.
1403 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1404 Handle per-OS dynamic-module compilation flags.
1406 PP/02 Let /dev/null have normal permissions.
1407 The 4.73 fixes were a little too stringent and complained about the
1408 permissions on /dev/null. Exempt it from some checks.
1409 Reported by Andreas M. Kirchwitz.
1411 PP/03 Report version information for many libraries, including
1412 Exim version information for dynamically loaded libraries. Created
1413 version.h, now support a version extension string for distributors
1414 who patch heavily. Dynamic module ABI change.
1416 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1417 privilege escalation vulnerability whereby the Exim run-time user
1418 can cause root to append content of the attacker's choosing to
1421 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1424 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1425 If dropping privileges for untrusted macros, we disabled normal logging
1426 on the basis that it would fail; for the Exim run-time user, this is not
1427 the case, and it resulted in successful deliveries going unlogged.
1428 Fixed. Reported by Andreas Metzler.
1434 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1435 only prepend for the Resent-* case. Fixes regression introduced in
1436 Exim 4.70 by NM/22 for Bugzilla 607.
1438 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1439 increasing numbers of administrators be bitten by this.
1441 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1443 PP/03 Bugzilla 994: added openssl_options main configuration option.
1445 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1447 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1449 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1451 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1452 without a peer certificate, leading to a segfault because of an
1453 assumption that peers always have certificates. Be a little more
1454 paranoid. Problem reported by Martin Tscholak.
1456 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1457 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1458 NB: ClamAV planning to remove STREAM in "middle of 2010".
1459 CL also introduces -bmalware, various -d+acl logging additions and
1460 more caution in buffer sizes.
1462 PP/09 Implemented reverse_ip expansion operator.
1464 PP/10 Bugzilla 937: provide a "debug" ACL control.
1466 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1468 PP/12 Bugzilla 973: Implement --version.
1470 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1472 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1474 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1476 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1477 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1478 ignore trailing whitespace.
1480 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1482 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1485 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1486 Notification from Dr Andrew Aitchison.
1488 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1489 ExtendedDetectionInfo response format.
1490 Notification from John Horne.
1492 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1495 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1496 XSL and documented dependency on system catalogs, with examples of how
1499 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1502 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1503 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1504 configuration file which is writeable by the Exim user or group.
1506 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1507 of configuration files to cover files specified with the -C option if
1508 they are going to be used with root privileges, not just the default
1511 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1512 option (effectively making it always true).
1514 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1515 files to be used while preserving root privileges.
1517 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1518 that rogue child processes cannot use them.
1520 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1521 run-time user, instead of root.
1523 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1524 Exim run-time user without dropping privileges.
1526 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1527 result string, instead of calling string_vformat() twice with the same
1530 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1531 for other users. Others should always drop root privileges if they use
1532 -C on the command line, even for a whitelisted configure file.
1534 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1536 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1542 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1543 $data_path, and $header_path variables; fixed documentation bugs and
1546 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1547 exipick to access non-standard spools, including the "frozen" queue
1550 NM/01 Bugzilla 965: Support mysql stored procedures.
1551 Patch from Alain Williams
1553 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1555 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1556 Patch from Andreas Metzler
1558 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1559 Patch from Kirill Miazine
1561 NM/05 Bugzilla 671: Added umask to procmail example.
1563 JJ/03 installed exipick 20100323.0, fixing doc bug
1565 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1566 directory. Notification and patch from Dan Rosenberg.
1568 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1570 TK/02 Improve log output when DKIM signing operation fails.
1572 MH/01 Treat the transport option dkim_domain as a colon separated
1573 list, not as a single string, and sign the message with each element,
1574 omitting multiple occurences of the same signer.
1576 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1577 Bugzilla 985, 986. Patch by Simon Arlott
1579 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1580 Patch by Simon Arlott
1582 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1583 MBX locking. Notification from Dan Rosenberg.
1589 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1591 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1593 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1595 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1597 NM/04 Bugzilla 915: Flag broken perl installation during build.
1603 TK/01 Added patch by Johannes Berg that expands the main option
1604 "spamd_address" if it starts with a dollar sign.
1606 TK/02 Write list of recipients to X-Envelope-Sender header when building
1607 the mbox-format spool file for content scanning (suggested by Jakob
1610 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1611 (http://www.dcc-servers.net/) support via dccifd. Activated by
1612 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1614 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1615 by Mark Daniel Reidel <mr@df.eu>.
1617 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1618 When building exim an external PCRE library is now needed -
1619 PCRE is a system library on the majority of modern systems.
1620 See entry on PCRE_LIBS in EDITME file.
1622 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1623 conversation. Added nologin parameter to request.
1624 Patch contributed by Kirill Miazine.
1626 TF/01 Do not log submission mode rewrites if they do not change the address.
1628 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1630 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1631 log files in place. Contributed by Roberto Lima.
1633 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1635 TF/03 Bugzilla 615: When checking the local_parts router precondition
1636 after a local_part_suffix or local_part_prefix option, Exim now
1637 does not use the address's named list lookup cache, since this
1638 contains cached lookups for the whole local part.
1640 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1641 Robert Millan. Documentation is in experimental-spec.txt.
1643 TF/04 Bugzilla 668: Fix parallel build (make -j).
1645 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1647 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1648 Patch provided by Jan Srzednicki.
1650 TF/05 Leading white space used to be stripped from $spam_report which
1651 wrecked the formatting. Now it is preserved.
1653 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1654 that they are available at delivery time.
1656 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1658 TF/08 TLS error reporting now respects the incoming_interface and
1659 incoming_port log selectors.
1661 TF/09 Produce a more useful error message if an SMTP transport's hosts
1662 setting expands to an empty string.
1664 NM/06 Bugzilla 744: EXPN did not work under TLS.
1665 Patch provided by Phil Pennock.
1667 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1668 Patch provided by Richard Godbee.
1670 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1671 acl_smtp_notquit, added index entry.
1673 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1674 Patch provided by Eugene Bujak.
1676 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1677 accept(). Patch provided by Maxim Dounin.
1679 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1680 Patch provided by Phil Pennock.
1682 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1684 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1685 Patch provided by Brad "anomie" Jorsch.
1687 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1688 Patch provided by Dean Brooks.
1690 TK/05 Add native DKIM support (does not depend on external libraries).
1692 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1693 Patch provided by Graeme Fowler.
1695 NM/16 Bugzilla 851: Documentation example syntax fix.
1697 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1699 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1702 NM/19 Bugzilla 745: TLS version reporting.
1703 Patch provided by Phil Pennock.
1705 NM/20 Bugzilla 167: bool: condition support.
1706 Patch provided by Phil Pennock.
1708 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1709 clients. Patch provided by Phil Pennock.
1711 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1712 Patch provided by Brad "anomie" Jorsch.
1714 NM/23 Bugzilla 687: Fix misparses in eximstats.
1715 Patch provided by Heiko Schlittermann.
1717 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1718 Patch provided by Heiko Schlittermann.
1720 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1721 plus update to original patch.
1723 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1725 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1726 Patch provided by David Brownlee.
1728 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1730 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1732 NM/30 Bugzilla 888: TLS documentation bugfixes.
1734 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1736 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1737 Unlike the original bugzilla I have changed all shell scripts in src tree.
1739 NM/33 Bugzilla 898: Transport filter timeout fix.
1740 Patch by Todd Rinaldo.
1742 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1743 Patch by Serge Demonchaux.
1745 NM/35 Bugzilla 39: Base64 decode bug fixes.
1746 Patch by Jakob Hirsch.
1748 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1750 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1752 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1754 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1760 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1761 ALT-N's libdkim that I have put here:
1762 http://duncanthrax.net/exim-experimental/
1764 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1765 were called 'true' and it seems that C99 defines that as a reserved
1766 keyword to be used with 'bool' variable types. That means you could
1767 not include C99-style headers which use bools without triggering
1768 build errors in sieve.c.
1770 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1771 as mailq or other aliases. Changed the --help handling significantly
1772 to do whats expected. exim_usage() emits usage/help information.
1774 SC/01 Added the -bylocaldomain option to eximstats.
1776 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1778 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1780 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1786 PH/01 Another patch from the Sieve maintainer.
1788 PH/02 When an IPv6 address is converted to a string for single-key lookup
1789 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1790 dots are used instead of colons so that keys in lsearch files need not
1791 contain colons. This was done some time before quoting was made available
1792 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1793 (notated using the quote facility) so as to distinguish them from IPv4
1794 keys. This meant that lookups for IP addresses in host lists did not work
1795 for iplsearch lookups.
1797 This has been fixed by arranging for IPv6 addresses to be expressed with
1798 colons if the lookup type is iplsearch. This is not incompatible, because
1799 previously such lookups could never work.
1801 The situation is now rather anomolous, since one *can* have colons in
1802 ordinary lsearch keys. However, making the change in all cases is
1803 incompatible and would probably break a number of configurations.
1805 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1808 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1809 conversion specification without a maximum field width, thereby enabling
1810 a rogue spamd server to cause a buffer overflow. While nobody in their
1811 right mind would setup Exim to query an untrusted spamd server, an
1812 attacker that gains access to a server running spamd could potentially
1813 exploit this vulnerability to run arbitrary code as the Exim user.
1815 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1816 $primary_hostname instead of what libspf2 thinks the hosts name is.
1818 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1819 a directory entry by the name of the lookup key. Previously, if a
1820 symlink pointed to a non-existing file or a file in a directory that
1821 Exim lacked permissions to read, a lookup for a key matching that
1822 symlink would fail. Now it is enough that a matching directory entry
1823 exists, symlink or not. (Bugzilla 503.)
1825 PH/03 The body_linecount and body_zerocount variables are now exported in the
1828 PH/04 Added the $dnslist_matched variable.
1830 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1831 This means they are set thereafter only if the connection becomes
1834 PH/06 Added the client_condition to authenticators so that some can be skipped
1835 by clients under certain conditions.
1837 PH/07 The error message for a badly-placed control=no_multiline_responses left
1838 "_responses" off the end of the name.
1840 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1842 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1843 (without spaces) instead of just copying the configuration text.
1845 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1847 PH/11 Added $max_received_linelength.
1849 PH/12 Added +ignore_defer and +include_defer to host lists.
1851 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1852 way in which PCRE > 7.0 is built.
1854 PH/14 Implemented queue_only_load_latch.
1856 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1857 MAIL command. The effect was to mangle the value on 64-bit systems.
1859 PH/16 Another patch from the Sieve maintainer.
1861 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1863 PH/18 If a system quota error occurred while trying to create the file for
1864 a maildir delivery, the message "Mailbox is full" was not appended to the
1865 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1866 only to a quota excession during the actual writing of the file.
1868 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1869 characters?) which causes problems in log lines. The DN values are now
1870 passed through string_printing() before being added to log lines.
1872 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1873 and InterBase are left for another time.)
1875 PH/21 Added message_body_newlines option.
1877 PH/22 Guard against possible overflow in moan_check_errorcopy().
1879 PH/23 POSIX allows open() to be a macro; guard against that.
1881 PH/24 If the recipient of an error message contained an @ in the local part
1882 (suitably quoted, of course), incorrect values were put in $domain and
1883 $local_part during the evaluation of errors_copy.
1889 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1890 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1893 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1894 issue a MAIL command.
1896 PH/02 In an ACL statement such as
1898 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1900 if a client was not listed at all, or was listed with a value other than
1901 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1902 the condition was not true (as it should be), so access was not denied.
1903 The bug was that the ! inversion was incorrectly passed on to the second
1904 item. This has been fixed.
1906 PH/03 Added additional dnslists conditions == and =& which are different from
1907 = and & when the dns lookup returns more than one IP address.
1909 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1910 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1912 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1913 FSYNC, which compiles an option called disable_fsync that allows for
1914 bypassing fsync(). The documentation is heavily laced with warnings.
1916 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1918 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1919 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1920 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1921 including adding "make clean"; (3) Added -fPIC when compiling the test
1922 dynamically loaded module, to get rid of a warning.
1924 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1925 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1926 = 0s. The bug is otherwise harmless.
1928 PH/07 There was a bug in the dovecot authenticator such that the value of
1929 $auth1 could be overwritten, and so not correctly preserved, after a
1930 successful authentication. This usually meant that the value preserved by
1931 the server_setid option was incorrect.
1933 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1935 PH/09 Installed PCRE release 7.0.
1937 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1938 run for batched SMTP input. It is now run at the start of every message
1939 in the batch. While fixing this I discovered that the process information
1940 (output by running exiwhat) was not always getting set for -bs and -bS
1941 input. This is fixed, and it now also says "batched" for BSMTP.
1943 PH/11 Added control=no_pipelining.
1945 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1946 patch, slightly modified), and move the expansion of helo_data till after
1947 the connection is made in the smtp transport (so it can use these
1950 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1952 PH/14 Added log_selector = +pid.
1954 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1956 PH/16 Add ${if forany and ${if forall.
1958 PH/17 Added dsn_from option to vary the From: line in DSNs.
1960 PH/18 Flush SMTP output before performing a callout, unless control =
1961 no_callout_flush is set.
1963 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1964 was true (the default) a successful delivery failed to delete the retry
1965 item, thus causing premature timeout of the address. The bug is now
1968 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1970 PH/21 Long custom messages for fakedefer and fakereject are now split up
1971 into multiline reponses in the same way that messages for "deny" and
1972 other ACL rejections are.
1974 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1975 with slight modification.
1977 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1978 draft, changing the syntax and factoring some duplicate code".
1980 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1981 for deliveries of the second and subsequent messages over the same SMTP
1984 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1985 ${reduce, with only minor "tidies".
1987 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1989 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1990 expansion side effects.
1992 PH/27 When a message times out after an over-quota error from an Exim-imposed
1993 quota, the bounce message says "mailbox is full". This message was not
1994 being given when it was a system quota that was exceeded. It now should
1997 MH/03 Made $recipients available in local_scan(). local_scan() already has
1998 better access to the recipient list through recipients_list[], but
1999 $recipients can be useful in postmaster-provided expansion strings.
2001 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2002 in the case of a MAIL command with additional options following the
2003 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2004 were accidentally chopped off.
2006 PH/29 SMTP synchronization checks are implemented when a command is read -
2007 there is a check that no more input is waiting when there shouldn't be
2008 any. However, for some commands, a delay in an ACL can mean that it is
2009 some time before the response is written. In this time, more input might
2010 arrive, invalidly. So now there are extra checks after an ACL has run for
2011 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2012 pipelining has not been advertised.
2014 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2016 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2017 returned in the SMTP response when smtp_return_error_details was false.
2018 This has been fixed.
2020 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2021 instead of the C I/O that was originally supplied, because problems were
2022 reported on Solaris.
2024 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2025 Exim which did not show up earlier: it was assuming that a call to
2026 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2027 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2028 was a macro that became an assignment, so it seemed to work. This has
2029 changed to a proper function call with a void return, hence the compile
2030 error. Exim's code has been fixed.
2032 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2035 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2037 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2039 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2040 because it turns out to be too big or there is a timeout) there is no
2041 "Completed" line in the log. When some messages of this type were
2042 selected by exigrep, they were listed as "not completed". Others were
2043 picked up by some special patterns. I have improved the selection
2044 criteria to be more general.
2046 PH/38 The host_find_failed option in the manualroute router can now be set
2047 to "ignore", to completely ignore a host whose IP address cannot be
2048 found. If all hosts are ignored, the behaviour is controlled by the new
2049 host_all_ignored option.
2051 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2052 homing or because of multiple MX records with /mx) generated more than
2053 one IP address, and the following item turned out to be the local host,
2054 all the secondary addresses of the first item were incorrectly removed
2055 from the list, along with the local host and any following hosts (which
2056 is what is supposed to happen).
2058 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2059 whoever called Exim into the -H file. In the case of the daemon it was
2060 behaving confusingly. When first started, it used values for whoever
2061 started the daemon, but after a SIGHUP it used the Exim user (because it
2062 calls itself on a restart). I have changed the code so that it now always
2065 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2066 message are rejected with the same error (e.g. no authentication or bad
2067 sender address), and a DATA command is nevertheless sent (as can happen
2068 with PIPELINING or a stupid MUA), the error message that was given to the
2069 RCPT commands is included in the rejection of the DATA command. This is
2070 intended to be helpful for MUAs that show only the final error to their
2073 PH/42 Another patch from the Sieve maintainer.
2075 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2076 Eximstats - Fixed some broken HTML links and added missing column headers
2078 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2079 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2081 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2083 PH/43 Yet another patch from the Sieve maintainer.
2085 PH/44 I found a way to check for a TCP/IP connection going away before sending
2086 the response to the final '.' that terminates a message, but only in the
2087 case where the client has not sent further data following the '.'
2088 (unfortunately, this is allowed). However, in many cases there won't be
2089 any further data because there won't be any more messages to send. A call
2090 to select() can be used: if it shows that the input is "ready", there is
2091 either input waiting, or the socket has been closed. An attempt to read
2092 the next input character can distinguish the two cases. Previously, Exim
2093 would have sent an OK response which the client would never have see.
2094 This could lead to message repetition. This fix should cure that, at
2095 least in a lot of common cases.
2097 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2098 advertised in response to EHLO.
2104 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2105 fixed by 4.65/MH/01 (is this a record?) are fixed:
2107 (i) An empty string was always treated as zero by the numeric comparison
2108 operators. This behaviour has been restored.
2110 (ii) It is documented that the numeric comparison operators always treat
2111 their arguments as decimal numbers. This was broken in that numbers
2112 starting with 0 were being interpreted as octal.
2114 While fixing these problems I realized that there was another issue that
2115 hadn't been noticed. Values of message_size_limit (both the global option
2116 and the transport option) were treated as octal if they started with 0.
2117 The documentation was vague. These values are now always treated as
2118 decimal, and I will make that clear in the documentation.
2124 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2125 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2128 MH/01 Don't check that the operands of numeric comparison operators are
2129 integers when their expansion is in "skipping" mode (fixes bug
2130 introduced by 4.64-PH/07).
2132 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2133 child addresses, Exim now panics and dies. Previously, because the count
2134 is held in a short int, deliveries were likely to be lost. As such a
2135 large number of recipients for a single message is ridiculous
2136 (performance will be very, very poor), I have chosen to impose a limit
2137 rather than extend the field.
2143 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2144 leftover -K file (the existence of which was triggered by #402).
2145 While we were at it, introduced process PID as part of the -K
2146 filename. This should rule out race conditions when creating
2149 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2150 processing considerably. Previous code took too long for large mails,
2151 triggering a timeout which in turn triggers #401.
2153 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2154 in the DK code in transports.c. sendfile() is not really portable,
2155 hence the _LINUX specificness.
2157 TF/01 In the add_headers option to the mail command in an Exim filter,
2158 there was a bug that Exim would claim a syntax error in any
2159 header after the first one which had an odd number of characters
2162 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2163 callout verification, Exim cached a "reject" for the entire domain. This
2164 is correct for most verifications, but it is not correct for a recipient
2165 verification with use_sender or use_postmaster set, because in that case
2166 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2167 case of MAIL FROM:<> rejection from other early rejections (e.g.
2168 rejection of HELO). When verifying a recipient using a non-null MAIL
2169 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2170 Whatever the result of the callout, the value of the domain cache is
2171 left unchanged (for any other kind of callout, getting as far as trying
2172 RCPT means that the domain itself is ok).
2174 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2177 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2178 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2179 session, the daemon ignores ECONNECT errors and logs others; it now
2180 ignores EPIPE as well.
2182 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2183 (quoted-printable decoding).
2185 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2186 later the small subsequent patch to fix an introduced bug.
2188 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2190 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2192 PH/08 An error is now given if message_size_limit is specified negative.
2194 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2195 to be given (somewhat) arbitrary names.
2197 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2200 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2201 miscellaneous code fixes
2203 PH/10 Added the log_reject_target ACL modifier to specify where to log
2206 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2207 hostname. This is wrong, because it relates to the incoming message (and
2208 probably the interface on which it is arriving) and not to the outgoing
2209 callout (which could be using a different interface). This has been
2210 changed to use the value of the helo_data option from the smtp transport
2211 instead - this is what is used when a message is actually being sent. If
2212 there is no remote transport (possible with a router that sets up host
2213 addresses), $smtp_active_hostname is used.
2215 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2216 tweaks were necessary in order to get it to work (see also 21 below):
2217 (a) The code assumed that strncpy() returns a negative number on buffer
2218 overflow, which isn't the case. Replaced with Exim's string_format()
2220 (b) There were several signed/unsigned issues. I just did the minimum
2221 hacking in of casts. There is scope for a larger refactoring.
2222 (c) The code used strcasecmp() which is not a standard C function.
2223 Replaced with Exim's strcmpic() function.
2224 (d) The code set only $1; it now sets $auth1 as well.
2225 (e) A simple test gave the error "authentication client didn't specify
2226 service in request". It would seem that Dovecot has changed its
2227 interface. Fortunately there's a specification; I followed it and
2228 changed what the client sends and it appears to be working now.
2230 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2233 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2234 address A is aliased to B and C, where B exists and C does not. Without
2235 -v the output is "A verified" because verification stops after a
2236 successful redirection if more than one address is generated. However,
2237 with -v the child addresses are also verified. Exim was outputting "A
2238 failed to verify" and then showing the successful verification for C,
2239 with its parentage. It now outputs "B failed to verify", showing B's
2240 parentage before showing the successful verification of C.
2242 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2243 look up a TXT record in a specific list after matching in a combined
2246 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2247 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2248 they consult the DNS. I had assumed they would set it the way they
2249 wanted; and indeed my experiments on Linux seem to show that in some
2250 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2251 To be on the safe side, however, I have now made the interface to
2252 host_find_byname() similar to host_find_bydns(), with an argument
2253 containing the DNS resolver options. The host_find_byname() function now
2254 sets these options at its start, just as host_find_bydns() does. The smtp
2255 transport options dns_qualify_single and dns_search_parents are passed to
2256 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2257 of host_find_byname() use the default settings of RES_DEFNAMES
2258 (qualify_single) but not RES_DNSRCH (search_parents).
2260 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2261 spool_read_header() do less string testing, by means of a preliminary
2262 switch on the second character of optional "-foo" lines. (This is
2263 overdue, caused by the large number of possibilities that now exist.
2264 Originally there were few.) While I was there, I also converted the
2265 str(n)cmp tests so they don't re-test the leading "-" and the first
2266 character, in the hope this might squeeze out yet more improvement.
2268 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2269 flag allowing group syntax was set by the header_syntax check but not
2270 turned off, possible causing trouble later; (2) The flag was not being
2271 set at all for the header_verify test, causing "group"-style headers to
2272 be rejected. I have now set it in this case, and also caused header_
2273 verify to ignore an empty address taken from a group. While doing this, I
2274 came across some other cases where the code for allowing group syntax
2275 while scanning a header line wasn't quite right (mostly, not resetting
2276 the flag correctly in the right place). These bugs could have caused
2277 trouble for malformed header lines. I hope it is now all correct.
2279 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2280 with the "reply" argument non-NULL. The code, however (which originally
2281 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2282 but it didn't always do it. This confused somebody who was copying the
2283 code for some other use. I have removed all the tests.
2285 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2286 feature that was used to support insecure browsers during the U.S. crypto
2287 embargo. It requires special client support, and Exim is probably the
2288 only MTA that supported it -- and would never use it because real RSA is
2289 always available. This code has been removed, because it had the bad
2290 effect of slowing Exim down by computing (never used) parameters for the
2291 RSA_EXPORT functionality.
2293 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2294 authenticator to fail if there's a tab character in the incoming data
2295 (there should never be unless someone is messing about, as it's supposed
2296 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2297 if the connection is using TLS or if the remote IP is the same as the
2298 local IP, and the "valid-client-cert option" if a client certificate has
2301 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2302 authenticators. This can be used for authorization after authentication
2303 succeeds. (In the case of plaintext, it servers for both authentication
2306 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2307 if any retry times were supplied.
2309 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2310 connection, where there is no client IP address to check. In this
2311 situation, the verify now always succeeds.
2313 PH/25 Applied John Jetmore's -Mset patch.
2315 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2317 PH/27 In a string expansion for a processed (not raw) header when multiple
2318 headers of the same name were present, leading whitespace was being
2319 removed from all of them, but trailing whitespace was being removed only
2320 from the last one. Now trailing whitespace is removed from each header
2321 before concatenation. Completely empty headers in a concatenation (as
2322 before) are ignored.
2324 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2325 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2327 PH/29 [Removed. This was a change that I later backed out, and forgot to
2328 correct the ChangeLog entry (that I had efficiently created) before
2329 committing the later change.]
2331 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2332 address errors (4xx response to RCPT) over the same connection as other
2333 messages routed to the same hosts. Such deliveries are always "forced",
2334 so retry times are not inspected. This resulted in far too many retries
2335 for the affected addresses. The effect occurred only when there were more
2336 hosts than the hosts_max_try setting in the smtp transport when it had
2337 the 4xx errors. Those hosts that it had tried were not added to the list
2338 of hosts for which the message was waiting, so if all were tried, there
2339 was no problem. Two fixes have been applied:
2341 (i) If there are any address or message errors in an SMTP delivery, none
2342 of the hosts (tried or untried) are now added to the list of hosts
2343 for which the message is waiting, so the message should not be a
2344 candidate for sending over the same connection that was used for a
2345 successful delivery of some other message. This seems entirely
2346 reasonable: after all the message is NOT "waiting for some host".
2347 This is so "obvious" that I'm not sure why it wasn't done
2348 previously. Hope I haven't missed anything, but it can't do any
2349 harm, as the worst effect is to miss an optimization.
2351 (ii) If, despite (i), such a delivery is accidentally attempted, the
2352 routing retry time is respected, so at least it doesn't keep
2353 hammering the server.
2355 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2356 in ${readsocket because some servers need this prod.
2358 PH/32 Added some extra debug output when updating a wait-xxx database.
2360 PH/33 The hint "could be header name not terminated by colon", which has been
2361 given for certain expansion errors for a long time, was not being given
2362 for the ${if def:h_colon_omitted{... case.
2364 PH/34 The spec says: "With one important exception, whenever a domain list is
2365 being scanned, $domain contains the subject domain." There was at least
2366 one case where this was not true.
2368 PH/35 The error "getsockname() failed: connection reset by peer" was being
2369 written to the panic log as well as the main log, but it isn't really
2370 panic-worthy as it just means the connection died rather early on. I have
2371 removed the panic log writing for the ECONNRESET error when getsockname()
2374 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2375 runs only) independently of the message's sender address. This meant
2376 that, if the 4xx error was in fact related to the sender, a different
2377 message to the same recipient with a different sender could confuse
2378 things. In particualar, this can happen when sending to a greylisting
2379 server, but other circumstances could also provoke similar problems.
2380 I have changed the default so that the retry time for these errors is now
2381 based a combination of the sender and recipient addresses. This change
2382 can be overridden by setting address_retry_include_sender=false in the
2385 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2386 remote server are returned as part of bounce messages. This was not
2387 happening for LMTP over a pipe (the lmtp transport), but now it is the
2388 same for both kinds of LMTP.
2390 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2391 in header lines that were in fact CNAMEs. This is no longer the case.
2393 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2394 and queue runs started by the daemon processed all messages. This has
2395 been fixed so that -R and -S can now usefully be given with -q<time>.
2397 PH/40 Import PCRE release 6.7 (fixes some bugs).
2399 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2401 PH/42 Give an error if -q is specified more than once.
2403 PH/43 Renamed the variables $interface_address and $interface_port as
2404 $received_ip_address and $received_port, to make it clear that these
2405 values apply to message reception, and not to the outgoing interface when
2406 a message is delivered. (The old names remain recognized, of course.)
2408 PH/44 There was no timeout on the connect() call when using a Unix domain
2409 socket in the ${readsocket expansion. There now is.
2411 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2412 be meaningful with "accept".
2414 SC/01 Eximstats V1.43
2415 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2417 SC/02 Eximstats V1.44
2418 Use a glob alias rather than an array ref in the generated
2419 parser. This improves both readability and performance.
2421 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2422 Collect SpamAssassin and rejection statistics.
2423 Don't display local sender or destination tables unless
2424 there is data to show.
2425 Added average volumes into the top table text output.
2427 SC/04 Eximstats V1.46
2428 Collect data on the number of addresses (recipients)
2429 as well as the number of messages.
2431 SC/05 Eximstats V1.47
2432 Added 'Message too big' to the list of mail rejection
2433 reasons (thanks to Marco Gaiarin).
2435 SC/06 Eximstats V1.48
2436 Mainlog lines which have GMT offsets and are too short to
2437 have a flag are now skipped.
2439 SC/07 Eximstats V1.49 (Alain Williams)
2440 Added the -emptyok flag.
2442 SC/08 Eximstats V1.50
2443 Fixes for obtaining the IP address from reject messages.
2445 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2446 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2447 whitesspace changes from 4.64-PH/27
2449 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2452 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2453 are found, allow negative numbers in numeric criteria)
2455 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2457 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2458 to match changes made in 4.64-PH/43
2460 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2462 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2463 "domain" is actually an IP address, and force a failure. This locks out
2464 those revolvers/nameservers that support "A-for-A" lookups, in
2465 contravention of the specifications.
2467 PH/48 When a host name was looked up from an IP address, and the subsequent
2468 forward lookup of the name timed out, the host name was left in
2469 $sender_host_name, contrary to the specification.
2471 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2472 restricted to single-key lookups, Exim was not diagnosing an error if
2473 * or *@ was used with a query-style lookup.
2475 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2477 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2478 long ago, but noone interested enough thought of it. Let's just say that
2479 the "1.1" means that there are some new functions that weren't there at
2480 some point in the past.
2482 PH/51 Error processing for expansion failure of helo_data from an smtp
2483 transport during callout processing was broken.
2485 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2486 tested/used via the -bh/-bhc/-bs options.
2488 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2489 bug, fixed in subsequent PCRE releases).
2491 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2492 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2494 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2500 SC/01 Use a glob alias rather than an array ref in eximstats generated
2501 parser. This improves both readability and performance.
2503 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2504 Don't display local sender or destination tables in eximstats unless
2505 there is data to show.
2506 Added average volumes into the eximstats top table text output.
2508 SC/03 Collect data on the number of addresses (recipients) as well
2509 as the number of messages in eximstats.
2511 TF/01 Correct an error in the documentation for the redirect router. Exim
2512 does (usually) call initgroups() when daemonizing.
2514 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2515 with consistent privilege compared to when running as a daemon.
2517 TF/03 Note in the spec that $authenticated_id is not set for local
2518 submissions from trusted users.
2520 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2521 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2523 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2524 by adding some example configuration directives to the default
2525 configuration file. A little bit of work is required to uncomment the
2526 directives and define how usernames and passwords are checked, but
2527 there is now a framework to start from.
2529 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2530 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2531 without this. I don't know how relevant this is to other LDAP libraries.
2533 PH/02 Add the verb name to the "unknown ACL verb" error.
2535 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2537 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2539 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2540 directory not expanded when it should be if an expanded home directory
2541 was set for the address (which is overridden by the transport).
2543 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2546 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2547 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2548 because it is too late at that time, and has no effect.
2550 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2551 security issue with \' (bugzilla #107). I could not use the
2552 PQescapeStringConn() function, because it needs a PGconn value as one of
2555 PH/08 When testing addresses using -bt, indicate those final addresses that
2556 are duplicates that would not cause an additional delivery. At least one
2557 person was confused, thinking that -bt output corresponded to deliveries.
2558 (Suppressing duplicates isn't a good idea as you lose the information
2559 about possibly different redirections that led to the duplicates.)
2561 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2562 systems where poll() doesn't work, in particular OS X.
2564 PH/10 Added more information to debugging output for retry time not reached.
2566 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2567 operations in malware.c.
2569 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2572 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2573 syslog_duplication set false, log lines that would normally be written
2574 both the the main log and to the reject log were not written to syslog at
2577 PH/14 In the default configuration, change the use of "message" in ACL warn
2578 statements to "add_header".
2580 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2581 not followed by a command (e.g. "seen endif").
2583 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2584 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2587 PH/17 Added extra conditions to the default value of delay_warning_condition
2591 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2592 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2593 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2596 The Auto-Submitted: and various List- headers are standardised, whereas I
2597 don't think Precedence: ever was.
2599 PH/18 Refactored debugging code in route_finduser() to show more information,
2600 in particular, the error code if getpwnam() issues one.
2602 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2603 This is apparently needed in addition to the PH/07 change above to avoid
2604 any possible encoding problems.
2606 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2607 but not after initializing Perl.
2609 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2610 output them only if debugging. By default they are written stderr,
2611 apparently, which is not desirable.
2613 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2616 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2619 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2621 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2622 authenticated or an ident call has been made. Suppress the default
2623 values for $authenticated_id and $authenticated_sender (but permit -oMai
2624 and -oMas) when testing with -bh.
2626 PH/24 Re-jigged the order of the tests in the default configuration so that the
2627 tests for valid domains and recipients precede the DNS black list and CSA
2628 tests, on the grounds that those ones are more expensive.
2630 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2631 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2632 This bug exists in every version of Exim that I still have, right back to
2635 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2636 However, an attempt to turn on case-sensitivity in a regex key by
2637 including (?-i) didn't work because the subject string was already
2638 lowercased, and the effects were non-intuitive. It turns out that a
2639 one-line patch can be used to allow (?-i) to work as expected.
2645 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2646 other effects) broke the use of negated acl sub-conditions.
2648 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2651 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2652 "Deny" causes Exim to reject the incoming connection with a 554 error.
2653 Unfortunately, if there is a major crisis, such as a disk failure,
2654 tcp-wrappers gives "deny", whereas what one would like would be some
2655 kind of temporary error. A kludge has been added to help with this.
2656 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2657 554 error is used if errno is still zero or contains ENOENT (which occurs
2658 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2661 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2663 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2664 errors. Otherwise a message that provokes a temporary error (when other
2665 messages do not) can cause a whole host to time out.
2667 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2668 addresses were routed directly to files or pipes from a redirect router.
2669 File deliveries just didn't batch; pipe deliveries might have suffered
2672 PH/06 A failure to get a lock for a hints database would erroneously always say
2673 "Failed to get write lock", even when it was really a read lock.
2675 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2676 of 0600. This has been changed to use the value of the lockfile_mode
2677 option (which defaults to 0600).
2679 PH/08 Applied small patch from the Sieve maintainer.
2681 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2682 folder from quota calculations, a direct delivery into this folder messed
2683 up the contents of the maildirsize file. This was because the regex was
2684 used only to exclude .Trash (or whatever) when the size of the mailbox
2685 was calculated. There was no check that a delivery was happening into an
2686 excluded directory. This bug has been fixed by ignoring all quota
2687 processing for deliveries into excluded directories.
2689 PH/10 Added the maildirfolder_create_regex option to appendfile.
2695 PH/01 The code for finding all the local interface addresses on a FreeBSD
2696 system running IPv6 was broken. This may well have applied to all BSD
2697 systems, as well as to others that have similar system calls. The broken
2698 code found IPv4 interfaces correctly, but gave incorrect values for the
2699 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2700 that it would not match correctly against @[] and not recognize the IPv6
2703 PH/02 The ipliteral router was not recognizing addresses of the form user@
2704 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2706 PH/03 Added disable_ipv6.
2708 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2709 decoded form, because it is most often used to construct To: headers
2710 lines in autoreplies, and the decoded form may well be syntactically
2711 invalid. However, $reply_address has leading white space removed, and all
2712 newlines turned into spaces so that the autoreply transport does not
2715 PH/05 If group was specified without a user on a router, and no group or user
2716 was specified on a transport, the group from the router was ignored.
2718 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2719 for visible compile-time settings that can be used to change these
2720 numbers, for those that want even more. Backwards compatibility with old
2721 spool files has been maintained. However, going back to a previous Exim
2722 release will lost any variables that are in spool files.
2724 PH/07 Two small changes when running in the test harness: increase delay when
2725 passing a TCP/IP connection to a new process, in case the original
2726 process has to generate a bounce, and remove special handling of
2727 127.0.0.2 (sic), which is no longer necessary.
2729 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2730 be the same on different OS.
2732 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2735 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2736 whether --show-vars was specified or not
2738 JJ/02 exipick: Added support for new ACL variable spool format introduced
2741 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2742 syntactically invalid From: or Reply-to: line, and a filter used this to
2743 generate an autoreply, and therefore failed to obtain an address for the
2744 autoreply, Exim could try to deliver to a non-existent relative file
2745 name, causing unrelated and misleading errors. What now happens is that
2746 it logs this as a hard delivery error, but does not attempt to create a
2749 PH/11 The exinext utility has a -C option for testing purposes, but although
2750 the given file was scanned by exinext itself; it wasn't being passed on
2751 when Exim was called.
2753 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2754 an end-of-file indication when reading a command response.
2756 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2757 compiled. In many other places in Exim, IPv6 addresses are always
2758 recognized, so I have changed this. It also means that IPv4 domain
2759 literals of the form [IPV4:n.n.n.n] are now always recognized.
2761 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2762 used if the router is not running as root, for example, when verifying at
2763 ACL time, or when using -bh. The debugging output from this situation was
2764 non-existent - all you got was a failure to exec. I have made two
2767 (a) Failures to set uid/gid, the current directory, or a process leader
2768 in a subprocess such as that created by queryprogram now generate
2769 suitable debugging ouput when -d is set.
2771 (b) The queryprogram router detects when it is not running as root,
2772 outputs suitable debugging information if -d is set, and then runs
2773 the subprocess without attempting to change uid/gid.
2775 PH/15 Minor change to Makefile for building test_host (undocumented testing
2778 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2779 additional section of a DNS packet that returns MX or SRV records.
2780 Instead, it always explicitly searches for A/AAAA records. This avoids
2781 major problems that occur when a DNS server includes only records of one
2782 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2783 fixed another bug: if SRV records were looked up and the corresponding
2784 address records were *not* found in the additional section, the port
2785 values from the SRV records were lost.
2787 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2788 using the correct key (the original address) when searching the retry
2789 rules in order to find which one to use for generating the retry hint.
2791 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2792 adding the default one. Similarly, if it contains a Reply-To: header, the
2793 errors_reply_to option, if set, is not used.
2795 PH/19 When calculating a retry time, Exim used to measure the "time since
2796 failure" by looking at the "first failed" field in the retry record. Now
2797 it does not use this if it is later than than the arrival time of the
2798 message. Instead it uses the arrival time. This makes for better
2799 behaviour in cases where some deliveries succeed, thus re-setting the
2800 "first failed" field. An example is a quota failure for a huge message
2801 when small messages continue to be delivered. Without this change, the
2802 "time since failure" will always be short, possible causing more frequent
2803 delivery attempts for the huge message than are intended.
2804 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2806 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2807 $1, $2, $3) because the numerical variables can be reset during some
2808 expansion items (e.g. "match"), thereby losing the authentication data.
2810 PH/21 Make -bV show the size of off_t variables so that the test suite can
2811 decide whether to run tests for quotas > 2G.
2813 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2814 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2815 filecount value is greater than 2G or if a quota value is greater than 2G
2816 on a system where the size of off_t is not greater than 4, a panic error
2819 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2820 never match. The debug and -bh output now contains an explicit error
2821 message indicating a malformed IPv4 address or mask.
2823 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2824 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2825 PH/23 above applies.
2827 PH/25 Do not write to syslog when running in the test harness. The only
2828 occasion when this arises is a failure to open the main or panic logs
2829 (for which there is an explicit test).
2831 PH/26 Added the /no_tell option to "control=freeze".
2833 PH/27 If a host name lookup failed very early in a connection, for example, if
2834 the IP address matched host_lookup and the reverse lookup yielded a name
2835 that did not have a forward lookup, an error message of the form "no IP
2836 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2837 could be logged. Now it outputs the IP address instead of "NULL".
2839 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2840 allows the sender and the authenticated sender to be set when
2841 submitting a message from within Exim. Since child_open_exim() is
2842 documented for local_scan(), the new function should be too.
2844 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2845 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2846 results in an empty string is now treated as unset.
2848 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2850 PH/31 Added sender_verify_fail logging option.
2852 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2853 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2854 tidied the source and removed it altogether.
2856 PH/33 When a queue run was abandoned because the load average was too high, a
2857 log line was always written; now it is written only if the queue_run log
2858 selector is set. In addition, the log line for abandonment now contains
2859 information about the queue run such as the pid. This is always present
2860 in "start" and "stop" lines but was omitted from the "abandon" line.
2862 PH/34 Omit spaces between a header name and the colon in the error message that
2863 is given when verify = headers_syntax fails (if there are lots of them,
2864 the message gets confusing).
2866 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2867 names, as there are now some PTR records that contain slashes. This check
2868 is only to protect against broken name servers that fall over on strange
2869 characters, so the fact that it applies to all lookups doesn't matter.
2871 PH/36 Now that the new test suite is complete, we can remove some of the
2872 special code in Exim that was needed for the old test suite. For example,
2873 sorting DNS records because real resolvers return them in an arbitrary
2874 order. The new test suite's fake resolver always returns records in the
2877 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2878 bounces) except when queue_only is set, to avoid logging races between
2879 the different processes.
2881 PH/38 Panic-die if .include specifies a non-absolute path.
2883 PH/39 A tweak to the "H" retry rule from its user.
2885 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2886 a label. They prevented compilation on older perls.
2888 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2889 a warning to be raised on newish perls.
2891 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2892 on queue. Changes to match documented behaviour of showing count of
2893 messages matching specified criteria.
2895 PH/40 Changed the default ident timeout from 30s to 5s.
2897 PH/41 Added support for the use of login_cap features, on those BSD systems
2898 that have them, for controlling the resources used by pipe deliveries.
2900 PH/42 The content-scanning code uses fopen() to create files in which to put
2901 message data. Previously it was not paying any attention to the mode of
2902 the files. Exim runs with umask(0) because the rest of the code creates
2903 files with open(), and sets the required mode explicitly. Thus, these
2904 files were ending up world-writeable. This was not a big issue, because,
2905 being within the spool directory, they were not world-accessible. I have
2906 created a function called modefopen, which takes an additional mode
2907 argument. It sets umask(777), creates the file, chmods it to the required
2908 mode, then resets the umask. All the relevant calls to fopen() in the
2909 content scanning code have been changed to use this function.
2911 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2912 to 24 hours. This avoids potential overflow problems when processing G
2913 and H retry rules. I suspect nobody ever tinkers with this value.
2915 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2917 PH/45 When the plaintext authenticator is running as a client, the server's
2918 challenges are checked to ensure they are valid base64 strings. By
2919 default, the authentication attempt is cancelled if an invalid string is
2920 received. Setting client_ignore_invalid_base64 true ignores these errors.
2921 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2922 they are received. Thus, the responses can be made to depend on the
2923 challenges. If an invalid string is ignored, an empty string is placed in
2926 PH/46 Messages that are created by the autoreply transport now contains a
2927 References: header, in accordance with RFCs 2822 and 3834.
2929 PH/47 Added authenticated_sender_force to the smtp transport.
2931 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2933 PH/49 Installed latest patch from the Sieve maintainer.
2935 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2936 was also set, the appendfile transport was unnecessarily scanning a
2937 directory of message files (e.g. for maildir delivery) to find the count
2938 of files (along with the size), even though it did not need this
2939 information. It now does the scan only if it needs to find either the
2940 size of the count of files.
2942 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2944 PH/52 Two bugs concerned with error handling when the smtp transport is
2947 (i) Exim was not creating retry information for temporary errors given
2948 for individual recipients after the DATA command when the smtp transport
2949 was used in LMTP mode. This meant that they could be retried too
2950 frequently, and not timed out correctly.
2952 (ii) Exim was setting the flag that allows error details to be returned
2953 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2954 recipients that were returned after the DATA command.
2956 PH/53 This is related to PH/52, but is more general: for any failing address,
2957 when detailed error information was permitted to be returned to the
2958 sender, but the error was temporary, then after the final timeout, only
2959 "retry timeout exceeded" was returned. Now it returns the full error as
2960 well as "retry timeout exceeded".
2962 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2963 do this, and (what is worse) MTAs that accept it.
2965 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2966 will now be deprecated.
2968 PH/56 New os.c-cygwin from the Cygwin maintainer.
2970 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2971 formats (previously only available in exim formats via -bpr, -bpru,
2972 and -bpra. Now also available in native and exiqgrep formats)
2974 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2975 with very large, slow to parse queues
2977 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2979 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2981 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2982 responses to authentication challenges, though it was showing the
2983 challenges; (ii) I've removed the CR characters from the debug output for
2986 PH/58 Allow for the insertion of a newline as well as a space when a string
2987 is turned into more than one encoded-word during RFC 2047 encoding. The
2988 Sieve code now uses this.
2990 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2991 data_4xx, lost_connection, tls_required.
2993 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2994 message was being sent as an SMTP response.
2996 PH/61 Add -l and -k options to exicyclog.
2998 PH/62 When verifying, if an address was redirected to one new address, so that
2999 verification continued, and the new address failed or deferred after
3000 having set something in $address_data, the value of $address_data was not
3001 passed back to the ACL. This was different to the case when no
3002 redirection occurred. The value is now passed back in both cases.
3004 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3005 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3006 use login_cap.h, so on its own it isn't the distinguishing feature. The
3007 new name refers directly to the setclassresources() function.
3009 PH/65 Added configuration files for NetBSD3.
3011 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3013 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3014 is preferred over IPv4.
3016 PH/68 The bounce_return_message and bounce_return_body options were not being
3017 honoured for bounces generated during the reception of non-SMTP messages.
3018 In particular, this applied to messages rejected by the ACL. This bug has
3019 been fixed. However, if bounce_return_message is true and bounce_return_
3020 body is false, the headers that are returned for a non-SMTP message
3021 include only those that have been read before the error was detected.
3022 (In the case of an ACL rejection, they have all been read.)
3024 PH/69 The HTML version of the specification is now built in a directory called
3025 spec_html instead of spec.html, because the latter looks like a path with
3026 a MIME-type, and this confuses some software.
3028 PH/70 Catch two compiler warnings in sieve.c.
3030 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3031 function verify_get_ident() calls ip_connect() to connect a socket, but
3032 if the "connect()" function timed out, ip_connect() used to close the
3033 socket. However, verify_get_ident() also closes the socket later, and in
3034 between Exim writes to the log, which may get opened at this point. When
3035 the socket was closed in ip_connect(), the log could get the same file
3036 descriptor number as the socket. This naturally causes chaos. The fix is
3037 not to close the socket in ip_connect(); the socket should be closed by
3038 the function that creates it. There was only one place in the code where
3039 this was missing, in the iplookup router, which I don't think anybody now
3040 uses, but I've fixed it anyway.
3042 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3043 well as to direct DNS lookups. Otherwise the handling of names in host
3044 lists is inconsistent and therefore confusing.
3050 PH/01 Two changes to the default runtime configuration:
3052 (1) Move the checks for relay_from_hosts and authenticated clients from
3053 after to before the (commented out) DNS black list checks.
3055 (2) Add control=submission to the relay_from_hosts and authenticated
3056 clients checks, on the grounds that messages accepted by these
3057 statements are most likely to be submissions.
3059 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3061 (1) Generate an error if the third argument for the ${prvs expansion is
3064 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3067 (3) Reset the variables that are obtained from the first argument of
3068 ${prvscheck and used in the second argument before leaving the code,
3069 because their memory is reclaimed, so using them afterwards may do
3072 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3073 one (it's much easier than Tom thought :-).
3075 (5) Because of (4), we can now allow for the use of $prvscheck_result
3076 inside the third argument.
3078 PH/03 For some reason, the default setting of PATH when running a command from
3079 a pipe transport was just "/usr/bin". I have changed it to
3082 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3083 anything to be listed in the output from -bV.
3085 PH/05 When a filter generated an autoreply, the entire To: header line was
3086 quoted in the delivery log line, like this:
3088 => >A.N.Other <ano@some.domain> <original@ddress> ...
3090 This has been changed so that it extracts the operative address. There
3091 may be more than one such address. If so, they are comma-separated, like
3094 => >ano@some.domain,ona@other.domain <original@ddress> ...
3096 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3097 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3098 not being looked up in the rDNS to get a host name, Exim was showing the
3099 IP address twice in Received: lines, even though the IP addresses were
3100 identical. For example:
3102 Received: from [1.2.3.4] (helo=[1.2.3.4])
3104 However, if the real host name was known, it was omitting the HELO data
3105 if it matched the actual IP address. This has been tidied up so that it
3106 doesn't show the same IP address twice.
3108 PH/07 When both +timestamp and +memory debugging was on, the value given by
3109 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3110 called by the debug printing, thereby overwriting the timestamp buffer.
3111 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3113 PH/08 When the original message was included in an autoreply transport, it
3114 always said "this is a copy of the message, including all the headers",
3115 even if body_only or headers_only was set. It now gives an appropriate
3118 PH/09 Applied a patch from the Sieve maintainer which:
3120 o fixes some comments
3121 o adds the (disabled) notify extension core
3122 o adds some debug output for the result of if/elsif tests
3123 o points to the current vacation draft in the documentation
3124 and documents the missing references header update
3128 o fixes a bug in processing the envelope test (when testing
3129 multiple envelope elements, the last element determinted the
3132 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3133 Electronic Mail") by including:
3135 Auto-submitted: auto-generated
3137 in the messages that it generates (bounce messages and others, such as
3138 warnings). In the case of bounce messages for non-SMTP mesages, there was
3139 also a typo: it was using "Auto_submitted" (underscore instead of
3140 hyphen). Since every message generated by Exim is necessarily in response
3141 to another message, thes have all been changed to:
3143 Auto-Submitted: auto-replied
3145 in accordance with these statements in the RFC:
3147 The auto-replied keyword:
3149 - SHOULD be used on messages sent in direct response to another
3150 message by an automatic process,
3152 - MUST NOT be used on manually-generated messages,
3154 - MAY be used on Delivery Status Notifications (DSNs) and Message
3155 Disposition Notifications (MDNs),
3157 - MUST NOT be used on messages generated by automatic or periodic
3158 processes, except for messages which are automatic responses to
3161 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3162 to the default Received: header definition.
3164 PH/12 Added log selector acl_warn_skipped (default on).
3166 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3167 variables because (a) they are in the wrong storage pool and (b) even if
3168 they were copied, it wouldn't work properly because of the caching.
3170 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3171 checking when decoding. Apparently there are clients that generate
3172 overlong encoded strings. Why am I not surprised?
3174 PH/15 If the first argument of "${if match_address" was not empty, but did not
3175 contain an "@" character, Exim crashed. Now it writes a panic log message
3176 and treats the condition as false.
3178 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3180 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3181 extension "envelope-auth". The code is finished and in agreement with
3182 other implementations, but there is no documentation so far and in fact,
3183 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3184 not changing the active code.
3186 Print executed "if" and "elsif" statements when debugging is used. This
3187 helps a great deal to understand what a filter does.
3189 Document more things not specified clearly in RFC3028. I had all this
3190 sorted out, when out of a sudden new issues came to my mind. Oops."
3192 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3195 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3196 canonical form (as documented). However, after a host name lookup from
3197 the IP address, check_host() was doing a simple string comparison with
3198 addresses acquired from the DNS when checking that the found name did
3199 have the original IP as one of its addresses. Since any found IPv6
3200 addresses are likely to be in abbreviated form, the comparison could
3201 fail. Luckily, there already exists a function for doing the comparison
3202 by converting both addresses to binary, so now that is used instead of
3203 the text comparison.
3205 PH/20 There was another similar case to PH/19, when a complete host name was
3206 given in a host list; looking up its IP address could give an abbreviated
3207 form, whereas the current host's name might or might not be abbreviated.
3208 The same fix has been applied.
3214 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3215 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3218 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3219 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3221 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3223 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3224 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3225 Also, for "Auto-Submitted", treat anything other than "no" as
3226 non-personal, in accordance with RFC 3834. (Previously it treated
3227 anything starting "auto-" as non-personal.)
3229 TF/01 The control=submission/name=... option had a problem with syntax
3230 errors if the name included a slash character. The /name= option
3231 now slurps the rest of the string, so it can include any characters
3232 but it must come last in the list of options (after /sender_retain
3235 PH/05 Some modifications to the interface to the fake nameserver for the new
3243 TK/01 Added the "success_on_redirect" address verification option. See
3244 NewStuff for rationale and an example.
3246 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3248 PH/02 Patch to exigrep to allow it to work on syslog lines.
3250 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3251 fread() to skip over the body file's header line, because in Cygwin the
3252 header line is locked and is inaccessible.
3254 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3255 co-exist for some time) to make it clear that it is the Exim ID that is
3256 referenced, not the Message-ID: header line.
3258 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3259 string_format() function, because snprintf() does not exist on all
3262 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3263 ${stat: expansion item.
3265 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3266 protocol synchronization error", to keep the pedants happy.
3268 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3269 well as for IRIX systems, when gcc is being used. See the host.c source
3272 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3274 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3277 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3278 transport and to the smtp transport in LMTP mode.
3280 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3282 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3283 run a filter in a subprocess. This could lead to confusion in subsequent
3284 lookups in the parent process. There should also be a search_tidyup() at
3285 the end of the subprocess.
3287 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3288 only if the host matched helo_try_verify_hosts, which caused the
3289 verification to occur when the EHLO/HELO command was issued. The ACL just
3290 tested the remembered result. Now, if a previous verification attempt has
3291 not happened, "verify = helo" does it there and then.
3293 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3295 TK/03 Fix log output including CR from clamd.
3297 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3298 exist provoked a memory error which could cause a segfault.
3300 PH/15 Installed PCRE 6.2
3302 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3304 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3305 of the problem. Specifically, suggested +O2 rather than +O1 for the
3308 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3310 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3311 record, the port was not correctly logged when the outgoing_port log
3312 selector was set (it logged the transort's default port).
3314 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3315 fallback_hosts, and "hosts" in the smtp transport.
3317 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3318 host errors such as "Connection refused".
3320 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3321 authentication with radiusclient 0.4.9:
3323 - Error returned from rc_read_config was caught wrongly
3324 - Username/password not passed on to radius server due to wrong length.
3326 The presumption is that some radiusclient API changes for 4.51/PH/17
3327 were not taken care of correctly. The code is still untested by me (my
3328 Linux distribution still has 0.3.2 of radiusclient), but it was
3329 contributed by a Radius user.
3331 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3332 expanding the "port" option of the smtp transport.
3334 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3335 while reading a MIME header. Thanks to Tom Hughes for a patch.
3337 PH/24 Include config.h inside local_scan.h so that configuration settings are
3340 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3341 that in an ACL for RCPT (for example), you can examine exactly what was
3344 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3345 commands, but it was not correctly comparing the address with the actual
3346 client host address. Thus, it would show the EHLO address in Received:
3347 header lines when this was not necessary.
3349 PH/27 Added the % operator to ${eval:}.
3351 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3352 it should be ignoring failures (because with -C, for example, it has lost
3353 privilege). It wasn't ignoring creation failures other than "already
3356 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3359 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3360 because an input line was too long, either on its own, or by virtue of
3361 too many continuations, the temporary file was not being removed, and the
3362 return code was incorrect.
3364 PH/31 Missing "BOOL" in function definition in filtertest.c.
3366 PH/32 Applied Sieve patches from the maintainer.
3368 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3370 PH/33 Added "verify = not_blind".
3372 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3373 Local/Makefile (with some defaults set). These are used in built scripts
3374 such as exicyclog, but they have never been used in the exim_install
3375 script (though there are many overriding facilities there). I have
3376 arranged that the exim_install script now takes note of these two
3379 PH/35 Installed configuration files for Dragonfly.
3381 PH/36 When a locally submitted message by a trusted user did not contain a
3382 From: header, and the sender address was obtained from -f or from an SMTP
3383 MAIL command, and the trusted user did not use -F to supply a sender
3384 name, $originator_name was incorrectly used when constructing a From:
3385 header. Furthermore, $originator_name was used for submission mode
3386 messages from external hosts without From: headers in a similar way,
3387 which is clearly wrong.
3389 PH/37 Added control=suppress_local_fixups.
3391 PH/38 When log_selector = +received_sender was set, and the addition of the
3392 sender made the log line's construction buffer exactly full, or one byte
3393 less than full, an overflow happened when the terminating "\n" was
3396 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3397 when the result of a list match is failure because a DNS lookup failed.
3399 PH/40 RM_COMMAND is now used in the building process.
3401 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3402 the "build-* directories that it finds.
3404 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3405 address in a domain literal was a prefix of an interface address.
3407 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3408 when verifying a sender address, unless rewrite_headers is false.
3410 PH/44 Wrote a long comment about why errors_to addresses are verified as
3411 recipients, not senders.
3413 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3414 the ratelimit ACL was added.
3416 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3418 PH/46 Added extra information about PostgreSQL errors to the error string.
3420 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3421 avoiding the need to install special zones in a real server. This is
3422 backwards compatible; if it can't find the fake resolver, it drops back.
3423 Thus, both old and new test suites can be run.
3425 TF/02 Added util/ratelimit.pl
3427 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3428 clock is set back in time.
3430 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3431 Candler <B.Candler@pobox.com>.
3433 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3434 OK for addresses that are the result of redirections.
3436 PH/48 A number of further additions for the benefit of the new test suite,
3437 including a fake gethostbyname() that interfaces to the fake DNS resolver
3440 TF/06 The fix for widen_domains has also been applied to qualify_single and
3441 search_parents which are the other dnslookup options that can cause
3444 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3447 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3449 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3450 service" if it encountered an MX record with an empty target hostname.
3451 The message is now "an MX or SRV record indicated no SMTP service".
3453 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3454 if the DNS of the sending site is misconfigured. This is quite a
3455 common situation. This change restores the behaviour of treating a
3456 helo verification defer as a failure.
3458 PH/51 If self=fail was set on a router, the bounce message did not include the
3459 actual error message.
3465 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3467 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3468 command itself ended in error, the underlying message about the transport
3469 filter timeout was being overwritten with the pipe command error. Now the
3470 underlying error message should be appended to the second error message.
3472 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3474 PH/02 Reduce the amount of output that "make" produces by default. Full output
3475 can still be requested.
3477 PH/03 The warning log line about a condition test deferring for a "warn" verb
3478 was being output only once per connection, rather than after each
3479 occurrence (because it was using the same function as for successful
3480 "warn" verbs). This seems wrong, so I have changed it.
3482 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3483 it should not have, which might have caused a crash in the right
3484 circumstances, but probably never did.
3486 PH/04 Installed a modified version of Tony Finch's patch to make submission
3487 mode fix the return path as well as the Sender: header line, and to
3488 add a /name= option so that you can make the user's friendly name appear
3491 TF/03 Added the control = fakedefer ACL modifier.
3493 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3494 Mark Lowes for thorough testing.
3496 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3498 TK/03 Merged latest SRS patch from Miles Wilton.
3500 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3501 with the definition in sysexits.h (which is #included earlier).
3502 Fortunately, Exim does not actually use EX_OK. The code used to try to
3503 preserve the sysexits.h value, by assumimg that macro definitions were
3504 scanned for macro replacements. I have been disabused of this notion,
3505 so now the code just undefines EX_OK before #including unistd.h.
3507 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3508 in the smtp transport. When a block could not be written in a single
3509 write() function, the timeout was being re-applied to each part-write.
3510 This seems wrong - if the receiver was accepting one byte at a time it
3511 would take for ever. The timeout is now adjusted when this happens. It
3512 doesn't have to be particularly precise.
3514 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3515 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3517 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3518 without a domain if the check to <postmaster@domain> fails.
3520 SC/01 Eximstats: added -xls and the ability to specify output files
3521 (patch written by Frank Heydlauf).
3523 SC/02 Eximstats: use FileHandles for outputing results.
3525 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3527 SC/04 Eximstats: fixed display of large numbers with -nvr option
3529 SC/05 Eximstats: fixed merging of reports with empty tables.
3531 SC/06 Eximstats: added the -include_original_destination flag
3533 SC/07 Eximstats: removed tabs and trailing whitespace.
3535 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3537 TK/06 MBOX spool code: Add real "From " MBOX separator line
3538 so the .eml file is really in mbox format (even though
3539 most programs do not really care). Patch from Alex Miller.
3541 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3542 The latter is generated from $received_to and is only set if the
3543 message has one envelope recipient. SA can use these headers,
3544 obviously out-of-the-box. Patch from Alex Miller.
3546 PH/08 The ${def test on a variable was returning false if the variable's
3547 value was "0", contrary to what the specification has always said!
3548 The result should be true unless the variable is empty.
3550 PH/09 The syntax error of a character other than { following "${if
3551 def:variable_name" (after optional whitespace) was not being diagnosed.
3552 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3553 accidental colon was present, for example, could give incorrect results.
3555 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3556 result is used (not including appendfile, where other changes are about
3559 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3560 This involved changing a lot of size variables from int to off_t. It
3561 should work with maildirs and everything.
3563 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3564 spamd dying while we are connected to it.
3566 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3569 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3570 function for BDB 4.3.
3572 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3574 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3575 the sources for PCRE have been re-arranged and more files are now
3578 PH/15 The code I had for printing potentially long long variables in PH/11
3579 above was not the best (it lost precision). The length of off_t variables
3580 is now inspected at build time, and an appropriate printing format (%ld
3581 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3582 to be "long long int" or "long int". This is needed for the internal
3583 formatting function string_vformat().
3585 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3586 the configuration file to be ":syslog", then the script "guesses" where
3587 the logs files are, rather than using the compiled in default. In our
3588 case the guess is not the same as the compiled default, so the script
3589 suddenly stopped working when I started to use syslog. The patch checks
3590 to see if log_file_path is "". If so, it attempts to read it from exim
3591 with no configuration file to get the compiled in version, before it
3592 falls back to the previous guessing code."
3594 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3595 implementing BATV in an Exim configuration. See NewStuff for the gory
3598 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3599 Makefile that are specific to HP-UX.
3601 PH/18 If the "use_postmaster" option was set for a recipient callout together
3602 with the "random" option, the postmaster address was used as the MAIL
3603 FROM address for the random test, but not for the subsequent recipient
3604 test. It is now used for both.
3606 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3607 patch removes a few documentation additions to RFC 3028, because the
3608 latest draft now contains them. It adds the new en;ascii-case comparator
3609 and a new error check for 8bit text in MIME parts. Comparator and
3610 require names are now matched exactly. I enabled the subaddress
3611 extension, but it is not well tested yet (read: it works for me)."
3613 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3614 rework some of the code of TK/09 above to avoid the hardwired use of
3615 "%lld" and "long long". Replaced the call to snprintf() with a call to
3618 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3619 records point to non-existent hosts", "retry timeout exceeded", and
3620 "retry time not reached for any host after a long failure period".
3622 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3623 experimental DomainKeys support:
3625 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3626 (2) On an error such as an illegally used "control", the wrong name for
3627 the control was given.
3629 These problems did NOT occur unless DomainKeys support was compiled.
3631 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3633 PH/24 Added ${if match_ip condition.
3635 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3636 databases so that it will be absolutely obvious if a crash occurs in the
3637 DB library. This is a regular occurrence (often caused by mis-matched
3640 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3641 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3642 picked up on a user's system that detects such things. There doesn't seem
3643 to be a gcc warning option for this - only an attribute that has to be
3644 put on the function's prototype. It seems that in Fedora Core 4 they have
3645 set this on a number of new functions. No doubt there will be more in due
3648 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3649 specify a transport. However, if an address that was verified by such a
3650 router was the subject of a callout, Exim crashed because it tried to
3651 read the rcpt_include_affixes from the non-existent transport. Now it
3652 just assumes that the setting of that option is false. This bug was
3653 introduced by 4.51/PH/31.
3655 PH/28 Changed -d+all to exclude +memory, because that information is very
3656 rarely of interest, but it makes the output a lot bigger. People tend to
3657 do -d+all out of habit.
3659 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3660 code in os-type was giving problems when libc.so lives in lib64, like on
3663 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3664 aren't the modern standard, and it seems that some systems' include files
3665 don't always have them. Exim was already checking for some of the newer
3666 ones like T_AAAA, and defining it itself. I've added checks for all the
3667 record types that Exim uses.
3669 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3670 not automatically generating a new one, as it is supposed to. This
3671 prevented TLS from working. If the file did exist, but contained invalid
3672 data, a new version was generated, as expected. It was only the case of a
3673 non-existent file that was broken.
3675 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3676 with a change in libdomainkeys > 0.64.
3678 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3679 from DNS. If the selector record carries the flag, it now has
3680 precedence over the domain-wide flag.
3682 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3684 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3685 the use of a port name that isn't defined in /etc/services) occurred, the
3686 message was deferred as in a normal delivery, and thus remained on the
3687 spool, instead of being failed because of the mua_wrapper setting. This
3688 is now fixed, and I tidied up some of the mua_wrapper messages at the
3691 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3692 the messages in a hash of arrays rather than using individual hashes.
3693 This is a bit cleaner and results in dramatic memory savings, albeit
3694 at a slight CPU cost.
3696 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3697 as requested by Marc Sherman.
3699 SC/10 Eximstats: added histograms for user specified patterns as requested
3702 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3704 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3705 fopen() in the content-scanning modules that did not already have it.
3711 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3712 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3714 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3716 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3718 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3719 to test Sieve filters that use "vacation".
3721 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3722 that changes the way the GnuTLS parameters are stored in the cache file.
3723 The new format can be generated externally. For backward compatibility,
3724 if the data in the cache doesn't make sense, Exim assumes it has read an
3725 old-format file, and it generates new data and writes a new file. This
3726 means that you can't go back to an older release without removing the
3729 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3730 work if there are any delivery delays because "one_time" forces the
3731 parent to be marked "delivered", so its unseen clone is never tried
3732 again. For this reason, Exim now forbids the simultaneous setting of
3735 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3736 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3737 entry if you want to know the details. The fix, however, neglected to
3738 consider the case when local delivery batching is involved. The test for
3739 "previously delivered" was not happening when checking to see if an
3740 address could be batched with a previous (undelivered) one; under
3741 certain circumstances this could lead to multiple deliveries to the same
3744 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3745 in its include files, and this causes problems building Exim.
3747 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3748 header_syntax) but Exim was just ignoring anything given after a slash.
3749 In particular, this caused confusion with an attempt to use "verify =
3750 reverse_host_lookup/defer_ok". An error is now given when options are
3751 supplied for verify items that do not have them. (Maybe reverse_host_
3752 lookup should have a defer_ok option, but that's a different point.)
3754 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3755 defined by RFC 821) to 2048, because there were problems with some AUTH
3756 commands, and RFC 1869 says the size should be increased for extended
3757 SMTP commands that take arguments.
3759 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3762 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3763 "unknown" error; now it says that the functionality isn't in the binary.
3765 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3766 an address' error message when a string expansion fails (syntax or
3767 whatever). Otherwise the password may appear in the log. Following change
3768 PH/42 below, there is no longer a chance of it appearing in a bounce
3771 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3773 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3774 of its addresses was ever tried. (Bugzilla bug #2.)
3776 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3777 the result incorrectly in the debug output. (It correctly added a newline
3778 to what was transported.)
3780 TF/01 Added $received_time.
3782 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3783 commented out examples of how to interface to a virus scanner and to
3784 SpamAssassin. Also added commented examples of av_scanner and
3785 spamd_address settings.
3787 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3788 and controls are allowed in which ACLs. There were a couple of minor
3789 errors. Some of the entries in the conditions table (which is a table of
3790 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3791 negation of where the condition IS allowed.
3793 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3795 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3796 header file does not have a version number, so I've had to invent a new
3797 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3798 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3799 radiusclient), but it was contributed by a Radius user.
3801 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3802 files or directories, for OpenSSL.
3804 PH/19 When an Exim process that is running as root has to create an Exim log
3805 file, it does so in a subprocess that runs as exim:exim so as to get the
3806 ownership right at creation (otherwise, other Exim processes might see
3807 the file with the wrong ownership). There was no test for failure of this
3808 fork() call, which would lead to the process getting stuck as it waited
3809 for a non-existent subprocess. Forks do occasionally fail when resources
3810 run out. I reviewed all the other calls to fork(); they all seem to check
3813 PH/20 When checking for unexpected SMTP input at connect time (before writing
3814 the banner), Exim was not dealing correctly with a non-positive return
3815 from the read() function. If the client had disconnected by this time,
3816 the result was a log entry for a synchronization error with an empty
3817 string after "input=" when read() returned zero. If read() returned -1
3818 (an event I could not check), uninitialized data bytes were printed.
3819 There were reports of junk text (parts of files, etc) appearing after
3822 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3824 PH/22 Added support for macro redefinition, and (re)definition in between
3825 driver and ACL definitions.
3827 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3828 forgetting to use the resulting value; it was using the unexpanded value.
3830 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3831 hadn't been configured. The fix is from Juergen Kreileder, who
3832 understands it better than I do:
3834 "Here's what I see happening with three configured cyrus_sasl
3835 authenticators configured (plain, login, cram-md5):
3837 On startup auth_cyrus_sasl_init() gets called for each of these.
3838 This means three calls to sasl_listmech() without a specified mech_list.
3839 => SASL tests which mechs of all available mechs actually work
3840 => three warnings about OTP not working
3841 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3843 With the patch, sasl_listmech() also gets called three times. But now
3844 SASL's mech_list option is set to the server_mech specified in the the
3845 authenticator. Or in other words, the answer from sasl_listmech()
3846 gets limited to just the mech you're testing for (which is different
3848 => the return list contains just 'plain' or 'login', 'cram-md5' or
3849 nothing depending on the value of ob->server_mech.
3851 I've just tested the patch: Authentication still works fine,
3852 unavailable mechs specified in the exim configuration are still
3853 caught, and the auth.log warnings about OTP are gone."
3855 PH/25 When debugging is enabled, the contents of the command line are added
3856 to the debugging output, even when log_selector=+arguments is not
3859 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3860 answer is "GNU", and only if the return is "GNU/something" is the answer
3863 PH/27 $acl_verify_message is now set immediately after the failure of a
3864 verification in an ACL, and so is available in subsequent modifiers. In
3865 particular, the message can be preserved by coding like this:
3867 warn !verify = sender
3868 set acl_m0 = $acl_verify_message
3870 Previously, $acl_verify_message was set only while expanding "message"
3871 and "log_message" when a very denied access.
3873 PH/28 Modified OS/os.c-Linux with
3875 -#ifndef OS_LOAD_AVERAGE
3876 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3878 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3879 nomenclature these days.)
3881 PH/29 Installed patch from the Sieve maintainer that adds the options
3882 sieve_useraddress and sieve_subaddress to the redirect router.
3884 PH/30 In these circumstances:
3885 . Two addresses routed to the same list of hosts;
3886 . First host does not offer TLS;
3887 . First host accepts first address;
3888 . First host gives temporary error to second address;
3889 . Second host offers TLS and a TLS session is established;
3890 . Second host accepts second address.
3891 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3892 and peerdn, if requested) that were in fact used only for the second
3895 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3896 attention to any local part prefix or suffix that was matched by the
3897 router that accepted the address. It now behaves in the same way as it
3898 does for delivery: the affixes are removed from the local part unless
3899 rcpt_include_affixes is set on the transport.
3901 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3902 timeout during the DATA phase of an incoming message.
3904 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3905 applied a patch sanctioned by the Sieve maintainer.
3907 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3908 the uid or gid is negative. A case of a negative gid caused this to be
3909 noticed. The fix allows for either to be negative.
3911 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3912 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3915 PH/36 The change PH/12 above was broken. Fixed it.
3917 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3918 the grounds that routing the same address twice would always produce the
3919 same answer. This might have been true once, but it is certainly no
3920 longer true now. Routing a child address may depend on the previous
3921 routing that produced that child. Some complicated redirection strategies
3922 went wrong when messages had multiple recipients, and made Exim's
3923 behaviour dependent on the order in which the addresses were given.
3925 I have moved the duplicate checking until after the routing is complete.
3926 Exim scans the addresses that are assigned to local and remote
3927 transports, and removes any duplicates. This means that more work will be
3928 done, as duplicates will always all be routed, but duplicates are
3929 presumably rare, so I don't expect this is of any significance.
3931 For deliveries to pipes, files, and autoreplies, the duplicate checking
3932 still happens during the routing process, since they are not going to be
3935 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3936 It corrects a timeout issue with spamd. This is Ian's comment: "The
3937 background is that sometimes spamd either never reads data from a
3938 connection it has accepted, or it never writes response data. The exiscan
3939 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3940 blindly assumes that writes won't block so it may never time out."
3942 PH/39 Allow G after quota size as well as K and M.
3944 PH/40 The value set for $authenticated_id in an authenticator may not contain
3945 binary zeroes or newlines because the value is written to log lines and
3946 to spool files. There was no check on this. Now the value is run through
3947 the string_printing() function so that such characters are converted to
3948 printable escape sequences.
3950 PH/41 $message_linecount is a new variable that contains the total number of
3951 lines in the message. Compare $body_linecount, which is the count for the
3954 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3955 bounce and delay warning messages, except in certain special cases, which
3958 (a) An SMTP error message from a remote host;
3959 (b) A message specified in a :fail: redirection;
3960 (c) A message specified in a "fail" command in a system filter;
3961 (d) A message specified in a FAIL return from the queryprogram router;
3962 (e) A message specified by the cannot_route_message router option.
3964 In these cases only, Exim does include the error details in bounce and
3965 warning messages. There are also a few cases where bland messages such
3966 as "unrouteable address" or "local delivery error" are given.
3968 PH/43 $value is now also set for the "else" part of a ${run expansion.
3970 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3971 being worked on, but at least Exim now implements the latest version to
3974 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3975 process to complete was treated as a delivery failure, a timeout while
3976 writing the message to the pipe was logged, but erroneously treated as a
3977 successful delivery. Such timeouts include transport filter timeouts. For
3978 consistency with the overall process timeout, these timeouts are now
3979 treated as errors, giving rise to delivery failures by default. However,
3980 there is now a new Boolean option for the pipe transport called
3981 timeout_defer, which, if set TRUE, converts the failures into defers for
3982 both kinds of timeout. A transport filter timeout is now identified in
3985 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3986 systems where "make" and "gmake" are different, calling "gmake" at top
3987 level broke things. I've arranged for the value of $(MAKE) to be passed
3988 from the Makefile to this script so that it can call the same version of
3992 A note about Exim versions 4.44 and 4.50
3993 ----------------------------------------
3995 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3996 changes of various kinds. As a consequence, a big documentation update was
3997 needed. This delayed the release for rather longer than seemed good, especially
3998 in the light of a couple of (minor) security issues. Therefore, the changes
3999 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4000 release. So 4.44 and 4.50 are in effect two different branches that both start
4003 I have left the 4.50 change log unchanged; it contains all the changes since
4004 4.43. The change log for 4.44 is below; many of its items are identical to
4005 those for 4.50. This seems to be the most sensible way to preserve the
4006 historical information.
4012 1. Minor wording change to the doc/README.SIEVE file.
4014 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4015 computation of the current number of files was incorrect.
4017 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4018 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4021 4. Give more explanation in the error message when the command for a transport
4022 filter fails to execute.
4024 5. There are several places where Exim runs a non-Exim command in a
4025 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4026 was being done only for the command run by the queryprogram router. It is
4027 now done for all such subprocesses. The other cases are: ${run, transport
4028 filters, and the commands run by the lmtp and pipe transports.
4030 6. Added CONFIGURE_GROUP build-time option.
4032 7. Some older OS have a limit of 256 on the maximum number of file
4033 descriptors. Exim was using setrlimit() to set 1000 as a large value
4034 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4035 systems. I've change it so that if it can't get 1000, it tries for 256.
4037 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4038 was an oversight, and furthermore, ever since the addition of extra
4039 controls (e.g. 4.43/32), the checks on when to allow different forms of
4040 "control" were broken. There should now be diagnostics for all cases when a
4041 control that does not make sense is encountered.
4043 9. Added the /retain_sender option to "control=submission".
4045 10. $recipients is now available in the predata ACL (oversight).
4047 11. Tidy the search cache before the fork to do a delivery from a message
4048 received from the command line. Otherwise the child will trigger a lookup
4049 failure and thereby defer the delivery if it tries to use (for example) a
4050 cached ldap connection that the parent has called unbind on.
4052 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4053 of $address_data from the recipient verification was clobbered by the
4054 sender verification.
4056 13. The value of address_data from a sender verification is now available in
4057 $sender_address_data in subsequent conditions in the ACL statement.
4059 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4061 15. Added a new option "connect=<time>" to callout options, to set a different
4064 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4065 was its contents. (It was OK if the option was not defined at all.)
4067 17. A "Completed" log line is now written for messages that are removed from
4068 the spool by the -Mrm option.
4070 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4071 information about exactly what failed.
4073 19. Added -dd to debug only the daemon process.
4075 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4076 handles timeouts, both on the server side and network timeouts. Renamed the
4077 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4079 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4080 It is now set to "smtps".
4082 22. $host_address is now set to the target address during the checking of
4083 ignore_target_hosts.
4085 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4086 being passed; this would have caused $sender_host_name to have been used if
4087 matching the list had actually called for a host name (not very likely,
4088 since this list is usually IP addresses). A host name is now passed as
4091 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4092 code that specifies a non-restarting handler (typically sigaction() in
4093 modern systems) in an attempt to fix a rare and obscure crash bug.
4095 25. Narrowed the window for a race in the daemon that could cause it to ignore
4096 SIGCHLD signals. This is not a major problem, because they are used only to
4097 wake it up if nothing else does.
4099 26. A malformed maildirsize file could cause Exim to calculate negative values
4100 for the mailbox size or file count. Odd effects could occur as a result.
4101 The maildirsize information is now recalculated if the size or filecount
4104 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4105 support for a long time. Removed HAVE_SYS_VFS_H.
4107 28. Installed the latest version of exipick from John Jetmore.
4109 29. In an address list, if the pattern was not a regular expression, an empty
4110 subject address (from a bounce message) matched only if the pattern was an
4111 empty string. Non-empty patterns were not even tested. This was the wrong
4112 because it is perfectly reasonable to use an empty address as part of a
4113 database query. An empty address is now tested by patterns that are
4114 lookups. However, all the other forms of pattern expect the subject to
4115 contain a local part and a domain, and therefore, for them, an empty
4116 address still always fails if the pattern is not itself empty.
4118 30. Exim went into a mad DNS loop when attempting to do a callout where the
4119 host was specified on an smtp transport, and looking it up yielded more
4120 than one IP address.
4122 31. Re-factored the code for checking spool and log partition space into a
4123 function that finds that data and another that does the check. The former
4124 is then used to implement four new variables: $spool_space, $log_space,
4125 $spool_inodes, and $log_inodes.
4127 32. The RFC2047 encoding function was originally intended for short strings
4128 such as real names; it was not keeping to the 75-character limit for
4129 encoded words that the RFC imposes. It now respects the limit, and
4130 generates multiple encoded words if necessary. To be on the safe side, I
4131 have increased the buffer size for the ${rfc2047: expansion operator from
4134 33. It is now permitted to omit both strings after an "if" condition; if the
4135 condition is true, the result is "true". As before, when the second string
4136 is omitted, a false condition yields an empty string. This makes it less
4137 cumbersome to write custom ACL and router conditions.
4139 34. Failure to deliver a bounce message always caused it to be frozen, even if
4140 there was an errors_to setting on the router. The errors_to setting is now
4143 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4144 canonical form (fully expanded) before being placed in
4145 $sender_host_address.
4147 36. The table in the code that translates DNS record types into text (T_A to
4148 "A" for instance) was missing entries for NS and CNAME. It is just possible
4149 that this could have caused confusion if both these types were looked up
4150 for the same domain, because the text type is used as part of Exim's
4151 per-process caching. But the chance of anyone hitting this buglet seems
4154 37. The dnsdb lookup has been extended in a number of ways.
4156 (1) There is a new type, "zns", which walks up the domain tree until it
4157 finds some nameserver records. It should be used with care.
4159 (2) There is a new type, "mxh", which is like "mx" except that it returns
4160 just the host names, not the priorities.
4162 (3) It is now possible to give a list of domains (or IP addresses) to be
4163 looked up. The behaviour when one of the lookups defers can be
4164 controlled by a keyword.
4166 (4) It is now possible to specify the separator character for use when
4167 multiple records are returned.
4169 38. The dnslists ACL condition has been extended: it is now possible to supply
4170 a list of IP addresses and/or domains to be looked up in a particular DNS
4173 39. Added log_selector=+queue_time_overall.
4175 40. When running the queue in the test harness, wait just a tad after forking a
4176 delivery process, to get repeatability of debugging output.
4178 41. Include certificate and key file names in error message when GnuTLS fails
4179 to set them up, because the GnuTLS error message doesn't include the name
4180 of the failing file when there is a problem reading it.
4182 42. Allow both -bf and -bF in the same test run.
4184 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4186 44. The "Exiscan patch" is now merged into the mainline Exim source.
4188 45. Sometimes the final signoff response after QUIT could fail to get
4189 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4190 < 0 before doing a fflush(). This bug looks as though it goes back to the
4191 introduction of TLS in release 3.20, but "sometimes" must have been rare
4192 because the tests only now provoked it.
4194 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4195 (this can affect the format of dates).
4197 47. exim_tidydb, when checking for the continued existence of a message for
4198 which it has found a message-specific retry record, was not finding
4199 messages that were in split spool directories. Consequently, it was
4200 deleting retry records that should have stayed in existence.
4202 48. Steve fixed some bugs in eximstats.
4204 49. The SPA authentication driver was not abandoning authentication and moving
4205 on to the next authenticator when an expansion was forced to fail,
4206 contradicting the general specification for all authenticators. Instead it
4207 was generating a temporary error. It now behaves as specified.
4209 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4210 (the order specifies the preference for clients). The order is now AES256,
4211 AES128, 3DES, ARCFOUR128.
4213 51. Small patch to Sieve code - explicitly set From: when generating an
4216 52. Exim crashed if a remote delivery caused a very long error message to be
4217 recorded - for instance if somebody sent an entire SpamAssassin report back
4218 as a large number of 550 error lines. This bug was coincidentally fixed by
4219 increasing the size of one of Exim's internal buffers (big_buffer) that
4220 happened as part of the Exiscan merge. However, to be on the safe side, I
4221 have made the code more robust (and fixed the comments that describe what
4224 53. Now that there can be additional text after "Completed" in log lines (if
4225 the queue_time_overall log selector is set), a one-byte patch to exigrep
4226 was needed to allow it to recognize "Completed" as not the last thing in
4229 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4230 patch that reportedly fixes this has been added. I am not expert enough to
4231 create a test for it. This is what the patch creator wrote:
4233 "I found a little strange behaviour of ldap code when working with
4234 Windows 2003 AD Domain, where users was placed in more than one
4235 Organization Units. When I tried to give exim partial DN, the exit code
4236 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4237 But simultaneously result of request was absolutely normal ldap result,
4238 so I produce this patch..."
4240 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4241 REFERENCE, so I have modified the code to exclude the patch when that macro
4244 55. Some experimental protocols are using DNS PTR records for new purposes. The
4245 keys for these records are domain names, not reversed IP addresses. The
4246 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4247 leaves it alone. Component reversal etc. now happens only for IP addresses.
4250 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4252 57. Double the size of the debug message buffer (to 2048) so that more of very
4253 long debug lines gets shown.
4255 58. The exicyclog utility now does better if the number of log files to keep
4256 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4258 59. Two changes related to the smtp_active_hostname option:
4260 (1) $smtp_active_hostname is now available as a variable.
4261 (2) The default for smtp_banner uses $smtp_active_hostname instead
4262 of $primary_hostname.
4264 60. The host_aton() function is supposed to be passed a string that is known
4265 to be a valid IP address. However, in the case of IPv6 addresses, it was
4266 not checking this. This is a hostage to fortune. Exim now panics and dies
4267 if the condition is not met. A case was found where this could be provoked
4268 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4269 components; fortuitously, this particular loophole had already been fixed
4270 by change 4.50/55 above.
4272 If there are any other similar loopholes, the new check in host_aton()
4273 itself should stop them being exploited. The report I received stated that
4274 data on the command line could provoke the exploit when Exim was running as
4275 exim, but did not say which command line option was involved. All I could
4276 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4277 running as the user.
4280 61. There was a buffer overflow vulnerability in the SPA authentication code
4281 (which came originally from the Samba project). I have added a test to the
4282 spa_base64_to_bits() function which I hope fixes it.
4285 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4286 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4288 63. The daemon start-up calls getloadavg() while still root for those OS that
4289 need the first call to be done as root, but it missed one case: when
4290 deliver_queue_load_max is set with deliver_drop_privilege. This is
4291 necessary for the benefit of the queue runner, because there is no re-exec
4292 when deliver_drop_privilege is set.
4294 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4295 This has been fixed.
4297 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4298 was in use, was not putting the data itself into the right store pool;
4299 consequently, it could be overwritten for a subsequent message in the same
4300 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4303 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4305 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4306 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4307 the function were treating the return as a boolean value, which happened to
4308 work because 0=false and not-0=true, but is not correct code.
4310 68. The host_aton() function was not handling scoped IPv6 addresses (those
4311 with, for example, "%eth0" on the end) correctly.
4313 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4314 negated items (that is, ~something) in unsigned ints. Some compilers
4315 apparently mutter when there is no cast.
4317 70. If an address verification called from an ACL failed, and did not produce a
4318 user-specific message (i.e. there was only a "system" message), nothing was
4319 put in $acl_verify_message. In this situation, it now puts the system
4322 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4323 session; change 4.31/43 added the unwanted input to the log line - except
4324 that it did not do this in the start of session case. It now does.
4326 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4327 This is wrong and can cause the other end to generate a synchronization
4328 error if it is another Exim or anything else that does the synchronization
4329 check. A QUIT command is no longer sent after a timeout.
4331 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4332 during host lookups.
4334 74. The defer_ok option of callout verification was not working if it was used
4335 when verifying addresses in header lines, that is, for this case:
4337 verify = header_sender/callout=defer_ok
4339 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4340 those file descriptors could be used for SMTP connections. If anything
4341 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4342 could be sent to the SMTP client, causing chaos. The daemon now opens
4343 stdin, stdout, and stderr to /dev/null when it puts itself into the
4346 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4347 log by default. The user can override this with suitable Perl magic.
4349 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4350 the log message when discard triggers, was not working for the DATA ACL or
4351 for the non-SMTP ACL.
4353 78. Error message wording change in sieve.c.
4355 79. If smtp_accept_max_per_host was set, the number of connections could be
4356 restricted to fewer than expected, because the daemon was trying to set up
4357 a new connection before checking whether the processes handling previous
4358 connections had finished. The check for completed processes is now done
4359 earlier. On busy systems, this bug wouldn't be noticed because something
4360 else would have woken the daemon, and it would have reaped the completed
4363 80. If a message was submitted locally by a user whose login name contained one
4364 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4365 It caused a spool format error. I have fixed the spool reading code. A
4366 related problem was that the "from" clause in the Received: line became
4367 illegal because of the space(s). It is now covered by ${quote_local_part.
4369 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4372 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4374 83. Patch from Sieve maintainer to fix unterminated string problem in
4375 "vacation" handling.
4377 84. Some minor changes to the Linux configuration files to help with other
4378 OS variants using glibc.
4380 85. One more patch for Sieve to update vacation handling to latest spec.
4383 ----------------------------------------------------
4384 See the note above about the 4.44 and 4.50 releases.
4385 ----------------------------------------------------
4391 1. Change 4.43/35 introduced a bug that caused file counts to be
4392 incorrectly computed when quota_filecount was set in an appendfile
4395 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4396 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4399 3. Give more explanation in the error message when the command for a transport
4400 filter fails to execute.
4402 4. There are several places where Exim runs a non-Exim command in a
4403 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4404 was being done only for the command run by the queryprogram router. It is
4405 now done for all such subprocesses. The other cases are: ${run, transport
4406 filters, and the commands run by the lmtp and pipe transports.
4408 5. Some older OS have a limit of 256 on the maximum number of file
4409 descriptors. Exim was using setrlimit() to set 1000 as a large value
4410 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4411 systems. I've change it so that if it can't get 1000, it tries for 256.
4413 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4414 was an oversight, and furthermore, ever since the addition of extra
4415 controls (e.g. 4.43/32), the checks on when to allow different forms of
4416 "control" were broken. There should now be diagnostics for all cases when a
4417 control that does not make sense is encountered.
4419 7. $recipients is now available in the predata ACL (oversight).
4421 8. Tidy the search cache before the fork to do a delivery from a message
4422 received from the command line. Otherwise the child will trigger a lookup
4423 failure and thereby defer the delivery if it tries to use (for example) a
4424 cached ldap connection that the parent has called unbind on.
4426 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4427 of $address_data from the recipient verification was clobbered by the
4428 sender verification.
4430 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4431 was its contents. (It was OK if the option was not defined at all.)
4433 11. A "Completed" log line is now written for messages that are removed from
4434 the spool by the -Mrm option.
4436 12. $host_address is now set to the target address during the checking of
4437 ignore_target_hosts.
4439 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4440 being passed; this would have caused $sender_host_name to have been used if
4441 matching the list had actually called for a host name (not very likely,
4442 since this list is usually IP addresses). A host name is now passed as
4445 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4446 code that specifies a non-restarting handler (typically sigaction() in
4447 modern systems) in an attempt to fix a rare and obscure crash bug.
4449 15. Narrowed the window for a race in the daemon that could cause it to ignore
4450 SIGCHLD signals. This is not a major problem, because they are used only to
4451 wake it up if nothing else does.
4453 16. A malformed maildirsize file could cause Exim to calculate negative values
4454 for the mailbox size or file count. Odd effects could occur as a result.
4455 The maildirsize information is now recalculated if the size or filecount
4458 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4459 support for a long time. Removed HAVE_SYS_VFS_H.
4461 18. Updated exipick to current release from John Jetmore.
4463 19. Allow an empty sender to be matched against a lookup in an address list.
4464 Previously the only cases considered were a regular expression, or an
4467 20. Exim went into a mad DNS lookup loop when doing a callout where the
4468 host was specified on the transport, if the DNS lookup yielded more than
4471 21. The RFC2047 encoding function was originally intended for short strings
4472 such as real names; it was not keeping to the 75-character limit for
4473 encoded words that the RFC imposes. It now respects the limit, and
4474 generates multiple encoded words if necessary. To be on the safe side, I
4475 have increased the buffer size for the ${rfc2047: expansion operator from
4478 22. Failure to deliver a bounce message always caused it to be frozen, even if
4479 there was an errors_to setting on the router. The errors_to setting is now
4482 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4483 canonical form (fully expanded) before being placed in
4484 $sender_host_address.
4486 24. Updated eximstats to version 1.33
4488 25. Include certificate and key file names in error message when GnuTLS fails
4489 to set them up, because the GnuTLS error message doesn't include the name
4490 of the failing file when there is a problem reading it.
4492 26. Expand error message when OpenSSL has problems setting up cert/key files.
4495 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4496 (this can affect the format of dates).
4498 28. exim_tidydb, when checking for the continued existence of a message for
4499 which it has found a message-specific retry record, was not finding
4500 messages that were in split spool directories. Consequently, it was
4501 deleting retry records that should have stayed in existence.
4503 29. eximstats updated to version 1.35
4504 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4505 1.35 - bugfix such that pie charts by volume are generated correctly
4507 30. The SPA authentication driver was not abandoning authentication and moving
4508 on to the next authenticator when an expansion was forced to fail,
4509 contradicting the general specification for all authenticators. Instead it
4510 was generating a temporary error. It now behaves as specified.
4512 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4513 (the order specifies the preference for clients). The order is now AES256,
4514 AES128, 3DES, ARCFOUR128.
4516 31. Small patch to Sieve code - explicitly set From: when generating an
4519 32. Exim crashed if a remote delivery caused a very long error message to be
4520 recorded - for instance if somebody sent an entire SpamAssassin report back
4521 as a large number of 550 error lines. This bug was coincidentally fixed by
4522 increasing the size of one of Exim's internal buffers (big_buffer) that
4523 happened as part of the Exiscan merge. However, to be on the safe side, I
4524 have made the code more robust (and fixed the comments that describe what
4527 33. Some experimental protocols are using DNS PTR records for new purposes. The
4528 keys for these records are domain names, not reversed IP addresses. The
4529 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4530 leaves it alone. Component reversal etc. now happens only for IP addresses.
4533 34. The host_aton() function is supposed to be passed a string that is known
4534 to be a valid IP address. However, in the case of IPv6 addresses, it was
4535 not checking this. This is a hostage to fortune. Exim now panics and dies
4536 if the condition is not met. A case was found where this could be provoked
4537 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4538 components; fortuitously, this particular loophole had already been fixed
4539 by change 4.50/55 or 4.44/33 above.
4541 If there are any other similar loopholes, the new check in host_aton()
4542 itself should stop them being exploited. The report I received stated that
4543 data on the command line could provoke the exploit when Exim was running as
4544 exim, but did not say which command line option was involved. All I could
4545 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4546 running as the user.
4549 35. There was a buffer overflow vulnerability in the SPA authentication code
4550 (which came originally from the Samba project). I have added a test to the
4551 spa_base64_to_bits() function which I hope fixes it.
4554 36. The daemon start-up calls getloadavg() while still root for those OS that
4555 need the first call to be done as root, but it missed one case: when
4556 deliver_queue_load_max is set with deliver_drop_privilege. This is
4557 necessary for the benefit of the queue runner, because there is no re-exec
4558 when deliver_drop_privilege is set.
4560 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4561 was in use, was not putting the data itself into the right store pool;
4562 consequently, it could be overwritten for a subsequent message in the same
4563 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4566 38. Sometimes the final signoff response after QUIT could fail to get
4567 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4568 < 0 before doing a fflush(). This bug looks as though it goes back to the
4569 introduction of TLS in release 3.20, but "sometimes" must have been rare
4570 because the tests only now provoked it.
4576 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4577 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4578 Now it returns an integer. A number of calls were still expecting a T/F
4579 return. Fortuitously, in all cases, the tests worked in OK situations,
4580 which is the norm. However, things would have gone wrong on any write
4581 failures on the smtp file descriptor. This function is used when sending
4582 messages over SMTP and also when doing verify callouts.
4584 2. When Exim is called to do synchronous delivery of a locally submitted
4585 message (the -odf or -odi options), it no longer closes stderr before doing
4588 3. Implemented the mua_wrapper option.
4590 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4592 5. Implemented the functions header_remove(), header_testname(),
4593 header_add_at_position(), and receive_remove_recipient(), and exported them
4596 6. If an ACL "warn" statement specified the addition of headers, Exim already
4597 inserted X-ACL-Warn: at the start if there was no header name. However, it
4598 was not making this test for the second and subsequent header lines if
4599 there were newlines in the string. This meant that an invalid header could
4600 be inserted if Exim was badly configured.
4602 7. Allow an ACL "warn" statement to add header lines at the start or after all
4603 the Received: headers, as well as at the end.
4605 8. Added the rcpt_4xx retry error code.
4607 9. Added postmaster_mailfrom=xxx to callout verification option.
4609 10. Added mailfrom=xxxx to the callout verification option, for verify=
4612 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4613 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4615 12. Inserted some casts to stop certain compilers warning when using pointer
4616 differences as field lengths or precisions in printf-type calls (mostly
4617 affecting debugging statements).
4619 13. Added optional readline() support for -be (dynamically loaded).
4621 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4622 same clock tick as a message's arrival, so that its received time was the
4623 same as the "first fail" time on the retry record, and that message
4624 remained on the queue past the ultimate address timeout, every queue runner
4625 would try a delivery (because it was past the ultimate address timeout) but
4626 after another failure, the ultimate address timeout, which should have then
4627 bounced the address, did not kick in. This was a "< instead of <=" error;
4628 in most cases the first failure would have been in the next clock tick
4629 after the received time, and all would be well.
4631 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4632 being recognized when the domain list was tested by the match_domain
4633 condition in an expansion string.
4635 16. Added the ${str2b64: operator.
4637 17. Exim was always calling setrlimit() to set a large limit for the number of
4638 processes, without checking whether the existing limit was already
4639 adequate. (It did check for the limit on file descriptors.) Furthermore,
4640 errors from getrlimit() and setrlimit() were being ignored. Now they are
4641 logged to the main and panic logs, but Exim does carry on, to try to do its
4642 job under whatever limits there are.
4644 18. Imported PCRE 5.0.
4646 19. Trivial typo in log message " temporarily refused connection" (the leading
4649 20. If the log selector return_path_on_delivery was set and an address was
4650 redirected to /dev/null, the delivery process crashed because it assumed
4651 that a return path would always be set for a "successful" delivery. In this
4652 case, the whole delivery is bypassed as an optimization, and therefore no
4655 21. Internal re-arrangement: the function for sending a challenge and reading
4656 a response while authentication was assuming a zero-terminated challenge
4657 string. It's now changed to take a pointer and a length, to allow for
4658 binary data in such strings.
4660 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4662 23. Exim was not respecting finduser_retries when seeking the login of the
4663 uid under which it was called; it was always trying 10 times. (The default
4664 setting of finduser_retries is zero.) Also, it was sleeping after the final
4665 failure, which is pointless.
4667 24. Implemented tls_on_connect_ports.
4669 25. Implemented acl_smtp_predata.
4671 26. If the domain in control=submission is set empty, Exim assumes that the
4672 authenticated id is a complete email address when it generates From: or
4673 Sender: header lines.
4675 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4676 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4677 chown and chgrp in /bin and hostname in /usr/bin.
4679 28. Exim was keeping the "process log" file open after each use, just as it
4680 does for the main log. This opens the possibility of it remaining open for
4681 long periods when the USR1 signal hits a daemon. Occasional processlog
4682 errors were reported, that could have been caused by this. Anyway, it seems
4683 much more sensible not to leave this file open at all, so that is what now
4686 29. The long-running daemon process does not normally write to the log once it
4687 has entered its main loop, and it closes the log before doing so. This is
4688 so that log files can straightforwardly be renamed and moved. However,
4689 there are a couple of unusual error situations where the daemon does write
4690 log entries, and I had neglected to close the log afterwards.
4692 30. The text of an SMTP error response that was received during a remote
4693 delivery was being truncated at 512 bytes. This is too short for some of
4694 the long messages that one sometimes sees. I've increased the limit to
4697 31. It is now possible to make retry rules that apply only when a message has a
4698 specific sender, in particular, an empty sender.
4700 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4701 it possible to be selective about when SMTP synchronization is enforced.
4703 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4705 32. Implemented hosts_connection_nolog.
4707 33. Added an ACL for QUIT.
4709 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4712 35. Added mailbox_size and mailbox_filecount to appendfile.
4714 36. Added control = no_multiline_responses to ACLs.
4716 37. There was a bug in the logic of the code that waits for the clock to tick
4717 in the case where the clock went backwards by a substantial amount such
4718 that the microsecond fraction of "now" was more than the microsecond
4719 fraction of "then" (but the whole seconds number was less).
4721 38. Added support for the libradius Radius client library this is found on
4722 FreeBSD (previously only the radiusclient library was supported).
4728 1. When certain lookups returned multiple values in the form name=value, the
4729 quoting of the values was not always being done properly. Specifically:
4730 (a) If the value started with a double quote, but contained no whitespace,
4732 (b) If the value contained whitespace other than a space character (i.e.
4733 tabs or newlines or carriage returns) it was not quoted.
4734 This fix has been applied to the mysql and pgsql lookups by writing a
4735 separate quoting function and calling it from the lookup code. The fix
4736 should probably also be applied to nisplus, ibase and oracle lookups, but
4737 since I cannot test any of those, I have not disturbed their existing code.
4739 2. A hit in the callout cache for a specific address caused a log line with no
4740 reason for rejecting RCPT. Now it says "Previous (cached) callout
4741 verification failure".
4743 3. There was an off-by-one bug in the queryprogram router. An over-long
4744 return line was truncated at 256 instead of 255 characters, thereby
4745 overflowing its buffer with the terminating zero. As well as fixing this, I
4746 have increased the buffer size to 1024 (and made a note to document this).
4748 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4749 when Exim is waiting for an SMTP response from a remote server, Exim
4750 restarts its select() call on the socket, thereby resetting its timeout.
4751 This is not a problem when such interrupts are rare. Somebody set up a cron
4752 job to run exiwhat every 2 minutes, which is less than the normal select()
4753 timeout (5 or 10 minutes). This meant that the select() timeout never
4754 kicked in because it was always reset. I have fixed this by comparing the
4755 time when an interrupt arrives with the time at the start of the first call
4756 to select(). If more time than the timeout has elapsed, the interrupt is
4757 treated as a timeout.
4759 5. Some internal re-factoring in preparation for the addition of Sieve
4760 extensions (by MH). In particular, the "personal" test is moved to a
4761 separate function, and given an option for scanning Cc: and Bcc: (which is
4762 not set for Exim filters).
4764 6. When Exim created an email address using the login of the caller as the
4765 local part (e.g. when creating a From: or Sender: header line), it was not
4766 quoting the local part when it contained special characters such as @.
4768 7. Installed new OpenBSD configuration files.
4770 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4771 try to make them clearer.
4773 9. Callout options, other than the timeout value, were being ignored when
4774 verifying sender addresses in header lines. For example, when using
4776 verify = header_sender/callout=no_cache
4778 the cache was (incorrectly) being used.
4780 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4781 only the Cygwin environment.
4783 11. When return_path_on_delivery was set as a log selector, if different remote
4784 addresses in the same message used different return paths and parallel
4785 remote delivery occurred, the wrong values would sometimes be logged.
4786 (Whenever a remote delivery process finished, the return path value from
4787 the most recently started remote delivery process was logged.)
4789 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4790 lines when AUTH and/or TLS are in use. This is the "received protocol"
4791 field. Exim used to use "asmtp" for authenticated SMTP, without any
4792 indication (in the protocol name) for TLS use. Now it follows the RFC and
4793 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4794 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4795 names appear in log lines as well as in Received: header lines.
4797 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4798 extensions, and comparison tests, and to fix some bugs.
4800 14. Changes to the "personal" filter test:
4802 (1) The test was buggy in that it was just doing the equivalent of
4803 "contains" tests on header lines. For example, if a user's address was
4804 anne@some.where, the "personal" test would incorrectly be true for
4806 To: susanne@some.where
4808 This test is now done by extracting each address from the header in turn,
4809 and checking the entire address. Other tests that are part of "personal"
4810 are now done using regular expressions (for example, to check local parts
4811 of addresses in From: header lines).
4813 (2) The list of non-personal local parts in From: addresses has been
4814 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4815 taken from the Sieve specification recommendations.
4817 (3) If the message contains any header line starting with "List-" it is
4818 treated as non-personal.
4820 (4) The test for "circular" in the Subject: header line has been removed
4821 because it now seems ill-conceived.
4823 15. Minor typos in src/EDITME comments corrected.
4825 16. Installed latest exipick from John Jetmore.
4827 17. If headers_add on a router specified a text string that was too long for
4828 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4829 of string_sprintf() is now avoided.
4831 18. $message_body_size was not set (it was always zero) when running the DATA
4832 ACL and the local_scan() function.
4834 19. For the "mail" command in an Exim filter, no default was being set for
4835 the once_repeat time, causing a random time value to be used if "once" was
4836 specified. (If the value happened to be <= 0, no repeat happened.) The
4837 default is now 0s, meaning "never repeat". The "vacation" command was OK
4838 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4839 (I found it when inspecting the code).
4841 20. There is now an overall timeout for performing a callout verification. It
4842 defaults to 4 times the callout timeout, which applies to individual SMTP
4843 commands during the callout. The overall timeout applies when there is more
4844 than one host that can be tried. The timeout is checked before trying the
4845 next host. This prevents very long delays if there are a large number of
4846 hosts and all are timing out (e.g. when the network connections are timing
4847 out). The value of the overall timeout can be changed by specifying an
4848 additional sub-option for "callout", called "maxwait". For example:
4850 verify = sender/callout=5s,maxwait=20s
4852 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4853 to the end before writing, but this should make it even safer).
4855 22. Exim was forgetting that it had advertised PIPELINING for the second and
4856 subsequent messages on an SMTP connection. It was also not resetting its
4857 memory on STARTTLS and an internal HELO.
4859 23. When Exim logs an SMTP synchronization error within a session, it now
4860 records whether PIPELINING has been advertised or not.
4862 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4863 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4864 rather than long int.
4866 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4868 26. Added the never_mail option to autoreply.
4874 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4875 crash if the getsockname() call failed; this can happen if a connection is
4876 closed very soon after it is established. The problem was simply in the
4877 order in which certain operations were done, causing Exim to try to write
4878 to the SMTP stream before it had set up the file descriptor. The bug has
4879 been fixed by making things happen in the correct order.
4885 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4886 before the connection was closed, thus losing the rejection response.
4888 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4889 some early Solaris releases, but causes trouble in current releases where
4890 socklen_t is defined.
4892 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4895 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4898 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4899 that releasing the top of it at the end releases what was used for sub-
4900 expansions (unless the block got too big). However, discard this block if
4901 the first thing is a variable or header, so that we can use its block when
4902 it is dynamic (useful for very large $message_headers, for example).
4904 6. Lookups now cache *every* query, not just the most recent. A new, separate
4905 store pool is used for this. It can be recovered when all lookup caches are
4906 flushed. Lookups now release memory at the end of their result strings.
4907 This has involved some general refactoring of the lookup sources.
4909 7. Some code has been added to the store_xxx() functions to reduce the amount
4910 of flapping under certain conditions.
4912 8. log_incoming_interface used to affect only the <= reception log lines. Now
4913 it causes the local interface and port to be added to several more SMTP log
4914 lines, for example "SMTP connection from", and rejection lines.
4916 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4918 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4920 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4921 could be overwritten at the end of the current message (or the start of a
4922 new message if it was set in a HELO ACL). The value is now preserved for
4923 the duration of the SMTP connection.
4925 12. If a transport had a headers_rewrite setting, and a matching header line
4926 contained an unqualified address, that address was qualified, even if it
4927 did not match any rewriting rules. The underlying bug was that the values
4928 of the flags that permit the existence of unqualified sender and recipient
4929 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4930 non-local messages, and by -bnq for local messages) were not being
4931 preserved with the message after it was received.
4933 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4934 "next input=" as part of the text comprising the host identity instead of
4935 the correct text. The code was using the same buffer for two different
4936 strings. However, depending on which order the printing function evaluated
4937 its arguments, the bug did not always show up. Under Linux, for example, my
4938 test suite worked just fine.
4940 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4941 doesn't work with some older versions of Perl. It has been changed to "my",
4942 which in any case is probably the better facility to use.
4944 15. A really picky compiler found some instances of statements for creating
4945 error messages that either had too many or two few arguments for the format
4948 16. The size of the buffer for calls to the DNS resolver has been increased
4949 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4950 for addresses that have a lot of PTR records. This alleviates a problem; it
4951 does not fully solve it.
4953 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4954 buffer now truncates the list and logs the incident, which is the same
4955 action as happens when Exim is looking up a host name and its aliases.
4956 Previously in this situation something unpredictable would happen;
4957 sometimes it was "internal error: store_reset failed".
4959 18. If a server dropped the connection unexpectedly when an Exim client was
4960 using GnuTLS and trying to read a response, the client delivery process
4961 crashed while trying to generate an error log message.
4963 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4964 string, for example:
4966 warn message = H1: something\nH2: something
4968 the text was added as a single header line from Exim's point of view
4969 though it ended up OK in the delivered message. However, searching for the
4970 second and subsequent header lines using $h_h2: did not work. This has been
4971 fixed. Similarly, if a system filter added multiple headers in this way,
4972 the routers could not see them.
4974 20. Expanded the error message when iplsearch is called with an invalid key to
4975 suggest using net-iplsearch in a host list.
4977 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4978 ACLs are no longer actually imposed (and a message to that effect is
4981 22. If a "gecos" field in a passwd entry contained escaped characters, in
4982 particular, if it contained a \" sequence, Exim got it wrong when building
4983 a From: or a Sender: header from that name. A second bug also caused
4984 incorrect handling when an unquoted " was present following a character
4985 that needed quoting.
4987 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4988 was not being matched caselessly.
4990 24. Arranged for all hyphens in the exim.8 source to be escaped with
4993 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4994 properly. Recipient callout cache records were still being keyed to include
4995 the sender, even when use_sender was set false. This led to far more
4996 callouts that were necessary. The sender is no longer included in the key
4997 when use_sender is false.
4999 26. Added "control = submission" modifier to ACLs.
5001 27. Added the ${base62d: operator to decode base 62 numbers.
5003 28. dnsdb lookups can now access SRV records.
5005 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5006 the configuration file.
5008 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5009 (-v) mode. This makes the output for a verbose queue run more intelligible.
5011 31. Added a use_postmaster feature to recipient callouts.
5013 32. Added the $body_zerocount variable, containing the number of binary zero
5014 bytes in the message body.
5016 33. The time of last modification of the "new" subdirectory is now used as the
5017 "mailbox time last read" when there is a quota error for a maildir
5020 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5022 35. Added +ignore_unknown as a special item in host lists.
5024 36. Code for decoding IPv6 addresses in host lists is now included, even if
5025 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5026 address was recognized as an IP address, but was then not correctly decoded
5027 into binary, causing unexpected and incorrect effects when compared with
5034 1. Very minor rewording of debugging text in manualroute to say "list of
5035 hosts" instead of "hostlist".
5037 2. If verify=header_syntax was set, and a header line with an unqualified
5038 address (no domain) and a large number of spaces between the end of the
5039 name and the colon was received, the reception process suffered a buffer
5040 overflow, and (when I tested it) crashed. This was caused by some obsolete
5041 code that should have been removed. The fix is to remove it!
5043 3. When running in the test harness, delay a bit after writing a bounce
5044 message to get a bit more predictability in the log output.
5046 4. Added a call to search_tidyup() just before forking a reception process. In
5047 theory, someone could use a lookup in the expansion of smtp_accept_max_
5048 per_host which, without the tidyup, could leave open a database connection.
5050 5. Added the variables $recipient_data and $sender_data which get set from a
5051 lookup success in an ACL "recipients" or "senders" condition, or a router
5052 "senders" option, similar to $domain_data and $local_part_data.
5054 6. Moved the writing of debug_print from before to after the "senders" test
5057 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5058 problems for message scanning, either using a data ACL, or using
5059 local_scan() because the Received: header was not generated till after they
5060 were called (in order to set the time as the time of reception completion).
5061 I have revised the way this works. The header is now generated after the
5062 body is received, but before the ACL or local_scan() are called. After they
5063 are run, the timestamp in the header is updated.
5069 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5070 before starting a queue runner without re-exec. This happened only when
5071 deliver_drop_privilege was set or when the Exim user was set to root. The
5072 effect of the bug was that timeouts during subsequent deliveries caused
5073 crashes instead of being properly handled. The handler is now left at its
5074 default (and expected) setting.
5076 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5077 message, again when deliver_drop_privilege is set or Exim is run as root.
5078 The bug described in (1) was not present in this case, but the tidying up
5079 of the other signals was missing. I have made the two cases consistent.
5081 3. The ignore_target_hosts setting on a manualroute router was being ignored
5082 for hosts that were looked up using the /MX notation.
5084 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5087 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5088 operated on the sender address. After changing the $sender_address to <>
5089 for the sender address verify, Exim was re-instated it as the original
5090 (before rewriting) address, but remembering that it had rewritten it, so it
5091 wasn't rewriting it again. This bug also had the effect of breaking the
5092 sender address verification caching when the sender address was rewritten.
5094 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5095 This has been changed so that if the ip literal address matches
5096 ignore_target_hosts, the router declines.
5098 7. Added expansion conditions match_domain, match_address, and match_local_
5099 part (NOT match_host).
5101 8. The placeholder for the Received: header didn't have a length field set.
5103 9. Added code to Exim itself and to exim_lock to test for a specific race
5104 condition that could lead to file corruption when using MBX delivery. The
5105 issue is with the lockfile that is created in /tmp. If this file is removed
5106 after a process has opened it but before that process has acquired a lock,
5107 there is the potential for a second process to recreate the file and also
5108 acquire a lock. This could lead to two Exim processes writing to the file
5109 at the same time. The added code performs the same test as UW imapd; it
5110 checks after acquiring the lock that its file descriptor still refers to
5111 the same named file.
5113 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5114 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5117 11. Added the smtp_active_hostname option. If used, this will typically be made
5118 to depend on the incoming interface address. Because $interface_address is
5119 not set up until the daemon has forked a reception process, error responses
5120 that can happen earlier (such as "too many connections") no longer contain
5123 12. If an expansion in a condition on a "warn" statement fails because a lookup
5124 defers, the "warn" statement is abandoned, and the next ACL statement is
5125 processed. Previously this caused the whole ACL to be aborted.
5127 13. Added the iplsearch lookup type.
5129 14. Added ident_timeout as a log selector.
5131 15. Added tls_certificate_verified as a log selector.
5133 16. Added a global option tls_require_ciphers (compare the smtp transport
5134 option of the same name). This controls incoming TLS connections.
5136 17. I finally figured out how to make tls_require_ciphers do a similar thing
5137 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5138 before starting the TLS session.
5140 18. Tabs are now shown as \t in -bP output.
5142 19. If the log selector return_path_on_delivery was set, Exim crashed when
5143 bouncing a message because it had too many Received: header lines.
5145 20. If two routers both had headers_remove settings, and the first one included
5146 a superfluous trailing colon, the final name in the first list and the
5147 first name in the second list were incorrectly joined into one item (with a
5148 colon in the middle).
5154 1. Added -C and -D options to the exinext utility, mainly to make it easier
5155 to include in the automated testing, but these could be helpful when
5156 multiple configurations are in use.
5158 2. The exinext utility was not formatting the output nicely when there was
5159 an alternate port involved in the retry record key, nor when there was a
5160 message id as well (for retries that were specific to a specific message
5161 and a specific host). It was also confused by IPv6 addresses, because of
5162 the additional colons they contain. I have fixed the IPv4 problem, and
5163 patched it up to do a reasonable job for IPv6.
5165 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5166 delivery, the log line now contains "pipelined" if PIPELINING was used.
5168 4. An SMTP transport process used to panic and die if the bind() call to set
5169 an explicit outgoing interface failed. This has been changed; it is now
5170 treated in the same way as a connect() failure.
5172 5. A reference to $sender_host_name in the part of a conditional expansion
5173 that was being skipped was still causing a DNS lookup. This no longer
5176 6. The def: expansion condition was not recognizing references to header lines
5177 that used bh_ and bheader_.
5179 7. Added the _cache feature to named lists.
5181 8. The code for checking quota_filecount in the appendfile transport was
5182 allowing one more file than it should have been.
5184 9. For compatibility with Sendmail, the command line option
5192 and sets the incoming protocol and host name (for trusted callers). The
5193 host name and its colon can be omitted when only the protocol is to be set.
5194 Note the Exim already has two private options, -pd and -ps, that refer to
5195 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5196 "s", but I don't think that's a major issue.
5198 10. A number of refactoring changes to the code, none of which should affect
5201 (a) The number of logging options was getting close to filling up the
5202 32-bit word that was used as a bit map. I have split them into two classes:
5203 those that are passed in the argument to log_write(), and those that are
5204 only ever tested independently outside of that function. These are now in
5205 separate 32-bit words, so there is plenty of room for expansion again.
5206 There is no change in the user interface or the logging behaviour.
5208 (b) When building, for example, log lines, the code previously used a
5209 macro that called string_cat() twice, in order to add two strings. This is
5210 not really sufficiently general. Furthermore, there was one instance where
5211 it was actually wrong because one of the argument was used twice, and in
5212 one call a function was used. (As it happened, calling the function twice
5213 did not affect the overall behaviour.) The macro has been replaced by a
5214 function that can join an arbitrary number of extra strings onto a growing
5217 (c) The code for expansion conditions now uses a table and a binary chop
5218 instead of a serial search (which was left over from when there were very
5219 few conditions). Also, it now recognizes conditions like "pam" even when
5220 the relevant support is not compiled in: a suitably worded error message is
5221 given if an attempt is made to use such a condition.
5223 11. Added ${time_interval:xxxxx}.
5225 12. A bug was causing one of the ddress fields not to be passed back correctly
5226 from remote delivery subprocesses. The field in question was not being
5227 subsequently used, so this caused to problems in practice.
5229 13. Added new log selectors queue_time and deliver_time.
5231 14. Might have fixed a bug in maildirsizefile handling that threw up
5232 "unexpected character" debug warnings, and recalculated the data
5233 unnecessarily. In any case, I expanded the warning message to give more
5236 15. Added the message "Restricted characters in address" to the statements in
5237 the default ACL that block characters like @ and % in local parts.
5239 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5240 Three changes have been made:
5242 (a) There was a serious bug; a negative response to MAIL caused the whole
5243 recipient domain to be cached as invalid, thereby blocking all messages
5244 to all local parts at the same domain, from all senders. This bug has
5245 been fixed. The domain is no longer cached after a negative response to
5246 MAIL if the sender used is not empty.
5248 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5251 (c) A new callout option, "use_sender" has been added for people who want
5252 the modified behaviour.
5258 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5261 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5262 indeed breaks things for older releases.
5264 3. Added additional logging to the case where there is a problem reading data
5265 from a filter that is running in a subprocess using a pipe, in order to
5266 try to track down a specific problem.
5268 4. Testing facility fudge: when running in the test harness and attempting
5269 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5270 getting "No route to host". Convert this to a timeout.
5272 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5275 6. Some OS don't have socklen_t but use size_t instead. This affects the
5276 fifth argument of getsockopt() amongst other things. This is now
5277 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5278 can be set for individual OS. I have set it for SunOS5, OSF1, and
5279 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5280 some earlier ones do not.
5282 7. Change 4.30/15 was not doing the test caselessly.
5284 8. The standard form for an IPv6 address literal was being rejected by address
5285 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5286 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5287 this, as well as the form without the "IPv6" on the front (but only when
5288 address literals are enabled, of course).
5290 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5292 10. Exim crashed if a message with an empty sender address specified by -f
5293 encountered a router with an errors_to setting. This could be provoked only
5294 by a command such as
5298 where an empty string was supplied; "<>" did not hit this bug.
5300 11. Installed PCRE release 4.5.
5302 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5303 remained set. It is now erased.
5305 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5306 times from message ids (which are base 36 rather than the normal 62).
5308 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5309 were being counted as actual protocol errors, and logged if the log
5310 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5311 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5312 and DATA following a set of rejected RCPTs do not count as protocol errors.
5313 In other words, Exim assumes they were pipelined, though this may not
5314 actually be the case. Of course, in all cases the client gets an
5315 appropriate error code.
5317 15. If a lookup fails in an ACL condition, a message about the failure may
5318 be available; it is used if testing the ACL cannot continue, because most
5319 such messages specify what the cause of the deferral is. However, some
5320 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5321 that caused an old message to be retained and used if a later statement
5322 caused a defer, replacing the real cause of the deferral.
5324 16. If an IP address had so many PTR records that the DNS lookup buffer
5325 was not large enough to hold them, Exim could crash while trying to process
5326 the truncated data. It now detects and logs this case.
5328 17. Further to 4.21/58, another change has been made: if (and only if) the
5329 first line of a message (the first header line) ends with CRLF, a bare LF
5330 in a subsequent header line has a space inserted after it, so as not to
5331 terminate the header.
5333 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5334 unnecessarily, used atoi() instead of strtol(), and didn't check the
5335 termination when getting file sizes from file names by regex.
5337 19. Completely re-implemented the support for maildirsize files, in the light
5338 of a number of problems with the previous contributed implementation
5339 (4.30/29). In particular:
5341 . If the quota is zero, the maildirsize file is maintained, but no quota is
5344 . If the maildir directory does not exist, it is created before any attempt
5345 to write a maildirsize file.
5347 . The quota value in the file is just a cache; if the quota is changed in
5348 the transport, the new value overrides.
5350 . A regular expression is available for excluding directories from the
5353 20. The autoreply transport checks the characters in options that define the
5354 message's headers; it allows continued headers, but it was checking with
5355 isspace() after an embedded newline instead of explicitly looking for a
5358 21. If all the "regular" hosts to which an address was routed had passed their
5359 expiry times, and had not reached their retry times, the address was
5360 bounced, even if fallback hosts were defined. Now Exim should go on to try
5363 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5364 equivalent code in the SMTP transport. Some hosts send humungous responses
5365 to HELO/EHLO, more than 1024 it seems.
5367 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5368 gives compiler warnings in some environments. I've now done it "properly",
5371 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5372 (because of problems with the built-in one) was declared to return uschar *
5373 instead of char *, causing compiler failure.
5375 25. Fixed a file descriptor leak when processing alias/forward files.
5377 26. Fixed a minor format string issue in dbfn.c.
5379 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5381 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5382 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5385 29. When a sender address is verified, it is cached, to save repeating the test
5386 when there is more than one recipient in a message. However, when the
5387 verification involves a callout, it is possible for different callout
5388 options to be set for different recipients. It is too complicated to keep
5389 track of this in the cache, so now Exim always runs a verification when a
5390 callout is required, relying on the callout cache for the optimization.
5391 The overhead is duplication of the address routing, but this should not be
5394 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5395 to be verified with callout=postmaster, and the main callout worked but the
5396 postmaster check failed, the verification correctly failed. However, if a
5397 subsequent RCPT command asked for sender verification *without* the
5398 postmaster check, incorrect caching caused this verification also to fail,
5401 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5402 it was not caching the DNS options (qualify_single, search_parents) that
5403 were used when the lookup failed. A subsequent lookup with different
5404 options therefore always gave the same answer, though there were cases
5405 where it should not have. (Example: a "domains = !$mx_any" option on a
5406 dnslookup router: the "domains" option is always processed without any
5407 widening, but the router might have qualify_single set.) Now Exim uses the
5408 cached value only when the same options are set.
5410 32. Added John Jetmore's "exipick" utility to the distribution.
5412 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5413 than a timeout (e.g. a certificate is required, and is not provided), an
5414 Exim server now closes the connection immediately. Previously it waited for
5415 the client to close - but if the client is SSL, it seems that they each
5416 wait for each other, leading to a delay before one of them times out.
5418 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5419 maintained 0.8.x compatibility because I don't think many are using it, and
5420 it is clearly obsolete.
5422 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5425 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5426 client certificate was expired. A simple patch fixes this, though I don't
5427 understand the full logic of why the verify callback is called multiple
5430 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5431 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5432 which causes problems with some clients (such as the Certicom SSL Plus
5433 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5434 disables the coutermeasure allowing Eudora to connect."
5436 38. Exim was not checking that a write() to a log file succeeded. This could
5437 lead to Bad Things if a log got too big, in particular if it hit a file
5438 size limit. Exim now panics and dies if it cannot write to a log file, just
5439 as it does if it cannot open a log file.
5441 39. Modified OS/Makefile-Linux so that it now contains
5443 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5445 The two -D definitions ensure that Exim is compiled with large file
5446 support, which makes it possible to handle log files that are bigger than
5449 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5450 instance) a domain was checked against a named list that involved a lookup,
5451 causing $domain_data to be set, then another domain was checked against the
5452 same list, then the first domain was re-checked, the value of $domain_data
5453 after the final check could be wrong. In particular, if the second check
5454 failed, it could be set empty. This bug probably also applied to
5457 41. The strip_trailing_dot option was not being applied to the address given
5458 with the -f command-line option.
5460 42. The code for reading a message's header from the spool was incrementing
5461 $received_count, but never initializing it. This meant that the value was
5462 incorrect (doubled) while delivering a message in the same process in which
5463 it was received. In the most common configuration of Exim, this never
5464 happens - a fresh exec is done - but it can happen when
5465 deliver_drop_privilege is set.
5467 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5468 it now includes up to 150 characters of the unexpected data in the log
5471 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5472 and building data strings. The size of both of these buffers was 10 000
5473 bytes - far larger than anybody would *ever* want, thought I. Needless to
5474 say, somebody hit the limit. I have increased the maximum line length to
5475 20 000 and the maximum data length of concatenated lines to 100 000. I have
5476 also fixed two bugs, because there was no checking on these buffers. Tsk,
5477 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5478 buffer is too small.
5480 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5481 lsearch lookups. Now it does.
5483 46. When parsing a route_list item in a manualroute router, a fixed-length
5484 buffer was used for the list of hosts. I made this 1024 bytes long,
5485 thinking that nobody would ever have a list of hosts that long. Wrong.
5486 Somebody had a whole pile of complicated expansion conditions, and the
5487 string was silently truncated, leading to an expansion error. It turns out
5488 that it is easier to change to an unlimited length (owing to other changes
5489 that have happened since this code was originally written) than to build
5490 structure for giving a limitation error. The length of the item that
5491 expands into the list of hosts is now unlimited.
5493 47. The lsearch lookup could not handle data where the length of text line was
5494 more than 4095 characters. Such lines were truncated, leading to shortened
5495 data being returned. It should now handle lines of any length.
5497 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5498 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5501 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5502 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5503 no longer happens in comment lines. A list of replacements is now placed
5504 at the head of all of the source files, except those whose only change is
5505 to replace PERL_COMMAND in the very first #! line.
5507 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5508 messages on the queue, with a bottom-up merge sort, using code contributed
5509 by Michael Haardt. This should make operations like -bp somewhat faster on
5510 large queues. It won't affect queue runners, except when queue_run_in_order
5513 51. Installed eximstats 1.31 in the distribution.
5515 52. Added support for SRV lookups to the dnslookup router.
5517 53. If an ACL referred to $message_body or $message_body_end, the value was not
5518 reset for any messages that followed in the same SMTP session.
5520 54. The store-handling optimization for building very long strings was not
5521 differentiating between the different store pools. I don't think this
5522 actually made any difference in practice, but I've tidied it.
5524 55. While running the routers to verify a sender address, $sender_address
5525 was still set to the sender address. This is wrong, because when routing to
5526 send a bounce to the sender, it would be empty. Therefore, I have changed
5527 it so that, while verifying a sender address, $sender_address is set to <>.
5528 (There is no change to what happens when verifying a recipient address.)
5530 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5531 target A or AAAA records (if not already returned) without resetting the
5532 qualify_single or search_parents options of the DNS resolver. These are
5533 inappropriate in this case because the targets of MX and SRV records must
5534 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5535 target that, when qualified, matched something in the local domain. These
5536 two options are now turned off when doing these lookups.
5538 57. It seems that at least some releases of Reiserfs (which does not have the
5539 concept of a fixed number of inodes) returns zero and not -1 for the
5540 number of available inodes. This interacted badly with check_spool_inodes,
5541 which assumed that -1 was the "no such thing" setting. What I have done is
5542 to check that the total number of inodes is greater than zero before doing
5543 the test of how many are available.
5545 58. When a "warn" ACL statement has a log_message modifier, the message is
5546 remembered, and not repeated. This is to avoid a lot of repetition when a
5547 message has many recipients that cause the same warning to be written.
5548 Howewer, Exim was preserving the list of already written lines for an
5549 entire SMTP session, which doesn't seem right. The memory is now reset if a
5550 new message is started.
5552 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5553 debugging output unless log_rewrite was also set.
5555 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5556 of (int)(handle) when we know that handle contains (void *)(-1).
5558 61. The Exim daemon panic-logs an error return when it closes the incoming
5559 connection. However "connection reset by peer" seems to be common, and
5560 isn't really an error worthy of noting specially, so that particular error
5563 62. When Exim is trying to find all the local interfaces, it used to panic and
5564 die if the ioctl to get the interface flags failed. However, it seems that
5565 on at least one OS (Solaris 9) it is possible to have an interface that is
5566 included in the list of interfaces, but for which you get a failure error
5567 for this call. This happens when the interface is not "plumbed" into a
5568 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5569 failure of the "get flags" call assumes that the interface is down.
5571 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5572 makes life easier for people who are doing arithmetic on fields extracted
5573 from dates, where you often get leading zeros that should not be
5574 interpreted as octal.
5576 64. Added qualify_domain to the redirect router, to override the global
5579 65. If a pathologically long header line contained very many addresses (the
5580 report of this problem mentioned 10 000) and each of them was rewritten,
5581 Exim could use up a very large amount of memory. (It kept on making new
5582 copies of the header line as it rewrote, and never released the old ones.)
5583 At the expense of a bit more processing, the header rewriting function has
5584 been changed so that it no longer eats memory in this way.
5586 66. The generation of the Received: header has been moved from the time that a
5587 message starts to be received, to the time that it finishes. The timestamp
5588 in the Received: header should now be very close to that of the <= log
5589 line. There are two side-effects of this change:
5591 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5592 logged header lines no longer include the local Received: line, because
5593 it has not yet been created. The same applies to a copy of the message
5594 that is returned to a non-SMTP sender when a message is rejected.
5596 (b) When a filter file is tested using -bf, no additional Received: header
5597 is added to the test message. After some thought, I decided that this
5600 This change does not affect the value of $received_for. It is still set
5601 after address rewriting, but before local_scan() is called.
5603 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5605 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5606 gave an unhelpful panic error message, and a defer error. I have managed to
5607 change this behaviour so that it now rejects any supplied certificate,
5608 which seems right, as the list of acceptable certificates is empty.
5610 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5611 gave an unhelpful defer error. I have not managed to make this reject any
5612 supplied certificates, but the error message it gives is "no certificate
5613 supplied", which is not helpful.
5615 70. exigrep's output now also includes lines that are not associated with any
5616 message, but which match the given pattern. Implemented by a patch from
5617 Martin Sluka, which also tidied up the Perl a bit.
5619 71. Recipient callout verification, like sender verification, was using <> in
5620 the MAIL FROM command. This isn't really the right thing, since the actual
5621 sender may affect whether the remote host accepts the recipient or not. I
5622 have changed it to use the actual sender in the callout; this means that
5623 the cache record is now keyed on a recipient/sender pair, not just the
5624 recipient address. There doesn't seem to be a real danger of callout loops,
5625 since a callout by the remote host to check the sender would use <>.
5626 [SEE ABOVE: changed after hitting problems.]
5628 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5629 temporary errors. However, in the case of such a code being given after
5630 the end of a data transmission (i.e. after ".") Exim was failing to write
5631 a retry record for the message. (Yes, there was some broken host that was
5632 actually sending 8xx at this point.)
5634 73. An unknown lookup type in a host list could cause Exim to panic-die when
5635 the list was checked. (An example that provoked this was putting <; in the
5636 middle of a list instead of at the start.) If this happened during a DATA
5637 ACL check, a -D file could be left lying around. This kind of configuration
5638 error no longer causes Exim to die; instead it causes a defer errror. The
5639 incident is still logged to the main and panic logs.
5641 74. Buglet left over from Exim 3 conversion. The message "too many messages
5642 in one connection" was written to the rejectlog but not the mainlog, except
5643 when address rewriting (yes!) was being logged.
5645 75. Added write_rejectlog option.
5647 76. When a system filter was run not as root (that is, when system_filter_user
5648 was set), the values of the $n variables were not being returned to the
5649 main process; thus, they were not subsequently available in the $sn
5652 77. Added +return_path_on_delivery log selector.
5654 78. A connection timeout was being treated differently from recipients deferred
5655 when testing hosts_max_try with a message that was older than the host's
5656 retry timeout. (The host should not be counted, thus allowing all hosts to
5657 be tried at least once before bouncing.) This may have been the cause of an
5658 occasionally reported bug whereby a message would remain on the queue
5659 longer than the retry timeout, but would be bounced if a delivery was
5660 forced. I say "may" because I never totally pinned down the problem;
5661 setting up timeout/retry tests is difficult. See also the next item.
5663 79. The ultimate address timeout was not being applied to errors that involved
5664 a combination of host plus message (for example, a timeout on a MAIL
5665 command). When an address resolved to a number of possible hosts, and they
5666 were not all tried for each delivery (e.g. because of hosts_max_try), a
5667 message could remain on the queue longer than the retry timeout.
5669 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5672 81. Fixed an obscure SMTP outgoing bug which required at least the following
5673 conditions: (a) there was another message waiting for the same server;
5674 (b) the server returned 5xx to all RCPT commands in the first message so
5675 that the message was not completed; (c) the server dropped the connection
5676 or gave a negative response to the RSET that Exim sends to abort the
5677 transaction. The observed case was a dropped connection after DATA that had
5678 been sent in pipelining mode. That is, the server had advertised PIPELINING
5679 but was not implementing it correctly. The effect of the bug was incorrect
5680 behaviour, such as trying another host, and this could lead to a crash.
5686 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5687 and daemon.c were passed as pointers to ints; they should have been
5688 pointers to socklen_t variables (which are typically unsigned ints).
5690 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5693 3. Fixed a really odd bug that affected only the testing scheme; patching a
5694 certain fixed string in the binary changed the value of another string that
5695 happened to be identical to the end of the original first string.
5697 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5698 name", it returns that address as the IP address. On some operating
5699 systems (e.g. Solaris), it also passes back the IP address string as the
5700 "host name". However, on others (e.g. Linux), it passes back an empty
5701 string. Exim wasn't checking for this, and was changing the host name to an
5702 empty string, assuming it had been canonicized.
5704 5. Although rare, it is permitted to have more than one PTR record for a given
5705 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5706 all the names associated with an address, because they do in Solaris.
5707 However, it seems that they do not in Linux for data that comes from the
5708 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5709 I found this out when I moved to a new Linux workstation and tried to run
5710 the Exim test suite.
5712 To get round this problem I have changed the code so that it now does its
5713 own call to the DNS to look up PTR records when searching for a host name.
5714 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5715 addresses that are only in /etc/hosts are still found.
5717 This behaviour is, however, controlled by an option called host_lookup_
5718 order, which defaults to "bydns:byaddr". If people want to use the other
5719 order, or indeed, just use one or the other means of lookup, they can
5720 specify it in this variable.
5722 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5723 some operating systems, this comes back from gethostbyaddr() as an empty
5724 string, and this is what Exim used to test for. However, it seems that in
5725 other systems, "." is yielded. Exim now tests for this case too.
5727 7. The values of check_spool_space and check_log_space are now held internally
5728 as a number of kilobytes instead of an absolute number of bytes. If a
5729 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5730 kilobyte. This means that much larger values can be stored.
5732 8. Exim monitor: an attempt to get the action menu when not actually pointing
5733 at a message produces an empty menu entitled "No message selected". This
5734 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5735 no entries in it ("Shell widget menu has zero width and/or height"). So I
5736 have added a single, blank menu entry in this case.
5738 9. Added ${quote_local_part.
5740 10. MIME decoding is now applied to the contents of Subject: header lines when
5743 11. Now that a reference to $sender_host_address automatically causes a reverse
5744 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5745 host lookup before query-style lookups in lists that might use this
5746 variable. This has therefore been abolished, and the "net-" prefix is no
5747 longer necessary for query-style lookups.
5749 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5750 have been a typo for LFLAGS, so it has been changed.
5752 13. The install script calls Exim with "-C /dev/null" in order to find the
5753 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5754 to be output. Howeve, since Exim outputs its version number before the
5755 error, it didn't break the script. It just looked ugly. I fixed this by
5756 always allowing "-C /dev/null" if the caller is root.
5758 14. Ignore overlarge ACL variable number when reading spool file - insurance
5759 against a later release with more variables having written the file.
5761 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5762 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5763 this, as well as the form without the "IPv6" on the front.
5765 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5766 OS/Makefile-Darwin file.
5768 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5769 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5771 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5772 variable. However, some OS use ints for inodes. I've added cast to long int
5773 to get rid of the compiler warning.
5775 19. I had forgotten to lock out "/../" in configuration file names when
5776 ALT_CONFIG_PREFIX was set.
5778 20. Routers used for verification do not need to specify transports. However,
5779 if such a router generated a host list, and callout was configured, Exim
5780 crashed, because it could not find a port number from the (non-existent)
5781 transport. It now assumes port 25 in this circumstance.
5783 21. Added the -t option to exigrep.
5785 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5786 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5787 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5788 them is set, LOOKUP_LSEARCH is forced.
5790 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5792 24. Added sender and host information to the "rejected by local_scan()" log
5793 line; previously there was no indication of these.
5795 25. Added .include_if_exists.
5797 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5798 Linux. It turns out that not all file systems support this. Apparently some
5799 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5800 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5801 not supported on the file descriptor, is now ignored when Exim is trying to
5802 sync a directory. This applies only to Linux.
5804 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5806 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5807 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5810 29. Added support for maildirsize files from supplied patch (modified a bit).
5812 30. The use of :fail: followed by an empty string could lead Exim to respond to
5813 sender verification failures with (e.g.):
5815 550 Verification failed for <xxx>
5816 550 Sender verify failed
5818 where the first response line was missing the '-' that indicates it is not
5819 the final line of the response.
5821 31. The loop for finding the name of the user that called Exim had a hardwired
5822 limit of 10; it now uses the value of finduser_retries, which is used for
5823 all other user lookups.
5825 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5828 33. Exim was neglecting to zero errno before one call of strtol() when
5829 expanding a string and expecting an integer value. On some systems this
5830 resulted in spurious "integer overflow" errors. Also, it was casting the
5831 result into an int without checking.
5833 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5834 did not work. The code looks as if it has *never* worked, though it appears
5835 to have been documented since at least releast 1.62. I have made it work.
5837 35. The "timeout_DNS" error in retry rules, also documented since at least
5838 1.62, also never worked. As it isn't clear exactly what this means, and
5839 clearly it isn't a major issue, I have abolished the feature by treating it
5840 as "timeout", and writing a warning to the main and panic logs.
5842 36. The display of retry rules for -brt wasn't always showing the error code
5845 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5846 timeout_connect_A, timeout_connect_MX.
5848 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5849 to the empty sender.
5851 39. The daemon was not analysing the content of -oX till after it had closed
5852 stderr and disconnected from the controlling terminal. This meant that any
5853 syntax errors were only noted on the panic log, and the return code from
5854 the command was 0. By re-arranging the code a little, I've made the
5855 decoding happen first, so such errors now appear on stderr, and the return
5856 code is 1. However, the actual setting up of the sockets still happens in
5857 the disconnected process, so errors there are still only recorded on the
5860 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5861 connections (as happens on some IP stacks) was logged at start up time as
5862 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5863 it from "IPv6 and IPv4", which means that two separate sockets are being
5866 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5867 says whether AF_INET or AF_INET6 was passed as an argument.
5869 42. Exiwhat output was messed up when time zones were included in log
5872 43. Exiwhat now gives more information about the daemon's listening ports,
5873 and whether -tls-on-connect was used.
5875 44. The "port" option of the smtp transport is now expanded.
5877 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5878 silently ignored. Now an error message is written to the main and panic
5881 46. There's a new ACL modifier called "logwrite" which writes to a log file
5882 as soon as it is encountered.
5884 47. Added $local_user_uid and $local_user_gid at routing time.
5886 48. Exim crashed when trying to verify a sender address that was being
5889 49. Exim was recognizing only a space character after ".include". It now also
5890 recognizes a tab character.
5892 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5893 extracting the relevant information from the specification. The man page no
5894 longer contains scrambled data for the -d option, and I've added a section
5895 at the front about calling Exim under different names.
5897 51. Added "extra_headers" argument to the "mail" command in filter files.
5899 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5902 53. Installed eximstats 1.29.
5904 54. Added transport_filter_timeout as a generic transport option.
5906 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5907 Cc: header lines. This was required by RFC 822, but it not required by RFC
5910 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5911 incoming messages that did not have them. Now it does so only if the
5912 message originates locally, that is, if there is no associated remote host
5913 address. When Resent- header lines are present, this applies to the Resent-
5914 lines rather than the non-Resent- lines.
5916 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5917 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5919 58. Messages for configuration errors now include the name of the main
5920 configuration file - useful now that there may be more than one file in a
5921 list (.included file names were always shown).
5923 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5924 for those rare installations that do not start the daemon as root or run it
5925 setuid root. I've cut out the call to initgroups() if the daemon is not
5928 60. The Exim user and group can now be bound into the binary as text strings
5929 that are looked up at the start of Exim's processing.
5931 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5933 62. Added $mailstore_basename variable.
5935 63. Installed patch to sieve.c from Michael Haardt.
5937 64. When Exim failed to open the panic log after failing to open the main log,
5938 the original message it was trying to log was written to stderr and debug
5939 output, but if they were not available (the usual case in production), it
5940 was lost. Now it is written to syslog before the two lines that record the
5941 failures to open the logs.
5943 65. Users' Exim filters run in subprocesses under the user's uid. It is
5944 possible for a "deliver" command or an alias in a "personal" command to
5945 provoke an address rewrite. If logging of address rewriting is configured,
5946 this fails because the process is not running as root or exim. There may be
5947 a better way of dealing with this, but for the moment (because 4.30 needs
5948 to be released), I have disabled address rewrite logging when running a
5949 filter in a non-root, non-exim process.
5955 1. The buildconfig auxiliary program wasn't quoting the value set for
5956 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5957 not defined. This bug was masked in 4.22 by the effect that was fixed in
5960 2. Some messages that were rejected after a message id was allocated were
5961 shown as "incomplete" by exigrep. It no longer does this for messages that
5962 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5964 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5965 have allow_domain_literals set, the ID did not get logged in the <= line.
5966 Domain literals are now always recognized in Message-ID: header lines.
5968 4. The first argument for a ${extract expansion item is the key name or field
5969 number. Leading and trailing spaces in this item were not being ignored,
5970 causing some misleading effects.
5972 5. When deliver_drop_privilege was set, single queue runner processes started
5973 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5974 same command in the process it spins off) were not dropping privilege.
5976 6. When the daemon running as "exim" started a queue runner, it always
5977 re-executed Exim in the spun-off process. This is a waste of effort when
5978 deliver_drop_privilege is set. The new process now just calls the
5979 queue-runner function directly.
5985 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5988 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5989 ignored. Though the use of -oP was forcing the writing of a pid file, it
5990 was always written to the default place.
5992 3. If the message "no IP address found for host xxxx" is generated during
5993 incoming verification, it is now followed by identification of the incoming
5994 connection (so you can more easily find what provoked it).
5996 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5998 5. Added some features to "harden" Exim a bit more against certain attacks:
6000 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6001 be put in Local/Makefile. This is like the never_users runtime option,
6002 but it cannot be overridden. The default setting is "root".
6004 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6005 prefix string with which any file named in a -C command line option
6008 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6009 is retained for -C and -D only if the caller of Exim is root. Without
6010 it, the exim user may also use -C and -D and retain privilege.
6012 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6013 command line option is disabled.
6015 6. Macro names set by the -D option must start with an upper case letter, just
6016 like macro names defined in the configuration file.
6018 7. Added "dereference=" facility to LDAP.
6020 8. Two instances of the typo "uknown" in the source files are fixed.
6022 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6023 the Configure-Makefile script screwed up while processing it.
6025 10. Incorporated PCRE 4.4.
6027 11. The SMTP synchronization check was not operating right at the start of an
6028 SMTP session. For example, it could not catch a HELO sent before the client
6029 waited for the greeting. There is now a check for outstanding input at the
6030 point when the greeting is written. Because of the duplex, asynchronous
6031 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6032 way, but not yet received, when the check is performed.
6034 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6035 on TCP/IP sockets, because this apparently causes some broken clients to
6038 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6039 unchanged) from the Cygwin maintainer.
6041 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6042 was supported instead of testing for mbx. Effectively a typo.
6044 15. The spa authenticator server code was not checking that the input it
6045 received was valid base64.
6047 16. The debug output line for the "set" modifier in ACLs was not showing the
6048 name of the variable that was being set.
6050 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6052 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6053 lookup on its own. Something else was needed to trigger the lookup. For
6054 example, a match in host_lookup or the need for a host name in a host list.
6055 Now, if $sender_host_name is referenced and the host name has not yet been
6056 looked up, a lookup is performed. If the lookup fails, the variable remains
6057 empty, and $host_lookup_failed is set to "1".
6059 19. Added "eqi" as a case-independent comparison operator.
6061 20. The saslauthd authentication condition could segfault if neither service
6062 nor realm was specified.
6064 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6065 error message that was logged was misleading, and incoming SMTP
6066 connections were dropped. The message is now more accurate, and temporary
6067 errors are given to SMTP connections.
6069 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6070 (except RSET) with a 421 error, until QUIT is received. However, it was
6071 failing to send a response to QUIT.
6073 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6074 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6075 ACL is now run after the helo_try_verify_hosts code.
6077 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6078 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6079 case-independent, so other case variants are also recognized). Apparently
6080 some systems use these upper case variants.
6082 25. If more than two messages were waiting for the same host, and a transport
6083 filter was specified for the transport, Exim sent two messages over the
6084 same TCP/IP connection, and then failed with "socket operation on non-
6085 socket" when it tried to send the third.
6087 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6089 27. The extern definition of crypt16() in expand.c was not being excluded when
6090 the OS had its own crypt16() function.
6092 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6093 as a preferred synonym for return_size_limit, both as an option and as an
6096 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6098 30. Changed the default configuration ACL to relax the local part checking rule
6099 for addresses that are not in any local domains. For these addresses,
6100 slashes and pipe symbols are allowed within local parts, but the sequence
6101 /../ is explicitly forbidden.
6103 31. SPA server authentication was not clearing the challenge buffer before
6106 32. log_message in a "warn" ACL statement was writing to the reject log as
6107 well as to the main log, which contradicts the documentation and doesn't
6108 seem right (because no rejection is happening). So I have stopped it.
6110 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6111 However, I am unable to do any testing of this.
6113 34. Fixed an infelicity in the appendfile transport. When checking directories
6114 for a mailbox, to see if any needed to be created, it was accidentally
6115 using path names with one or more superfluous leading slashes; tracing
6116 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6118 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6119 contents are added to the log line that is written for every discarded
6120 recipient. (Previously a log_message setting was ignored.)
6122 36. The ${quote: operator now quotes the string if it is empty.
6124 37. The install script runs exim in order to find its version number. If for
6125 some reason other than non-existence or emptiness, which it checks, it
6126 could not run './exim', it was installing it with an empty version number,
6127 i.e. as "exim-". This error state is now caught, and the installation is
6130 38. An argument was missing from the function that creates an error message
6131 when Exim fails to connect to the socket for saslauthd authentication.
6132 This could cause Exim to crash, or give a corrupted message.
6134 39. Added isip, isip4, and isip6 to ${if conditions.
6136 40. The ACL variables $acl_xx are now saved with the message, and can be
6137 accessed later in routers, transports, and filters.
6139 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6140 strings in the file are not string-expanded.
6142 42. If a MAIL command specified a SIZE value that was too large to fit into an
6143 int variable, the check against message_size_limit failed. Such values are
6144 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6145 day this will have to be increased, but I don't think I want to be around
6146 when emails are that large.
6153 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6154 iconv() is not standard in FreeBSD.
6156 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6157 IPv6 enabled. The observed symptom was a segmentation fault on return from
6158 the function os_common_find_running_interfaces() in src/os.c.
6160 3. In the check_special_case() function in daemon.c I had used "errno" as an
6161 argument name, which causes warnings on some systems. This was basically a
6162 typo, since it was named "eno" in the comments!
6164 4. The code that waits for the clock to tick (at a resolution of some fraction
6165 of a second) so as to ensure message-id uniqueness was always waiting for
6166 at least one whole tick, when it could have waited for less. [This is
6167 almost certainly not relevant at current processor speeds, where it is
6168 unlikely to ever wait at all. But we try to future-proof.]
6170 5. The function that sleeps for a time interval that includes fractions of a
6171 second contained a race. It did not block SIGALRM between setting the
6172 timer, and suspending (a couple of lines later). If the interval was short
6173 and the sigsuspend() was delayed until after it had expired, the suspension
6174 never ended. On busy systems this could lead to processes getting stuck for
6177 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6178 process, before it forks any delivery processes. The open lookup caching
6179 mechanism meant that the open file or database connection was passed into
6180 the delivery process. The problem was that delivery processes always tidy
6181 up cached lookup data. This could cause a problem for the next delivery
6182 process started by the queue runner, because the external queue runner
6183 process does not know about the closure. So the next delivery process
6184 still has data in the lookup cache. In the case of a file lookup, there was
6185 no problem because closing a file descriptor in a subprocess doesn't affect
6186 the parent. However, if the lookup was caching a connection to a database,
6187 the connection was closed, and the second delivery process was likely to
6188 see errors such as "PGSQL: query failed: server closed the connection
6189 unexpectedly". The problem has been fixed by closing all cached lookups
6190 in a queue runner before running a delivery process.
6192 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6193 seem to have the "const" qualifier which it has on other OS. I've
6196 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6197 *of the same type* (client or server) with the same public name that an
6198 error should be diagnosed.
6200 9. When Exim looked up a host name for an IP address, but failed to find the
6201 original IP address when looking up the host name (a safety check), it
6202 output the message "<ip address> does not match any IP for NULL", which was
6203 confusing, to say the least. The bug was that the host name should have
6204 appeared instead of "NULL".
6206 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6207 user that is built into the binary, and the -C or -D options is used, root
6208 privilege is dropped before the configuration file is read. In addition,
6209 logging is switched to stderr instead of the normal log files. If the
6210 configuration then re-defines the Exim user, the unprivileged environment
6211 is probably not what is expected, so Exim logs a panic warning message (but
6214 However, if deliver_drop_privilege is set, the unprivileged state may well
6215 be exactly what is intended, so the warning has been cut out in that case,
6216 and Exim is allowed to try to write to its normal log files.
6222 1. smtp_return_error_details was not giving details for temporary sender
6223 or receiver verification errors.
6225 2. Diagnose a configuration error if two authenticators have the same public
6228 3. Exim used not to create the message log file for a message until the first
6229 delivery attempt. This could be confusing when incoming messages were held
6230 for policy or load reasons. The message log file is now created at the time
6231 the message is received, and an initial "Received" line is written to it.
6233 4. The automatically generated man page for command line options had a minor
6234 bug that caused no ill effects; however, a more serious problem was that
6235 the procedure for building the man page automatically didn't always
6236 operate. Consequently, release 4.20 contains an out-of-date version. This
6237 shouldn't happen again.
6239 5. When building Exim with embedded Perl support, the script that builds the
6240 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6241 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6243 6. The freeze_tell option was not being used for messages that were frozen on
6244 arrival, either by an ACL or by local_scan().
6246 7. Added the smtp_incomplete_transaction log selector.
6248 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6249 was accepting AUTH without a new EHLO.
6251 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6252 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6255 10. Logging of TCP/IP connections (when configured) now happens in the main
6256 daemon process instead of the child process, so that the TCP/IP connection
6257 count is more accurate (but it can never be perfect).
6259 11. The use of "drop" in a nested ACL was not being handled correctly in the
6260 outer ACL. Now, if condition failure induced by the nested "drop" causes
6261 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6262 or "require"), the connection is dropped.
6264 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6265 that yield "discard" can now be used with an "accept" or a "discard" verb,
6266 but an error is generated for any others (because I can't see a useful way
6267 to define what should happen).
6269 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6270 are now processed in the same way as lines in the Exim configuration file.
6271 In particular, continuation lines are supported.
6273 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6275 15. Added -ti meaning -t -i.
6277 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6278 domains, and warn by logging if others are found.
6280 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6281 returned from GIFCONF when Exim is trying to find the list of interfaces on
6282 a host. The code in os.c has been modified to copy each ifreq to an aligned
6283 structure in all cases.
6285 Also, in some cases, the returned ifreq's were being copied to a 'struct
6286 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6287 means the last couple of bytes of an IPv6 address could be chopped if the
6288 ifreq contained only a normal sockaddr (14 bytes storage).
6290 18. Named domain lists were not supported in the hosts_treat_as_local option.
6291 An entry such as +xxxx was not recognized, and was treated as a literal
6294 19. Ensure that header lines added by a DATA ACL are included in the reject log
6295 if the ACL subsequently rejects the message.
6297 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6298 MD5 (which is deprecated).
6300 21. When testing a filter file using -bf, Exim was writing a message when it
6301 took the sender from a "From " line in the message, but it was not doing so
6302 when it took $return_path from a Return-Path: header line. It now does.
6304 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6305 with a valid header line field name (a series of printing characters
6306 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6308 23. Changed "disc" in the source to "disk" to conform to the documentation and
6309 the book and for uniformity.
6311 24. Ignore Sendmail's -Ooption=value command line item.
6313 25. When execve() failed while trying to run a command in a pipe transport,
6314 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6315 could be confused with a return value of 69 from the command itself. This
6316 has been changed to 127, the value the shell returns if it is asked to run
6317 a non-existent command. The wording for the related log line suggests a
6318 non-existent command as the problem.
6320 26. If received_header_text expands to an empty string, do not add a Received:
6321 header line to the message. (Well, it adds a token one on the spool, but
6322 marks it "old" so that it doesn't get used or transmitted.)
6324 27. Installed eximstats 1.28 (addition of -nt option).
6326 28. There was no check for failure on the call to getsockname() in the daemon
6327 code. This can fail if there is a shortage of resources on the system, with
6328 ENOMEM, for example. A temporary error is now given on failure.
6330 29. Contrary to the C standard, it seems that in some environments, the
6331 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6332 program. Exim now does this explicitly; it affects the formatting of
6333 timestamps using strftime().
6335 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6336 complaints. I've now initialized all the variables, to avoid this.
6338 32. Header lines added by a system filter were not being "seen" during
6339 transport-time rewrites.
6341 33. The info_callback() function passed to OpenSSL is set up with type void
6342 (*)(SSL *, int, int), as described somewhere. However, when calling the
6343 function (actually a macro) that sets it up, the type void(*)() is
6344 expected. I've put in a cast to prevent warnings from picky compilers.
6346 34. If a DNS black list lookup found a CNAME record, but there were no A
6347 records associated with the domain it pointed at, Exim crashed.
6349 35. If a DNS black list lookup returned more than one A record, Exim ignored
6350 all but the first. It now scans all returned addresses if a particular IP
6351 value is being sought. In this situation, the contents of the
6352 $dnslist_value variable are a list of all the addresses, separated by a
6355 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6356 to accept a host name and all its aliases if the forward lookup for any of
6357 them yielded the IP address of the incoming connection. Now it accepts only
6358 those names whose forward lookup yields the correct IP address. Any other
6359 names are discarded. This closes a loophole whereby a rogue DNS
6360 administrator could create reverse DNS records to break through a
6361 wildcarded host restriction in an ACL.
6363 37. If a user filter or a system filter that ran in a subprocess used any of
6364 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6365 the wrong values were passed to the pipe command ($thisaddress had the
6366 value of $0, $0 had the value of $1, etc). This bug was introduced by
6367 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6369 38. Improved the line breaking for long SMTP error messages from ACLs.
6370 Previously, if there was no break point between 40 and 75 characters, Exim
6371 left the rest of the message alone. Two changes have been made: (a) I've
6372 reduced the minimum length to 35 characters; (b) if it can't find a break
6373 point between 35 and 75 characters, it looks ahead and uses the first one
6374 that it finds. This may give the occasional overlong line, but at least the
6375 remaining text gets split now.
6377 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6378 file descriptors might be low, and that setting 1000 would always raise it.
6379 It turns out that in some environments, the limit is already over 1000 and
6380 that lowering it causes trouble. So now Exim takes care not to decrease it.
6382 40. When delivering a message, the value of $return_path is set to $sender_
6383 address at the start of routing (routers may change the value). By an
6384 oversight, this default was not being set up when an address was tested by
6385 -bt or -bv, which affected the outcome if any router or filter referred to
6388 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6389 log only when "message" or "log_message" are set. However, if one of the
6390 conditions was an address verification, or a call to a nested ACL, the
6391 messages generated by the underlying test were being passed through. This
6392 no longer happens. The underlying message is available in $acl_verify_
6393 message for both "message" and "log_message" expansions, so it can be
6394 passed through if needed.
6396 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6397 new expansion $bh_ to give the encoded byte string without charset
6398 translation. Translation happens only if iconv() is available; HAVE_ICONV
6399 indicates this at build time. HEADERS_CHARSET gives the charset to
6400 translate to; headers_charset can change it in the configuration, and
6401 "headers charset" can change it in an individual filter file.
6403 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6404 that creates RFC 2047 encoded "words" labels them as that charset instead
6405 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6406 expansion operator; (ii) when Exim creates a From: line for a local
6407 message; (iii) when a header line is rewritten to include a "phrase" part.
6409 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6410 buggy, causing it to skip the first lines of messages whose message ID
6411 ended in 'D'. This would not have bitten before Exim release 4.14, because
6412 message IDs were unlikely to end in 'D' before then. The effect was to have
6413 incorrect size information for certain domains.
6415 45. #include "config.h" was missing at the start of the crypt16.c module. This
6416 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6419 46. If there was a timeout during a "random" callout check, Exim treated it as
6420 a failure of the random address, and carried on sending RSET and the real
6421 address. If the delay was just some slowness somewhere, the response to the
6422 original RCPT would be taken as a response to RSET and so on, causing
6423 mayhem of various kinds.
6425 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6426 when I implemented it. It didn't allow for the fact that some option values
6427 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6428 the right test for positive values.
6430 48. Domain names in DNS records are case-independent. Exim always looks them up
6431 in lower case. Some resolvers return domain names in exactly the case they
6432 appear in the zone file, that is, they may contain uppercase letters. Not
6433 all resolvers do this - some return always lower case. Exim was treating a
6434 change of case by a resolver as a change of domain, similar to a widening
6435 of a domain abbreviation. This triggered its re-routing code and so it was
6436 trying to route what was effectively the same domain again. This normally
6437 caused routing to fail (because the router wouldn't handle the domain
6438 twice). Now Exim checks for this case specially, and just changes the
6439 casing of the domain that it ultimately uses when it transmits the message
6442 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6445 50. If a filter generated a file delivery with a non-absolute name (possible if
6446 no home directory exists for the router), the forbid_file option was not
6449 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6450 the existing equality matching.
6452 52. Exim was using ints instead of ino_t variables in some places where it was
6453 dealing with inode numbers.
6455 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6456 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6457 called TMPDIR, and if it finds it is different, it changes its value.
6459 54. The smtp_printf() function is now made available to local_scan() so
6460 additional output lines can be written before returning. There is also an
6461 smtp_fflush() function to enable the detection of a dropped connection.
6462 The variables smtp_input and smtp_batched_input are exported to
6465 55. Changed the default runtime configuration: the message "Unknown user"
6466 has been removed from the ACL, and instead placed on the localuser router,
6467 using the cannot_route_message feature. This means that any verification
6468 failures that generate their own messages won't get overridden. Similarly,
6469 the "Unrouteable address" message that was in the ACL for unverifiable
6470 relay addresses has also been removed.
6472 56. Added hosts_avoid_esmtp to the smtp transport.
6474 57. The exicyclog script was not checking for the esoteric option
6475 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6476 will work only if exicyclog is run under the appropriate euid.
6478 58. Following a discussion on the list, the rules by which Exim recognises line
6479 endings on incoming messages have been changed. The -dropcr and drop_cr
6480 options are now no-ops, retained only for backwards compatibility. The
6481 following line terminators are recognized: LF CRLF CR. However, special
6482 processing applies to CR:
6484 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6485 nor a local message in the state where . is a terminator.
6487 (ii) If a bare CR is encountered in a header line, an extra space is added
6488 after the line terminator so as not to end the header. The reasoning
6489 behind this is that bare CRs in header lines are most likely either
6490 to be mistakes, or people trying to play silly games.
6492 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6493 was being incorrectly given as 18 bytes larger than it should have been.
6494 This is a VOB (very old bug).
6496 60. This may never have affected anything current, but just in case it has:
6497 When the local host is found other than at the start of a list of hosts,
6498 the local host, those with the same MX, and any that follow, are discarded.
6499 When the list in question was part of a longer list of hosts, the following
6500 hosts (not currently being processed) were also being discarded. This no
6501 longer happens. I'm not sure if this situation could ever has previously
6504 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6507 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6508 header. This is something that is recommended in a new Internet Draft, and
6509 is something that is documented as being done by Sendmail. There are two
6510 possible values. For messages generated by the autoreply transport, Exim
6513 Auto-Submitted: auto-replied
6515 whereas for all other generated messages (e.g. bounces) it adds
6517 Auto-Submitted: auto-generated
6519 63. The "personal" condition in filters now includes a test for the
6520 Auto-Submitted: header. If it contains the string "auto-" the message it
6521 not considered personal.
6523 64. Added rcpt_include_affixes as a generic transport option.
6525 65. Added queue_only_override (default true).
6527 66. Added the syslog_duplication option.
6529 67. If what should have been the first header line of a message consisted of
6530 a space followed by a colon, Exim was mis-interpreting it as a header line.
6531 It isn't of course - it is syntactically invalid and should therefore be
6532 treated as the start of the message body. The misbehaviour could have
6533 caused a number of strange effects, including loss of data in subsequent
6534 header lines, and spool format errors.
6536 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6537 client host had authenticated. This control can now be exercised by an ACL
6538 for more flexibility.
6540 69. By default, callouts do not happen when testing with -bh. There is now a
6541 variant, -bhc, which does actually run the callout code, including
6542 consulting and updating the callout cache.
6544 70. Added support for saslauthd authentication, courtesy of Alexander
6547 71. If statvfs() failed on the spool or log directories while checking their
6548 size for availability, Exim confusingly gave the error "space shortage".
6549 Furthermore, in debugging mode it crashed with a floating point exception.
6550 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6551 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6552 serious problem, Exim now writes to the main and panic logs when this
6553 happens, with details of the failure. It then refuses to accept the
6554 incoming message, giving the message "spool directory problem" or "log
6555 directory problem" with a 421 code for SMTP messages.
6557 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6558 0, 1, and 2 exist, because some OS complain for execs without them (see
6559 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6560 descriptors. However, the code omitted to check that the open succeeded,
6561 causing mysterious errors if for some reason the permissions on /dev/null
6562 got screwed. Now Exim writes a message to the main and panic logs, and
6563 bombs out if it can't open /dev/null.
6565 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6566 interact so that it is all more flexible. It is supposed to remain
6567 backwards compatible. Also added extra_local_interfaces.
6569 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6570 to bomb out with an assertion failure - to the client this appears as a
6571 connection drop. This problem occurs in the part of the code that was taken
6572 from the Samba project. Fortunately, the assertion is in a very simple
6573 function, so I have fixed this by reproducing the function inline in the
6574 one place where it is called, and arranging for authentication to fail
6575 instead of killing the process with assert().
6577 75. The SPA client code was not working when the server requested OEM rather
6578 than Unicode encoding.
6580 76. Added code to make require_files with a specific uid setting more usable in
6581 the case where statting the file as root fails - usually a non-root-mounted
6582 NFS file system. When this happens and the failure is EACCES, Exim now
6583 forks a subprocess and does the per-uid checking as the relevant uid.
6585 77. Added process_log_path.
6587 78. If log_file_path was not explicitly set, a setting of check_log_space or
6588 check_log_inodes was ignored.
6590 79. If a space check for the spool or log partitions fails, the incident is now
6591 logged. Of course, in the latter case the data may get lost...
6593 80. Added the %p formatting code to string_format() so that it can be used to
6594 print addresses in debug_print(). Adjusted all the address printing in the
6595 debugging in store.c to use %p rather than %d.
6597 81. There was a concern that a line of code in smtp_in.c could overflow a
6598 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6599 initially expressed, the concern was not well-founded, because trailing
6600 spaces are removed early. However, if the trailing spaces were followed by
6601 a NULL, they did not get removed, so the overflow was possible. Two fixes
6604 (a) I re-wrote the offending code in a cleaner fashion.
6605 (b) If an incoming SMTP command contains a NULL character, it is rejected
6608 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6609 runs initgroups(), so that if the Exim user is in any additional groups,
6610 they will be used during message reception.
6616 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob