1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 self-signed certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the source code
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
102 JH/27 Fix a possible security hole, wherein a process operating with the Exim
103 UID can gain a root shell. Credit to http://www.halfdog.net/ for
104 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
107 JH/28 Enable {spool,log} filesystem space and inode checks as default.
108 Main config options check_{log,spool}_{inodes,space} are now
109 100 inodes, 10MB unless set otherwise in the configuration.
111 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
112 Previously it only applied to the main-section connection policy
115 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
117 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
118 by me. Added RFC7919 DH primes as an alternative.
120 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
121 are not in the system include path.
123 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
124 GnuTLS, when a session startup failed (eg because the client disconnected)
125 Exim did stdio operations after fclose. This was exposed by a recent
126 change which nulled out the file handle after the fclose.
131 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
132 and 3.4.4 - once the server is enabled to respond to an OCSP request
133 it does even when not requested, resulting in a stapling non-aware
134 client dropping the TLS connection.
136 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
137 support variable-length bit vectors. No functional change.
139 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
140 The I= interface field on outgoing lines is now after the H= remote
141 host field, same as incoming lines. There is a separate
142 outgoing_interface log selector which allows you to disable the
145 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
146 If not running log_selector +smtp_connection the mainlog would be held
147 open indefinitely after a "too many connections" event, including to a
148 deleted file after a log rotate. Leave the per net connection logging
149 leaving it open for efficiency as that will be quickly detected by the
150 check on the next write.
152 HS/01 Bug 1671: Fix post transport crash.
153 Processing the wait-<transport> messages could crash the delivery
154 process if the message IDs didn't exist for some reason. When
155 using 'split_spool_directory=yes' the construction of the spool
156 file name failed already, exposing the same netto behaviour.
158 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
159 mime_regex ACL conditions.
161 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
162 to DSN fail messages (bounces): remote IP, remote greeting, remote response
163 to HELO, local diagnostic string.
165 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
166 log line to debug. Even when configured with a tls authenticator many
167 client connections are expected to not authenticate in this way, so
168 an authenticate fail is not an error.
170 HS/02 Add the Exim version string to the process info. This way exiwhat
171 gives some more detail about the running daemon.
173 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
174 matter for fast-change records such as DNSBLs.
176 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
177 retry record, even if constant. There may be multiple transports with
178 different interface settings and the retry behaviour needs to be kept
181 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
183 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
185 JH/10 Bug 840: fix log_defer_output option of pipe transport
187 JH/11 Bug 830: use same host for all RCPTS of a message, even under
188 hosts_randomize. This matters a lot when combined with mua_wrapper.
190 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
191 ${quote_pgsql:<string>} operator.
193 JH/13 Bug 1708: avoid misaligned access in cached lookup.
195 JH/14 Change header file name for freeradius-client. Relevant if compiling
196 with Radius support; from the Gentoo tree and checked under Fedora.
198 JH/15 Bug 1712: Introduce $prdr_requested flag variable
200 JH/16 Bug 1714: Permit an empty string as expansion result for transport
201 option transport_filter, meaning no filtering.
203 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
205 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
206 defaults to "*" (all hosts). The variable is now available when not built
207 with TLS, default unset, mainly to enable keeping the testuite sane.
208 If a server certificate is not supplied (via tls_certificate) an error is
209 logged, and clients will find TLS connections fail on startup. Presumably
210 they will retry in-clear.
211 Packagers of Exim are strongly encouraged to create a server certificate
212 at installation time.
214 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
215 with the $config_file variable.
217 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
218 in transport context, after the attempt, and per-recipient. The latter type
219 is per host attempted. The event data is the error message, and the errno
220 information encodes the lookup type (A vs. MX) used for the (first) host,
221 and the trailing two digits of the smtp 4xx reponse.
223 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
224 to write to mainlog (or rejectlog, paniclog) in the window between file
225 creation and permissions/ownership being changed. Particularly affects
226 installations where exicyclog is run as root, rather than exim user;
227 result is that the running daemon panics and dies.
229 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
231 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
232 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
233 "pri" and "weight". Note that the previous implicit priority given by the
234 list order is no longer honoured.
236 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
239 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
240 by defining SUPPORT_SOCKS.
242 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
243 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
244 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
245 variables are renamed to proxy_{local,external}_{address,port}.
247 JH/27 Move Internationalisation support from Experimental to mainline, enabled
248 for a build by defining SUPPORT_I18N
250 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
251 of the query string, and make ${quote_redis:} do that quoting.
253 JH/29 Move Events support from Experimental to mainline, enabled by default
254 and removable for a build by defining DISABLE_EVENT.
256 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
258 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
259 cached by the daemon.
261 JH/32 Move Redis support from Experimental to mainline, enabled for a build
262 by defining LOOKUP_REDIS. The libhiredis library is required.
264 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
265 keys are given for lookup.
267 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
268 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
269 only supported when built with TLS support. The PolarSSL SHA routines
270 are still used when the TLS library is too old for convenient support.
272 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
273 openssl_options), for security. OpenSSL forces this from version 1.1.0
274 server-side so match that on older versions.
276 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
277 allocation for $value could be released as the expansion processing
278 concluded, but leaving the global pointer active for it.
280 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
281 and to use the domains and local_parts ACL conditions.
283 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
284 incorrectly not doubled on cutthrough transmission, hence seen as a
285 body-termination at the receiving system - resulting in truncated mails.
286 Commonly the sender saw a TCP-level error, and retransmitted the message
287 via the normal store-and-forward channel. This could result in duplicates
288 received - but deduplicating mailstores were liable to retain only the
289 initial truncated version.
291 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
293 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
295 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
296 we're in there, support oversigning also; bug 1309.
298 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
300 HS/04 Add support for keep_environment and add_environment options.
302 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
303 either intentional arithmetic overflow during PRNG, or testing config-
306 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
307 delivery resulted in actual delivery. Cancel cutthrough before DATA
310 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
311 rejects a recipient: pass the reject to the originator.
313 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
314 Many were false-positives and ignorable, but it's worth fixing the
317 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
318 for the new environment-manipulation done at startup. Move the routines
319 from being local to tls.c to being global via the os.c file.
321 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
322 an extract embedded as result-arg for a map, the first arg for extract
323 is unavailable so we cannot tell if this is a numbered or keyed
324 extraction. Accept either.
330 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
333 JH/02 The smtp transport option "multi_domain" is now expanded.
335 JH/03 The smtp transport now requests PRDR by default, if the server offers
338 JH/04 Certificate name checking on server certificates, when exim is a client,
339 is now done by default. The transport option tls_verify_cert_hostnames
340 can be used to disable this per-host. The build option
341 EXPERIMENTAL_CERTNAMES is withdrawn.
343 JH/05 The value of the tls_verify_certificates smtp transport and main options
344 default to the word "system" to access the system default CA bundle.
345 For GnuTLS, only version 3.0.20 or later.
347 JH/06 Verification of the server certificate for a TLS connection is now tried
348 (but not required) by default. The verification status is now logged by
349 default, for both outbound TLS and client-certificate supplying inbound
352 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
355 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
356 Status Notification (bounce) messages are now MIME format per RFC 3464.
357 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
358 under the control of the dsn_advertise_hosts option, and routers may
359 have a dsn_lasthop option.
361 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
362 default, modifiable by a malware= option. The list separator for
363 the options can now be changed in the usual way. Bug 68.
365 JH/10 The smtp_receive_timeout main option is now expanded before use.
367 JH/11 The incoming_interface log option now also enables logging of the
368 local interface on delivery outgoing connections.
370 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
371 if the interface and destination host and port all match.
373 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
376 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
377 Patch from Andrew Lewis.
379 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
380 now supports optional time-restrictions, weighting, and priority
381 modifiers per server. Patch originally by <rommer@active.by>.
383 JH/16 The spamd_address main option now supports a mixed list of local
384 and remote servers. Remote servers can be IPv6 addresses, and
385 specify a port-range.
387 JH/17 Bug 68: The spamd_address main option now supports an optional
388 timeout value per server.
390 JH/18 Bug 1581: Router and transport options headers_add/remove can
391 now have the list separator specified.
393 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
396 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
399 JH/21 Support for the A6 type of dns record is withdrawn.
401 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
402 rather than the verbs used.
404 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
405 from 255 to 1024 chars.
407 JH/24 Verification callouts now attempt to use TLS by default.
409 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
410 are generic router options now. The defaults didn't change.
412 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
413 Original patch from Alexander Shikoff, worked over by JH.
415 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
416 files if ZCAT_COMMAND is not executable.
418 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
420 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
422 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
423 Normally benign, it bites when the pair was led to by a CNAME;
424 modern usage is to not canoicalize the domain to a CNAME target
425 (and we were inconsistent anyway for A-only vs AAAA+A).
427 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
429 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
430 when evaluating $sender_host_dnssec.
432 JH/31 Check the HELO verification lookup for DNSSEC, adding new
433 $sender_helo_dnssec variable.
435 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
437 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
439 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
441 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
442 documented as working, but never had. Support all but $spam_report.
444 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
445 added for tls authenticator.
447 HS/03 Add perl_taintmode main config option
452 TL/01 When running the test suite, the README says that variables such as
453 no_msglog_check are global and can be placed anywhere in a specific
454 test's script, however it was observed that placement needed to be near
455 the beginning for it to behave that way. Changed the runtest perl
456 script to read through the entire script once to detect and set these
457 variables, reset to the beginning of the script, and then run through
458 the script parsing/test process like normal.
460 TL/02 The BSD's have an arc4random API. One of the functions to induce
461 adding randomness was arc4random_stir(), but it has been removed in
462 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
463 function when detected.
465 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
466 cause callback expansion.
468 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
469 syntax errors in an expansion can be treated as a string instead of
470 logging or causing an error, due to the internal use of bool_lax
471 instead of bool when processing it.
473 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
474 server certificates when making smtp deliveries.
476 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
478 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
480 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
482 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
483 Merged patch from Sebastian Wiedenroth.
485 JH/05 Fix results-pipe from transport process. Several recipients, combined
486 with certificate use, exposed issues where response data items split
487 over buffer boundaries were not parsed properly. This eventually
488 resulted in duplicates being sent. This issue only became common enough
489 to notice due to the introduction of conection certificate information,
490 the item size being so much larger. Found and fixed by Wolfgang Breyha.
492 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
493 size buffer was used, resulting in syntax errors when an expansion
496 JH/07 Add support for directories of certificates when compiled with a GnuTLS
497 version 3.3.6 or later.
499 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
500 is EXPERIMENTAL_EVENT, the main-configuration and transport options
501 both become "event_action", the variables become $event_name, $event_data
502 and $event_defer_errno. There is a new variable $verify_mode, usable in
503 routers, transports and related events. The tls:cert event is now also
504 raised for inbound connections, if the main configuration event_action
507 TL/06 In test suite, disable OCSP for old versions of openssl which contained
508 early OCSP support, but no stapling (appears to be less than 1.0.0).
510 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
511 server certificate names available under the smtp transport option
512 "tls_verify_cert_hostname" now do not permit multi-component wildcard
515 JH/10 Time-related extraction expansions from certificates now use the main
516 option "timezone" setting for output formatting, and are consistent
517 between OpenSSL and GnuTLS compilations. Bug 1541.
519 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
520 encoded parameter in the incoming message. Bug 1558.
522 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
523 include certificate info, eximon was claiming there were spoolfile
526 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
528 JH/14 Log delivery-related information more consistently, using the sequence
529 "H=<name> [<ip>]" wherever possible.
531 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
532 are problematic for Debian distribution, omit them from the release
535 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
537 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
540 JH/17 Fix a null-indirection in certextract expansions when a nondefault
541 output list separator was used.
546 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
547 checkers that were complaining about end of non-void function with no
550 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
551 This was a regression intruduced in 4.83 by another bugfix.
553 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
555 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
556 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
562 TF/01 Correctly close the server side of TLS when forking for delivery.
564 When a message was received over SMTP with TLS, Exim failed to clear up
565 the incoming connection properly after forking off the child process to
566 deliver the message. In some situations the subsequent outgoing
567 delivery connection happened to have the same fd number as the incoming
568 connection previously had. Exim would try to use TLS and fail, logging
569 a "Bad file descriptor" error.
571 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
572 utilities have not been installed.
574 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
575 temporary space as the ACL may create new global variables.
577 TL/01 LDAP support uses per connection or global context settings, depending
578 upon the detected version of the libraries at build time.
580 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
581 to extract and use the src ip:port in logging and expansions as if it
582 were a direct connection from the outside internet. PPv2 support was
583 updated based on HAProxy spec change in May 2014.
585 JH/02 Add ${listextract {number}{list}{success}{fail}}.
587 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
588 Properly escape header and check for NULL return.
590 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
593 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
595 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
596 characters in header names, implemented as a verify condition.
597 Contributed by Michael Fischer v. Mollard.
599 TL/05 Rename SPF condition results err_perm and err_temp to standardized
600 results permerror and temperror. Previous values are deprecated but
601 still accepted. In a future release, err_perm and err_temp will be
602 completely removed, which will be a backward incompatibility if the
603 ACL tests for either of these two old results. Patch contributed by
604 user bes-internal on the mailing list.
606 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
608 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
609 selectors, in both main and reject logs.
611 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
614 JH/07 Add malware type "sock" for talking to simple daemon.
616 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
618 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
619 routers/transports under cutthrough routing.
621 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
622 numbers. Touch up "bool" conditional to keep the same definition.
624 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
626 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
628 JH/12 Expand items in router/transport headers_add or headers_remove lists
629 individually rather than the list as a whole. Bug 1452.
631 Required for reasonable handling of multiple headers_ options when
632 they may be empty; requires that headers_remove items with embedded
633 colons must have them doubled (or the list-separator changed).
635 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
636 view the policy declared in the DMARC record. Currently, $dmarc_status
637 is a combined value of both the record presence and the result of the
640 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
642 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
643 dnslookup router and the smtp transport (applying to the forward
646 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
647 of ldap servers used for a specific lookup. Patch provided by Heiko
650 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
651 New variable $lookup_dnssec_authenticated for observability.
653 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
654 Patch submitted by Lars Timman.
656 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
658 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
659 Requires trusted mode and valid format message id, aborts otherwise.
660 Patch contributed by Heiko Schlichting.
662 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
663 certextract with support for various fields. Bug 1358.
665 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
666 is requested by default, modifiable by smtp transport option
669 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
670 operate on certificate variables to give certificate fingerprints
671 Also new ${sha256:cert_variable}.
673 JH/23 The PRDR feature is moved from being Experimental into the mainline.
675 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
676 Christian Aistleitner.
678 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
680 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
681 file. Patch from Wolfgang Breyha.
683 JH/25 Expand the coverage of the delivery $host and $host_address to
684 client authenticators run in verify callout. Bug 1476.
686 JH/26 Port service names are now accepted for tls_on_connect_ports, to
687 align with daemon_smtp_ports. Bug 72.
689 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
690 support and error reporting did not work properly.
692 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
693 and is readable. Patch from Andrew Colin Kissa.
695 TL/14 Enhance documentation of ${run expansion and how it parses the
696 commandline after expansion, particularly in the case when an
697 unquoted variable expansion results in an empty value.
699 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
701 PP/02 Fix internal collision of T_APL on systems which support RFC3123
702 by renaming away from it. Addresses GH issue 15, reported by
705 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
707 TL/15 SECURITY: prevent double expansion in math comparison functions
708 (can expand unsanitized data). Not remotely exploitable.
715 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
717 PP/02 Make -n do something, by making it not do something.
718 When combined with -bP, the name of an option is not output.
720 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
723 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
724 $sender_host_name and config options to manage this, and basic check
727 PP/05 DSCP support for outbound connections and control modifier for inbound.
729 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
730 (Only plugin which currently uses this is kerberos4, which nobody should
731 be using, but we should make it available and other future plugins might
732 conceivably use it, even though it would break NAT; stuff *should* be
733 using channel bindings instead).
735 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
736 name; added for Sendmail compatibility; requires admin caller.
737 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
738 just ignore it); requires trusted caller.
739 Also parse but ignore: -Ac -Am -X<logfile>
742 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
744 TL/02 Add +smtp_confirmation as a default logging option.
746 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
747 Patch by Magnus Holmgren from 2007-02-20.
749 TL/04 Bugzilla 1281 - Spec typo.
750 Bugzilla 1283 - Spec typo.
751 Bugzilla 1290 - Spec grammar fixes.
753 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
755 TL/06 Add Experimental DMARC support using libopendmarc libraries.
757 TL/07 Fix an out of order global option causing a segfault. Reported to dev
758 mailing list by by Dmitry Isaikin.
760 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
762 JH/02 Support "G" suffix to numbers in ${if comparisons.
764 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
766 NM/01 Bugzilla 1197 - Spec typo
767 Bugzilla 1196 - Spec examples corrections
769 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
771 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
772 gnutls_enable_pkcs11, but renamed to more accurately indicate its
775 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
776 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
778 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
779 "acl {{name}{arg}...}", and optional args on acl condition
782 JH/05 Permit multiple router/transport headers_add/remove lines.
784 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
786 JH/07 Avoid using a waiting database for a single-message-only transport.
787 Performance patch from Paul Fisher. Bugzilla 1262.
789 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
792 JH/09 Add $headers_added variable, with content from use of ACL modifier
793 add_header (but not yet added to the message). Bugzilla 199.
795 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
796 Pulled from Bugzilla 817 by Wolfgang Breyha.
798 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
800 (nb: this is the same fix as in Exim 4.80.1)
802 JH/11 Add A= logging on delivery lines, and a client_set_id option on
805 JH/12 Add optional authenticated_sender logging to A= and a log_selector
808 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
810 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
811 advertise SMTP AUTH mechanism to us, instead of a generic
812 protocol violation error. Also, make Exim more robust to bad
813 data from the Dovecot auth socket.
815 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
817 When a queue runner is handling a message, Exim first routes the
818 recipient addresses, during which it prunes them based on the retry
819 hints database. After that it attempts to deliver the message to
820 any remaining recipients. It then updates the hints database using
823 So if a recipient address works intermittently, it can get repeatedly
824 deferred at routing time. The retry hints record remains fresh so the
825 address never reaches the final cutoff time.
827 This is a fairly common occurrence when a user is bumping up against
828 their storage quota. Exim had some logic in its local delivery code
829 to deal with this. However it did not apply to per-recipient defers
830 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
832 This change adds a proper retry rule check during routing so that the
833 final cutoff time is checked against the message's age. We only do
834 this check if there is an address retry record and there is not a
835 domain retry record; this implies that previous attempts to handle
836 the address had the retry_use_local_parts option turned on. We use
837 this as an approximation for the destination being like a local
838 delivery, as in LMTP.
840 I suspect this new check makes the old local delivery cutoff check
841 redundant, but I have not verified this so I left the code in place.
843 TF/02 Correct gecos expansion when From: is a prefix of the username.
845 Test 0254 submits a message to Exim with the header
849 When I ran the test suite under the user fanf2, Exim expanded
850 the header to contain my full name, whereas it should have added
851 a Resent-Sender: header. It erroneously treats any prefix of the
852 username as equal to the username.
854 This change corrects that bug.
856 GF/01 DCC debug and logging tidyup
857 Error conditions log to paniclog rather than rejectlog.
858 Debug lines prefixed by "DCC: " to remove any ambiguity.
860 TF/03 Avoid unnecessary rebuilds of lookup-related code.
862 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
863 Bug spotted by Jeremy Harris; was flawed since initial commit.
864 Would have resulted in OCSP responses post-SNI triggering an Exim
865 NULL dereference and crash.
867 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
869 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
870 Bug detection, analysis and fix by Samuel Thibault.
871 Bugzilla 1331, Debian bug #698092.
873 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
875 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
876 Server implementation by Todd Lyons, client by JH.
877 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
878 config variable "prdr_enable" controls whether the server
879 advertises the facility. If the client requests PRDR a new
880 acl_data_smtp_prdr ACL is called once for each recipient, after
881 the body content is received and before the acl_smtp_data ACL.
882 The client is controlled by bolth of: a hosts_try_prdr option
883 on the smtp transport, and the server advertisement.
884 Default client logging of deliveries and rejections involving
885 PRDR are flagged with the string "PRDR".
887 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
888 fclose(). Diagnosis by Todd Lyons.
890 PP/17 Update configure.default to handle IPv6 localhost better.
891 Patch by Alain Williams (plus minor tweaks).
894 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
895 This is now consistent with GnuTLS, and is now documented: the
896 previous undocumented portable approach to treating the option as
897 unset was to force an expansion failure. That still works, and
898 an empty string is now equivalent.
900 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
901 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
902 not performing validation itself.
904 PP/20 Added force_command boolean option to pipe transport.
905 Patch from Nick Koston, of cPanel Inc.
907 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
910 TF/04 Added udpsend ACL modifer and hexquote expansion operator
912 PP/21 Fix eximon continuous updating with timestamped log-files.
913 Broken in a format-string cleanup in 4.80, missed when I repaired the
914 other false fix of the same issue.
915 Report and fix from Heiko Schlichting.
918 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
919 Report from Prashanth Katuri.
921 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
922 It's SecureTransport, so affects any MacOS clients which use the
923 system-integrated TLS libraries, including email clients.
925 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
926 using a MIME ACL for non-SMTP local injection.
927 Report and assistance in diagnosis by Warren Baker.
929 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
931 JH/16 Fix comparisons for 64b. Bugzilla 1385.
933 TL/09 Add expansion variable $authenticated_fail_id to keep track of
934 last id that failed so it may be referenced in subsequent ACL's.
936 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
939 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
940 ldap library initialization, allowing self-signed CA's to be
941 used. Also properly sets require_cert option later in code by
942 using NULL (global ldap config) instead of ldap handle (per
943 session). Bug diagnosis and testing by alxgomz.
945 TL/12 Enhanced documentation in the ratelimit.pl script provided in
946 the src/util/ subdirectory.
948 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
949 renamed to Transport Post Delivery Action by Jeremy Harris, as
952 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
953 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
954 redis_servers = needs to be configured which will be used by the redis
955 lookup. Patch from Warren Baker, of The Packet Hub.
957 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
959 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
960 hostname or reverse DNS when processing a host list. Used suggestions
961 from multiple comments on this bug.
963 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
965 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
966 Missed a few lines, added it to make the runtest require no keyboard
969 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
970 contains upper case chars. Make router use caseful_local_part.
972 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
973 support when GnuTLS has been built with p11-kit.
979 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
981 This, or similar/improved, will also be change PP/11 of 4.82.
987 PP/01 Handle short writes when writing local log-files.
988 In practice, only affects FreeBSD (8 onwards).
989 Bugzilla 1053, with thanks to Dmitry Isaikin.
991 NM/01 Bugzilla 949 - Documentation tweak
993 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
996 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
998 PP/02 Implemented gsasl authenticator.
1000 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1002 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1003 `pkg-config foo` for cflags/libs.
1005 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1006 with rest of GSASL and with heimdal_gssapi.
1008 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1009 `pkg-config foo` for cflags/libs for the TLS implementation.
1011 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1012 properties get this fed in as external SSF. A number of robustness
1013 and debugging improvements to the cyrus_sasl authenticator.
1015 PP/08 cyrus_sasl server now expands the server_realm option.
1017 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1018 Patch by Jeremy Harris.
1020 PP/10 Added dbmjz lookup type.
1022 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1024 PP/12 MAIL args handles TAB as well as SP, for better interop with
1025 non-compliant senders.
1026 Analysis and variant patch by Todd Lyons.
1028 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1029 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1030 Patch from Dirk Mueller <dmueller@suse.com>
1032 PP/13 tls_peerdn now print-escaped for spool files.
1033 Observed some $tls_peerdn in wild which contained \n, which resulted
1034 in spool file corruption.
1036 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1037 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1038 or write after TLS renegotiation, which otherwise led to messages
1041 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1042 as a tracking header (ie: a signed header comes before the signature).
1043 Patch from Wolfgang Breyha.
1045 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1046 comma-sep list; embedded commas doubled.
1048 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1050 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1052 Report and patch from Dmitry Banschikov.
1054 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1055 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1056 failures. We appear to now support TLS1.1+ with Exim.
1058 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1059 lets Exim select keys and certificates based upon TLS SNI from client.
1060 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1061 before an outbound SMTP session. New log_selector, +tls_sni.
1063 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1064 NULL dereference. Report and patch from Alun Jones.
1066 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1067 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1068 Not seeing resolver debug output on NetBSD, but suspect this is a
1069 resolver implementation change.
1071 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1072 Left warnings. Added "eximon gdb" invocation mode.
1074 PP/21 Defaulting "accept_8bitmime" to true, not false.
1076 PP/22 Added -bw for inetd wait mode support.
1078 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1079 locate the relevant includes and libraries. Made this the default.
1081 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1082 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1084 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1085 This may cause build issues on older platforms.
1087 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1088 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1089 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1090 Added SNI support via GnuTLS too.
1091 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1093 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1095 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1096 Applied second patch from Janne, implementing suggestion to default
1097 multiple-strings-in-record handling to match SPF spec.
1099 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1101 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1102 read-only, out of scope).
1103 Patch from Wolfgang Breyha, report from Stuart Northfield.
1105 PP/29 Fix three issues highlighted by clang analyser static analysis.
1106 Only crash-plausible issue would require the Cambridge-specific
1107 iplookup router and a misconfiguration.
1108 Report from Marcin Mirosław.
1110 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1112 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1113 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1114 As part of this, removing so much warning spew let me fix some minor
1115 real issues in debug logging.
1117 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1118 assignment on my part. Fixed.
1120 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1121 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1122 Janne Snabb (who went above and beyond: thank you).
1124 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1125 string otherwise requires a connection and a bunch more work and it's
1126 relatively easy to get wrong. Should also expose TLS library linkage
1129 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1130 64-bit ${eval} (JH/03).
1132 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1133 GNU libc to support some of the 64-bit stuff, should not lead to
1134 conflicts. Defined before os.h is pulled in, so if a given platform
1135 needs to override this, it can.
1137 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1138 protection layer was required, which is not implemented.
1139 Bugzilla 1254, patch from Wolfgang Breyha.
1141 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1142 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1143 tls_dhparam take prime identifiers. Also unbreak combination of
1144 OpenSSL+DH_params+TLSSNI.
1146 PP/39 Disable SSLv2 by default in OpenSSL support.
1152 PP/01 Solaris build fix for Oracle's LDAP libraries.
1153 Bugzilla 1109, patch from Stephen Usher.
1155 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1157 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1160 TF/02 Fix a couple more cases where we did not log the error message
1161 when unlink() failed. See also change 4.74-TF/03.
1163 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1164 lock up or crash if it happened to be inside a call to libc when it
1165 got a SIGUSR1 from exiwhat.
1167 The SIGUSR1 handler appends the current process status to the process
1168 log which is later printed by exiwhat. It used to use the general
1169 purpose logging code to do this, but several functions it calls are
1170 not safe for signals.
1172 The new output code in the SIGUSR1 handler is specific to the process
1173 log, and simple enough that it's easy to inspect for signal safety.
1174 Removing some special cases also simplifies the general logging code.
1175 Removing the spurious timestamps from the process log simplifies
1178 TF/04 Improved ratelimit ACL condition.
1180 The /noupdate option has been deprecated in favour of /readonly which
1181 has clearer semantics. The /leaky, /strict, and /readonly update modes
1182 are mutually exclusive. The update mode is no longer included in the
1183 database key; it just determines when the database is updated. (This
1184 means that when you upgrde Exim will forget old rate measurements.)
1186 Exim now checks that the per_* options are used with an update mode that
1187 makes sense for the current ACL. For example, when Exim is processing a
1188 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1189 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1190 must specify per_mail/readonly. If you omit the update mode it defaults to
1191 /leaky where that makes sense (as before) or /readonly where required.
1193 The /noupdate option is now undocumented but still supported for
1194 backwards compatibility. It is equivalent to /readonly except that in
1195 ACLs where /readonly is required you may specify /leaky/noupdate or
1196 /strict/noupdate which are treated the same as /readonly.
1198 A useful new feature is the /count= option. This is a generalization
1199 of the per_byte option, so that you can measure the throughput of other
1200 aggregate values. For example, the per_byte option is now equivalent
1201 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1203 The per_rcpt option has been generalized using the /count= mechanism
1204 (though it's more complicated than the per_byte equivalence). When it is
1205 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1206 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1207 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1208 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1209 Note that using per_rcpt with a non-readonly update mode in more than
1210 one ACL will cause the recipients to be double-counted. (The per_mail
1211 and per_byte options don't have this problem.)
1213 The handling of very low rates has changed slightly. If the computed rate
1214 is less than the event's count (usually one) then this event is the first
1215 after a long gap. In this case the rate is set to the same as this event's
1216 count, so that the first message of a spam run is counted properly.
1218 The major new feature is a mechanism for counting the rate of unique
1219 events. The new per_addr option counts the number of different
1220 recipients that someone has sent messages to in the last time period. It
1221 behaves like per_rcpt if all the recipient addresses are different, but
1222 duplicate recipient addresses do not increase the measured rate. Like
1223 the /count= option this is a general mechanism, so the per_addr option
1224 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1225 example, measure the rate that a client uses different sender addresses
1226 with the options per_mail/unique=$sender_address. There are further
1227 details in the main documentation.
1229 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1231 TF/06 Removed a few PCRE remnants.
1233 TF/07 Automatically extract Exim's version number from tags in the git
1234 repository when doing development or release builds.
1236 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1237 Bugzilla 879. Patch from Paul Fisher.
1239 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1240 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1243 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1245 PP/05 Variable $av_failed, true if the AV scanner deferred.
1246 Bugzilla 1078. Patch from John Horne.
1248 PP/06 Stop make process more reliably on build failure.
1249 Bugzilla 1087. Patch from Heiko Schlittermann.
1251 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1252 Bugzilla 1089. Patch from Heiko Schlittermann.
1254 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1255 Bugzilla 1131. Patch from Holger Weiß.
1257 PP/09 Handle IPv6 addresses with SPF.
1258 Bugzilla 860. Patch from Wolfgang Breyha.
1260 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1262 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1265 PP/11 match_* no longer expand right-hand-side by default.
1266 New compile-time build option, EXPAND_LISTMATCH_RHS.
1267 New expansion conditions, "inlist", "inlisti".
1269 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1271 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1273 PP/14 fix log_write() format string regression from TF/03.
1274 Bugzilla 1152. Patch from Dmitry Isaikin.
1280 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1282 PP/02 Harmonised TLS library version reporting; only show if debugging.
1283 Layout now matches that introduced for other libraries in 4.74 PP/03.
1285 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1287 PP/04 New "dns_use_edns0" global option.
1289 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1292 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1293 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1295 TK/01 Updated PolarSSL code to 0.14.2.
1296 Bugzilla 1097. Patch from Andreas Metzler.
1298 PP/07 Catch divide-by-zero in ${eval:...}.
1299 Fixes bugzilla 1102.
1301 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1304 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1305 format-string attack -- SECURITY: remote arbitrary code execution.
1307 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1308 time unintentionally subject to list matching rules, letting the header
1309 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1310 arbitrary string expansion). This allowed for information disclosure.
1312 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1313 INT_MIN/-1 -- value coerced to INT_MAX.
1319 NM/01 Workround for PCRE version dependency in version reporting
1322 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1323 This fixes portability to compilers other than gcc, notably
1324 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1326 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1327 makefiles for portability to HP-UX and POSIX correctness.
1329 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1330 Also via indented variable definition in the Makefile.
1331 (Debugging by Oliver Heesakkers).
1333 PP/02 Restore caching of spamd results with expanded spamd_address.
1334 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1336 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1337 Improves build reliability. Fix from: Frank Elsner
1339 NM/02 Fix wide character breakage in the rfc2047 coding
1340 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1342 NM/03 Allow underscore in dnslist lookups
1343 Fixes bug 1026. Patch from Graeme Fowler
1345 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1346 Code patches from Adam Ciarcinski of NetBSD.
1348 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1351 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1352 is logged, to avoid truncation. Patch from John Horne.
1354 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1355 Patch from Jakob Hirsch.
1357 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1358 SQL string expansion failure details.
1359 Patch from Andrey Oktyabrski.
1361 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1362 Patch from Simon Arlott.
1364 PP/09 New lookups functionality failed to compile on old gcc which rejects
1365 extern declarations in function scope.
1366 Patch from Oliver Fleischmann
1368 PP/10 Use sig_atomic_t for flags set from signal handlers.
1369 Check getgroups() return and improve debugging.
1370 Fixed developed for diagnosis in bug 927 (which turned out to be
1373 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1374 Patch from Mark Zealey.
1376 PP/12 Bugzilla 1056: Improved spamd server selection.
1377 Patch from Mark Zealey.
1379 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1380 Based on patch from Heiko Schlittermann.
1382 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1383 Patch from Uwe Doering, sign-off by Michael Haardt.
1385 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1386 variable declaration deep within a block. Bug and patch from
1389 PP/15 lookups-Makefile IRIX compatibilty coercion.
1391 PP/16 Make DISABLE_DKIM build knob functional.
1393 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1394 Patch by Simon Arlott
1396 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1397 variable argument macros. Our copy now differs from upstream.
1403 TF/01 Failure to get a lock on a hints database can have serious
1404 consequences so log it to the panic log.
1406 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1407 controlled using the smtp_confirmation log selector.
1409 TF/03 Include the error message when we fail to unlink a spool file.
1411 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1412 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1413 for maintaining out-of-tree patches for some time.
1415 PP/01 Bugzilla 139: Documentation and portability issues.
1416 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1417 Handle per-OS dynamic-module compilation flags.
1419 PP/02 Let /dev/null have normal permissions.
1420 The 4.73 fixes were a little too stringent and complained about the
1421 permissions on /dev/null. Exempt it from some checks.
1422 Reported by Andreas M. Kirchwitz.
1424 PP/03 Report version information for many libraries, including
1425 Exim version information for dynamically loaded libraries. Created
1426 version.h, now support a version extension string for distributors
1427 who patch heavily. Dynamic module ABI change.
1429 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1430 privilege escalation vulnerability whereby the Exim run-time user
1431 can cause root to append content of the attacker's choosing to
1434 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1437 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1438 If dropping privileges for untrusted macros, we disabled normal logging
1439 on the basis that it would fail; for the Exim run-time user, this is not
1440 the case, and it resulted in successful deliveries going unlogged.
1441 Fixed. Reported by Andreas Metzler.
1447 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1448 only prepend for the Resent-* case. Fixes regression introduced in
1449 Exim 4.70 by NM/22 for Bugzilla 607.
1451 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1452 increasing numbers of administrators be bitten by this.
1454 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1456 PP/03 Bugzilla 994: added openssl_options main configuration option.
1458 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1460 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1462 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1464 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1465 without a peer certificate, leading to a segfault because of an
1466 assumption that peers always have certificates. Be a little more
1467 paranoid. Problem reported by Martin Tscholak.
1469 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1470 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1471 NB: ClamAV planning to remove STREAM in "middle of 2010".
1472 CL also introduces -bmalware, various -d+acl logging additions and
1473 more caution in buffer sizes.
1475 PP/09 Implemented reverse_ip expansion operator.
1477 PP/10 Bugzilla 937: provide a "debug" ACL control.
1479 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1481 PP/12 Bugzilla 973: Implement --version.
1483 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1485 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1487 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1489 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1490 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1491 ignore trailing whitespace.
1493 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1495 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1498 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1499 Notification from Dr Andrew Aitchison.
1501 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1502 ExtendedDetectionInfo response format.
1503 Notification from John Horne.
1505 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1508 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1509 XSL and documented dependency on system catalogs, with examples of how
1512 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1515 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1516 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1517 configuration file which is writeable by the Exim user or group.
1519 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1520 of configuration files to cover files specified with the -C option if
1521 they are going to be used with root privileges, not just the default
1524 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1525 option (effectively making it always true).
1527 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1528 files to be used while preserving root privileges.
1530 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1531 that rogue child processes cannot use them.
1533 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1534 run-time user, instead of root.
1536 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1537 Exim run-time user without dropping privileges.
1539 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1540 result string, instead of calling string_vformat() twice with the same
1543 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1544 for other users. Others should always drop root privileges if they use
1545 -C on the command line, even for a whitelisted configure file.
1547 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1549 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1555 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1556 $data_path, and $header_path variables; fixed documentation bugs and
1559 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1560 exipick to access non-standard spools, including the "frozen" queue
1563 NM/01 Bugzilla 965: Support mysql stored procedures.
1564 Patch from Alain Williams
1566 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1568 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1569 Patch from Andreas Metzler
1571 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1572 Patch from Kirill Miazine
1574 NM/05 Bugzilla 671: Added umask to procmail example.
1576 JJ/03 installed exipick 20100323.0, fixing doc bug
1578 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1579 directory. Notification and patch from Dan Rosenberg.
1581 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1583 TK/02 Improve log output when DKIM signing operation fails.
1585 MH/01 Treat the transport option dkim_domain as a colon separated
1586 list, not as a single string, and sign the message with each element,
1587 omitting multiple occurences of the same signer.
1589 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1590 Bugzilla 985, 986. Patch by Simon Arlott
1592 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1593 Patch by Simon Arlott
1595 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1596 MBX locking. Notification from Dan Rosenberg.
1602 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1604 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1606 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1608 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1610 NM/04 Bugzilla 915: Flag broken perl installation during build.
1616 TK/01 Added patch by Johannes Berg that expands the main option
1617 "spamd_address" if it starts with a dollar sign.
1619 TK/02 Write list of recipients to X-Envelope-Sender header when building
1620 the mbox-format spool file for content scanning (suggested by Jakob
1623 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1624 (http://www.dcc-servers.net/) support via dccifd. Activated by
1625 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1627 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1628 by Mark Daniel Reidel <mr@df.eu>.
1630 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1631 When building exim an external PCRE library is now needed -
1632 PCRE is a system library on the majority of modern systems.
1633 See entry on PCRE_LIBS in EDITME file.
1635 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1636 conversation. Added nologin parameter to request.
1637 Patch contributed by Kirill Miazine.
1639 TF/01 Do not log submission mode rewrites if they do not change the address.
1641 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1643 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1644 log files in place. Contributed by Roberto Lima.
1646 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1648 TF/03 Bugzilla 615: When checking the local_parts router precondition
1649 after a local_part_suffix or local_part_prefix option, Exim now
1650 does not use the address's named list lookup cache, since this
1651 contains cached lookups for the whole local part.
1653 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1654 Robert Millan. Documentation is in experimental-spec.txt.
1656 TF/04 Bugzilla 668: Fix parallel build (make -j).
1658 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1660 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1661 Patch provided by Jan Srzednicki.
1663 TF/05 Leading white space used to be stripped from $spam_report which
1664 wrecked the formatting. Now it is preserved.
1666 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1667 that they are available at delivery time.
1669 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1671 TF/08 TLS error reporting now respects the incoming_interface and
1672 incoming_port log selectors.
1674 TF/09 Produce a more useful error message if an SMTP transport's hosts
1675 setting expands to an empty string.
1677 NM/06 Bugzilla 744: EXPN did not work under TLS.
1678 Patch provided by Phil Pennock.
1680 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1681 Patch provided by Richard Godbee.
1683 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1684 acl_smtp_notquit, added index entry.
1686 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1687 Patch provided by Eugene Bujak.
1689 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1690 accept(). Patch provided by Maxim Dounin.
1692 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1693 Patch provided by Phil Pennock.
1695 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1697 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1698 Patch provided by Brad "anomie" Jorsch.
1700 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1701 Patch provided by Dean Brooks.
1703 TK/05 Add native DKIM support (does not depend on external libraries).
1705 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1706 Patch provided by Graeme Fowler.
1708 NM/16 Bugzilla 851: Documentation example syntax fix.
1710 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1712 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1715 NM/19 Bugzilla 745: TLS version reporting.
1716 Patch provided by Phil Pennock.
1718 NM/20 Bugzilla 167: bool: condition support.
1719 Patch provided by Phil Pennock.
1721 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1722 clients. Patch provided by Phil Pennock.
1724 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1725 Patch provided by Brad "anomie" Jorsch.
1727 NM/23 Bugzilla 687: Fix misparses in eximstats.
1728 Patch provided by Heiko Schlittermann.
1730 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1731 Patch provided by Heiko Schlittermann.
1733 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1734 plus update to original patch.
1736 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1738 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1739 Patch provided by David Brownlee.
1741 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1743 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1745 NM/30 Bugzilla 888: TLS documentation bugfixes.
1747 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1749 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1750 Unlike the original bugzilla I have changed all shell scripts in src tree.
1752 NM/33 Bugzilla 898: Transport filter timeout fix.
1753 Patch by Todd Rinaldo.
1755 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1756 Patch by Serge Demonchaux.
1758 NM/35 Bugzilla 39: Base64 decode bug fixes.
1759 Patch by Jakob Hirsch.
1761 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1763 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1765 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1767 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1773 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1774 ALT-N's libdkim that I have put here:
1775 http://duncanthrax.net/exim-experimental/
1777 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1778 were called 'true' and it seems that C99 defines that as a reserved
1779 keyword to be used with 'bool' variable types. That means you could
1780 not include C99-style headers which use bools without triggering
1781 build errors in sieve.c.
1783 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1784 as mailq or other aliases. Changed the --help handling significantly
1785 to do whats expected. exim_usage() emits usage/help information.
1787 SC/01 Added the -bylocaldomain option to eximstats.
1789 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1791 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1793 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1799 PH/01 Another patch from the Sieve maintainer.
1801 PH/02 When an IPv6 address is converted to a string for single-key lookup
1802 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1803 dots are used instead of colons so that keys in lsearch files need not
1804 contain colons. This was done some time before quoting was made available
1805 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1806 (notated using the quote facility) so as to distinguish them from IPv4
1807 keys. This meant that lookups for IP addresses in host lists did not work
1808 for iplsearch lookups.
1810 This has been fixed by arranging for IPv6 addresses to be expressed with
1811 colons if the lookup type is iplsearch. This is not incompatible, because
1812 previously such lookups could never work.
1814 The situation is now rather anomolous, since one *can* have colons in
1815 ordinary lsearch keys. However, making the change in all cases is
1816 incompatible and would probably break a number of configurations.
1818 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1821 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1822 conversion specification without a maximum field width, thereby enabling
1823 a rogue spamd server to cause a buffer overflow. While nobody in their
1824 right mind would setup Exim to query an untrusted spamd server, an
1825 attacker that gains access to a server running spamd could potentially
1826 exploit this vulnerability to run arbitrary code as the Exim user.
1828 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1829 $primary_hostname instead of what libspf2 thinks the hosts name is.
1831 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1832 a directory entry by the name of the lookup key. Previously, if a
1833 symlink pointed to a non-existing file or a file in a directory that
1834 Exim lacked permissions to read, a lookup for a key matching that
1835 symlink would fail. Now it is enough that a matching directory entry
1836 exists, symlink or not. (Bugzilla 503.)
1838 PH/03 The body_linecount and body_zerocount variables are now exported in the
1841 PH/04 Added the $dnslist_matched variable.
1843 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1844 This means they are set thereafter only if the connection becomes
1847 PH/06 Added the client_condition to authenticators so that some can be skipped
1848 by clients under certain conditions.
1850 PH/07 The error message for a badly-placed control=no_multiline_responses left
1851 "_responses" off the end of the name.
1853 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1855 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1856 (without spaces) instead of just copying the configuration text.
1858 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1860 PH/11 Added $max_received_linelength.
1862 PH/12 Added +ignore_defer and +include_defer to host lists.
1864 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1865 way in which PCRE > 7.0 is built.
1867 PH/14 Implemented queue_only_load_latch.
1869 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1870 MAIL command. The effect was to mangle the value on 64-bit systems.
1872 PH/16 Another patch from the Sieve maintainer.
1874 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1876 PH/18 If a system quota error occurred while trying to create the file for
1877 a maildir delivery, the message "Mailbox is full" was not appended to the
1878 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1879 only to a quota excession during the actual writing of the file.
1881 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1882 characters?) which causes problems in log lines. The DN values are now
1883 passed through string_printing() before being added to log lines.
1885 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1886 and InterBase are left for another time.)
1888 PH/21 Added message_body_newlines option.
1890 PH/22 Guard against possible overflow in moan_check_errorcopy().
1892 PH/23 POSIX allows open() to be a macro; guard against that.
1894 PH/24 If the recipient of an error message contained an @ in the local part
1895 (suitably quoted, of course), incorrect values were put in $domain and
1896 $local_part during the evaluation of errors_copy.
1902 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1903 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1906 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1907 issue a MAIL command.
1909 PH/02 In an ACL statement such as
1911 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1913 if a client was not listed at all, or was listed with a value other than
1914 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1915 the condition was not true (as it should be), so access was not denied.
1916 The bug was that the ! inversion was incorrectly passed on to the second
1917 item. This has been fixed.
1919 PH/03 Added additional dnslists conditions == and =& which are different from
1920 = and & when the dns lookup returns more than one IP address.
1922 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1923 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1925 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1926 FSYNC, which compiles an option called disable_fsync that allows for
1927 bypassing fsync(). The documentation is heavily laced with warnings.
1929 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1931 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1932 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1933 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1934 including adding "make clean"; (3) Added -fPIC when compiling the test
1935 dynamically loaded module, to get rid of a warning.
1937 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1938 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1939 = 0s. The bug is otherwise harmless.
1941 PH/07 There was a bug in the dovecot authenticator such that the value of
1942 $auth1 could be overwritten, and so not correctly preserved, after a
1943 successful authentication. This usually meant that the value preserved by
1944 the server_setid option was incorrect.
1946 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1948 PH/09 Installed PCRE release 7.0.
1950 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1951 run for batched SMTP input. It is now run at the start of every message
1952 in the batch. While fixing this I discovered that the process information
1953 (output by running exiwhat) was not always getting set for -bs and -bS
1954 input. This is fixed, and it now also says "batched" for BSMTP.
1956 PH/11 Added control=no_pipelining.
1958 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1959 patch, slightly modified), and move the expansion of helo_data till after
1960 the connection is made in the smtp transport (so it can use these
1963 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1965 PH/14 Added log_selector = +pid.
1967 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1969 PH/16 Add ${if forany and ${if forall.
1971 PH/17 Added dsn_from option to vary the From: line in DSNs.
1973 PH/18 Flush SMTP output before performing a callout, unless control =
1974 no_callout_flush is set.
1976 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1977 was true (the default) a successful delivery failed to delete the retry
1978 item, thus causing premature timeout of the address. The bug is now
1981 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1983 PH/21 Long custom messages for fakedefer and fakereject are now split up
1984 into multiline reponses in the same way that messages for "deny" and
1985 other ACL rejections are.
1987 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1988 with slight modification.
1990 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1991 draft, changing the syntax and factoring some duplicate code".
1993 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1994 for deliveries of the second and subsequent messages over the same SMTP
1997 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1998 ${reduce, with only minor "tidies".
2000 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2002 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2003 expansion side effects.
2005 PH/27 When a message times out after an over-quota error from an Exim-imposed
2006 quota, the bounce message says "mailbox is full". This message was not
2007 being given when it was a system quota that was exceeded. It now should
2010 MH/03 Made $recipients available in local_scan(). local_scan() already has
2011 better access to the recipient list through recipients_list[], but
2012 $recipients can be useful in postmaster-provided expansion strings.
2014 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2015 in the case of a MAIL command with additional options following the
2016 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2017 were accidentally chopped off.
2019 PH/29 SMTP synchronization checks are implemented when a command is read -
2020 there is a check that no more input is waiting when there shouldn't be
2021 any. However, for some commands, a delay in an ACL can mean that it is
2022 some time before the response is written. In this time, more input might
2023 arrive, invalidly. So now there are extra checks after an ACL has run for
2024 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2025 pipelining has not been advertised.
2027 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2029 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2030 returned in the SMTP response when smtp_return_error_details was false.
2031 This has been fixed.
2033 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2034 instead of the C I/O that was originally supplied, because problems were
2035 reported on Solaris.
2037 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2038 Exim which did not show up earlier: it was assuming that a call to
2039 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2040 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2041 was a macro that became an assignment, so it seemed to work. This has
2042 changed to a proper function call with a void return, hence the compile
2043 error. Exim's code has been fixed.
2045 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2048 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2050 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2052 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2053 because it turns out to be too big or there is a timeout) there is no
2054 "Completed" line in the log. When some messages of this type were
2055 selected by exigrep, they were listed as "not completed". Others were
2056 picked up by some special patterns. I have improved the selection
2057 criteria to be more general.
2059 PH/38 The host_find_failed option in the manualroute router can now be set
2060 to "ignore", to completely ignore a host whose IP address cannot be
2061 found. If all hosts are ignored, the behaviour is controlled by the new
2062 host_all_ignored option.
2064 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2065 homing or because of multiple MX records with /mx) generated more than
2066 one IP address, and the following item turned out to be the local host,
2067 all the secondary addresses of the first item were incorrectly removed
2068 from the list, along with the local host and any following hosts (which
2069 is what is supposed to happen).
2071 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2072 whoever called Exim into the -H file. In the case of the daemon it was
2073 behaving confusingly. When first started, it used values for whoever
2074 started the daemon, but after a SIGHUP it used the Exim user (because it
2075 calls itself on a restart). I have changed the code so that it now always
2078 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2079 message are rejected with the same error (e.g. no authentication or bad
2080 sender address), and a DATA command is nevertheless sent (as can happen
2081 with PIPELINING or a stupid MUA), the error message that was given to the
2082 RCPT commands is included in the rejection of the DATA command. This is
2083 intended to be helpful for MUAs that show only the final error to their
2086 PH/42 Another patch from the Sieve maintainer.
2088 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2089 Eximstats - Fixed some broken HTML links and added missing column headers
2091 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2092 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2094 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2096 PH/43 Yet another patch from the Sieve maintainer.
2098 PH/44 I found a way to check for a TCP/IP connection going away before sending
2099 the response to the final '.' that terminates a message, but only in the
2100 case where the client has not sent further data following the '.'
2101 (unfortunately, this is allowed). However, in many cases there won't be
2102 any further data because there won't be any more messages to send. A call
2103 to select() can be used: if it shows that the input is "ready", there is
2104 either input waiting, or the socket has been closed. An attempt to read
2105 the next input character can distinguish the two cases. Previously, Exim
2106 would have sent an OK response which the client would never have see.
2107 This could lead to message repetition. This fix should cure that, at
2108 least in a lot of common cases.
2110 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2111 advertised in response to EHLO.
2117 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2118 fixed by 4.65/MH/01 (is this a record?) are fixed:
2120 (i) An empty string was always treated as zero by the numeric comparison
2121 operators. This behaviour has been restored.
2123 (ii) It is documented that the numeric comparison operators always treat
2124 their arguments as decimal numbers. This was broken in that numbers
2125 starting with 0 were being interpreted as octal.
2127 While fixing these problems I realized that there was another issue that
2128 hadn't been noticed. Values of message_size_limit (both the global option
2129 and the transport option) were treated as octal if they started with 0.
2130 The documentation was vague. These values are now always treated as
2131 decimal, and I will make that clear in the documentation.
2137 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2138 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2141 MH/01 Don't check that the operands of numeric comparison operators are
2142 integers when their expansion is in "skipping" mode (fixes bug
2143 introduced by 4.64-PH/07).
2145 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2146 child addresses, Exim now panics and dies. Previously, because the count
2147 is held in a short int, deliveries were likely to be lost. As such a
2148 large number of recipients for a single message is ridiculous
2149 (performance will be very, very poor), I have chosen to impose a limit
2150 rather than extend the field.
2156 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2157 leftover -K file (the existence of which was triggered by #402).
2158 While we were at it, introduced process PID as part of the -K
2159 filename. This should rule out race conditions when creating
2162 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2163 processing considerably. Previous code took too long for large mails,
2164 triggering a timeout which in turn triggers #401.
2166 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2167 in the DK code in transports.c. sendfile() is not really portable,
2168 hence the _LINUX specificness.
2170 TF/01 In the add_headers option to the mail command in an Exim filter,
2171 there was a bug that Exim would claim a syntax error in any
2172 header after the first one which had an odd number of characters
2175 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2176 callout verification, Exim cached a "reject" for the entire domain. This
2177 is correct for most verifications, but it is not correct for a recipient
2178 verification with use_sender or use_postmaster set, because in that case
2179 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2180 case of MAIL FROM:<> rejection from other early rejections (e.g.
2181 rejection of HELO). When verifying a recipient using a non-null MAIL
2182 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2183 Whatever the result of the callout, the value of the domain cache is
2184 left unchanged (for any other kind of callout, getting as far as trying
2185 RCPT means that the domain itself is ok).
2187 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2190 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2191 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2192 session, the daemon ignores ECONNECT errors and logs others; it now
2193 ignores EPIPE as well.
2195 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2196 (quoted-printable decoding).
2198 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2199 later the small subsequent patch to fix an introduced bug.
2201 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2203 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2205 PH/08 An error is now given if message_size_limit is specified negative.
2207 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2208 to be given (somewhat) arbitrary names.
2210 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2213 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2214 miscellaneous code fixes
2216 PH/10 Added the log_reject_target ACL modifier to specify where to log
2219 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2220 hostname. This is wrong, because it relates to the incoming message (and
2221 probably the interface on which it is arriving) and not to the outgoing
2222 callout (which could be using a different interface). This has been
2223 changed to use the value of the helo_data option from the smtp transport
2224 instead - this is what is used when a message is actually being sent. If
2225 there is no remote transport (possible with a router that sets up host
2226 addresses), $smtp_active_hostname is used.
2228 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2229 tweaks were necessary in order to get it to work (see also 21 below):
2230 (a) The code assumed that strncpy() returns a negative number on buffer
2231 overflow, which isn't the case. Replaced with Exim's string_format()
2233 (b) There were several signed/unsigned issues. I just did the minimum
2234 hacking in of casts. There is scope for a larger refactoring.
2235 (c) The code used strcasecmp() which is not a standard C function.
2236 Replaced with Exim's strcmpic() function.
2237 (d) The code set only $1; it now sets $auth1 as well.
2238 (e) A simple test gave the error "authentication client didn't specify
2239 service in request". It would seem that Dovecot has changed its
2240 interface. Fortunately there's a specification; I followed it and
2241 changed what the client sends and it appears to be working now.
2243 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2246 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2247 address A is aliased to B and C, where B exists and C does not. Without
2248 -v the output is "A verified" because verification stops after a
2249 successful redirection if more than one address is generated. However,
2250 with -v the child addresses are also verified. Exim was outputting "A
2251 failed to verify" and then showing the successful verification for C,
2252 with its parentage. It now outputs "B failed to verify", showing B's
2253 parentage before showing the successful verification of C.
2255 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2256 look up a TXT record in a specific list after matching in a combined
2259 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2260 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2261 they consult the DNS. I had assumed they would set it the way they
2262 wanted; and indeed my experiments on Linux seem to show that in some
2263 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2264 To be on the safe side, however, I have now made the interface to
2265 host_find_byname() similar to host_find_bydns(), with an argument
2266 containing the DNS resolver options. The host_find_byname() function now
2267 sets these options at its start, just as host_find_bydns() does. The smtp
2268 transport options dns_qualify_single and dns_search_parents are passed to
2269 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2270 of host_find_byname() use the default settings of RES_DEFNAMES
2271 (qualify_single) but not RES_DNSRCH (search_parents).
2273 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2274 spool_read_header() do less string testing, by means of a preliminary
2275 switch on the second character of optional "-foo" lines. (This is
2276 overdue, caused by the large number of possibilities that now exist.
2277 Originally there were few.) While I was there, I also converted the
2278 str(n)cmp tests so they don't re-test the leading "-" and the first
2279 character, in the hope this might squeeze out yet more improvement.
2281 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2282 flag allowing group syntax was set by the header_syntax check but not
2283 turned off, possible causing trouble later; (2) The flag was not being
2284 set at all for the header_verify test, causing "group"-style headers to
2285 be rejected. I have now set it in this case, and also caused header_
2286 verify to ignore an empty address taken from a group. While doing this, I
2287 came across some other cases where the code for allowing group syntax
2288 while scanning a header line wasn't quite right (mostly, not resetting
2289 the flag correctly in the right place). These bugs could have caused
2290 trouble for malformed header lines. I hope it is now all correct.
2292 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2293 with the "reply" argument non-NULL. The code, however (which originally
2294 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2295 but it didn't always do it. This confused somebody who was copying the
2296 code for some other use. I have removed all the tests.
2298 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2299 feature that was used to support insecure browsers during the U.S. crypto
2300 embargo. It requires special client support, and Exim is probably the
2301 only MTA that supported it -- and would never use it because real RSA is
2302 always available. This code has been removed, because it had the bad
2303 effect of slowing Exim down by computing (never used) parameters for the
2304 RSA_EXPORT functionality.
2306 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2307 authenticator to fail if there's a tab character in the incoming data
2308 (there should never be unless someone is messing about, as it's supposed
2309 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2310 if the connection is using TLS or if the remote IP is the same as the
2311 local IP, and the "valid-client-cert option" if a client certificate has
2314 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2315 authenticators. This can be used for authorization after authentication
2316 succeeds. (In the case of plaintext, it servers for both authentication
2319 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2320 if any retry times were supplied.
2322 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2323 connection, where there is no client IP address to check. In this
2324 situation, the verify now always succeeds.
2326 PH/25 Applied John Jetmore's -Mset patch.
2328 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2330 PH/27 In a string expansion for a processed (not raw) header when multiple
2331 headers of the same name were present, leading whitespace was being
2332 removed from all of them, but trailing whitespace was being removed only
2333 from the last one. Now trailing whitespace is removed from each header
2334 before concatenation. Completely empty headers in a concatenation (as
2335 before) are ignored.
2337 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2338 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2340 PH/29 [Removed. This was a change that I later backed out, and forgot to
2341 correct the ChangeLog entry (that I had efficiently created) before
2342 committing the later change.]
2344 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2345 address errors (4xx response to RCPT) over the same connection as other
2346 messages routed to the same hosts. Such deliveries are always "forced",
2347 so retry times are not inspected. This resulted in far too many retries
2348 for the affected addresses. The effect occurred only when there were more
2349 hosts than the hosts_max_try setting in the smtp transport when it had
2350 the 4xx errors. Those hosts that it had tried were not added to the list
2351 of hosts for which the message was waiting, so if all were tried, there
2352 was no problem. Two fixes have been applied:
2354 (i) If there are any address or message errors in an SMTP delivery, none
2355 of the hosts (tried or untried) are now added to the list of hosts
2356 for which the message is waiting, so the message should not be a
2357 candidate for sending over the same connection that was used for a
2358 successful delivery of some other message. This seems entirely
2359 reasonable: after all the message is NOT "waiting for some host".
2360 This is so "obvious" that I'm not sure why it wasn't done
2361 previously. Hope I haven't missed anything, but it can't do any
2362 harm, as the worst effect is to miss an optimization.
2364 (ii) If, despite (i), such a delivery is accidentally attempted, the
2365 routing retry time is respected, so at least it doesn't keep
2366 hammering the server.
2368 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2369 in ${readsocket because some servers need this prod.
2371 PH/32 Added some extra debug output when updating a wait-xxx database.
2373 PH/33 The hint "could be header name not terminated by colon", which has been
2374 given for certain expansion errors for a long time, was not being given
2375 for the ${if def:h_colon_omitted{... case.
2377 PH/34 The spec says: "With one important exception, whenever a domain list is
2378 being scanned, $domain contains the subject domain." There was at least
2379 one case where this was not true.
2381 PH/35 The error "getsockname() failed: connection reset by peer" was being
2382 written to the panic log as well as the main log, but it isn't really
2383 panic-worthy as it just means the connection died rather early on. I have
2384 removed the panic log writing for the ECONNRESET error when getsockname()
2387 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2388 runs only) independently of the message's sender address. This meant
2389 that, if the 4xx error was in fact related to the sender, a different
2390 message to the same recipient with a different sender could confuse
2391 things. In particualar, this can happen when sending to a greylisting
2392 server, but other circumstances could also provoke similar problems.
2393 I have changed the default so that the retry time for these errors is now
2394 based a combination of the sender and recipient addresses. This change
2395 can be overridden by setting address_retry_include_sender=false in the
2398 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2399 remote server are returned as part of bounce messages. This was not
2400 happening for LMTP over a pipe (the lmtp transport), but now it is the
2401 same for both kinds of LMTP.
2403 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2404 in header lines that were in fact CNAMEs. This is no longer the case.
2406 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2407 and queue runs started by the daemon processed all messages. This has
2408 been fixed so that -R and -S can now usefully be given with -q<time>.
2410 PH/40 Import PCRE release 6.7 (fixes some bugs).
2412 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2414 PH/42 Give an error if -q is specified more than once.
2416 PH/43 Renamed the variables $interface_address and $interface_port as
2417 $received_ip_address and $received_port, to make it clear that these
2418 values apply to message reception, and not to the outgoing interface when
2419 a message is delivered. (The old names remain recognized, of course.)
2421 PH/44 There was no timeout on the connect() call when using a Unix domain
2422 socket in the ${readsocket expansion. There now is.
2424 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2425 be meaningful with "accept".
2427 SC/01 Eximstats V1.43
2428 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2430 SC/02 Eximstats V1.44
2431 Use a glob alias rather than an array ref in the generated
2432 parser. This improves both readability and performance.
2434 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2435 Collect SpamAssassin and rejection statistics.
2436 Don't display local sender or destination tables unless
2437 there is data to show.
2438 Added average volumes into the top table text output.
2440 SC/04 Eximstats V1.46
2441 Collect data on the number of addresses (recipients)
2442 as well as the number of messages.
2444 SC/05 Eximstats V1.47
2445 Added 'Message too big' to the list of mail rejection
2446 reasons (thanks to Marco Gaiarin).
2448 SC/06 Eximstats V1.48
2449 Mainlog lines which have GMT offsets and are too short to
2450 have a flag are now skipped.
2452 SC/07 Eximstats V1.49 (Alain Williams)
2453 Added the -emptyok flag.
2455 SC/08 Eximstats V1.50
2456 Fixes for obtaining the IP address from reject messages.
2458 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2459 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2460 whitesspace changes from 4.64-PH/27
2462 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2465 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2466 are found, allow negative numbers in numeric criteria)
2468 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2470 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2471 to match changes made in 4.64-PH/43
2473 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2475 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2476 "domain" is actually an IP address, and force a failure. This locks out
2477 those revolvers/nameservers that support "A-for-A" lookups, in
2478 contravention of the specifications.
2480 PH/48 When a host name was looked up from an IP address, and the subsequent
2481 forward lookup of the name timed out, the host name was left in
2482 $sender_host_name, contrary to the specification.
2484 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2485 restricted to single-key lookups, Exim was not diagnosing an error if
2486 * or *@ was used with a query-style lookup.
2488 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2490 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2491 long ago, but noone interested enough thought of it. Let's just say that
2492 the "1.1" means that there are some new functions that weren't there at
2493 some point in the past.
2495 PH/51 Error processing for expansion failure of helo_data from an smtp
2496 transport during callout processing was broken.
2498 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2499 tested/used via the -bh/-bhc/-bs options.
2501 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2502 bug, fixed in subsequent PCRE releases).
2504 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2505 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2507 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2513 SC/01 Use a glob alias rather than an array ref in eximstats generated
2514 parser. This improves both readability and performance.
2516 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2517 Don't display local sender or destination tables in eximstats unless
2518 there is data to show.
2519 Added average volumes into the eximstats top table text output.
2521 SC/03 Collect data on the number of addresses (recipients) as well
2522 as the number of messages in eximstats.
2524 TF/01 Correct an error in the documentation for the redirect router. Exim
2525 does (usually) call initgroups() when daemonizing.
2527 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2528 with consistent privilege compared to when running as a daemon.
2530 TF/03 Note in the spec that $authenticated_id is not set for local
2531 submissions from trusted users.
2533 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2534 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2536 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2537 by adding some example configuration directives to the default
2538 configuration file. A little bit of work is required to uncomment the
2539 directives and define how usernames and passwords are checked, but
2540 there is now a framework to start from.
2542 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2543 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2544 without this. I don't know how relevant this is to other LDAP libraries.
2546 PH/02 Add the verb name to the "unknown ACL verb" error.
2548 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2550 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2552 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2553 directory not expanded when it should be if an expanded home directory
2554 was set for the address (which is overridden by the transport).
2556 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2559 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2560 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2561 because it is too late at that time, and has no effect.
2563 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2564 security issue with \' (bugzilla #107). I could not use the
2565 PQescapeStringConn() function, because it needs a PGconn value as one of
2568 PH/08 When testing addresses using -bt, indicate those final addresses that
2569 are duplicates that would not cause an additional delivery. At least one
2570 person was confused, thinking that -bt output corresponded to deliveries.
2571 (Suppressing duplicates isn't a good idea as you lose the information
2572 about possibly different redirections that led to the duplicates.)
2574 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2575 systems where poll() doesn't work, in particular OS X.
2577 PH/10 Added more information to debugging output for retry time not reached.
2579 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2580 operations in malware.c.
2582 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2585 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2586 syslog_duplication set false, log lines that would normally be written
2587 both the the main log and to the reject log were not written to syslog at
2590 PH/14 In the default configuration, change the use of "message" in ACL warn
2591 statements to "add_header".
2593 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2594 not followed by a command (e.g. "seen endif").
2596 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2597 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2600 PH/17 Added extra conditions to the default value of delay_warning_condition
2604 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2605 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2606 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2609 The Auto-Submitted: and various List- headers are standardised, whereas I
2610 don't think Precedence: ever was.
2612 PH/18 Refactored debugging code in route_finduser() to show more information,
2613 in particular, the error code if getpwnam() issues one.
2615 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2616 This is apparently needed in addition to the PH/07 change above to avoid
2617 any possible encoding problems.
2619 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2620 but not after initializing Perl.
2622 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2623 output them only if debugging. By default they are written stderr,
2624 apparently, which is not desirable.
2626 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2629 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2632 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2634 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2635 authenticated or an ident call has been made. Suppress the default
2636 values for $authenticated_id and $authenticated_sender (but permit -oMai
2637 and -oMas) when testing with -bh.
2639 PH/24 Re-jigged the order of the tests in the default configuration so that the
2640 tests for valid domains and recipients precede the DNS black list and CSA
2641 tests, on the grounds that those ones are more expensive.
2643 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2644 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2645 This bug exists in every version of Exim that I still have, right back to
2648 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2649 However, an attempt to turn on case-sensitivity in a regex key by
2650 including (?-i) didn't work because the subject string was already
2651 lowercased, and the effects were non-intuitive. It turns out that a
2652 one-line patch can be used to allow (?-i) to work as expected.
2658 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2659 other effects) broke the use of negated acl sub-conditions.
2661 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2664 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2665 "Deny" causes Exim to reject the incoming connection with a 554 error.
2666 Unfortunately, if there is a major crisis, such as a disk failure,
2667 tcp-wrappers gives "deny", whereas what one would like would be some
2668 kind of temporary error. A kludge has been added to help with this.
2669 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2670 554 error is used if errno is still zero or contains ENOENT (which occurs
2671 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2674 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2676 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2677 errors. Otherwise a message that provokes a temporary error (when other
2678 messages do not) can cause a whole host to time out.
2680 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2681 addresses were routed directly to files or pipes from a redirect router.
2682 File deliveries just didn't batch; pipe deliveries might have suffered
2685 PH/06 A failure to get a lock for a hints database would erroneously always say
2686 "Failed to get write lock", even when it was really a read lock.
2688 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2689 of 0600. This has been changed to use the value of the lockfile_mode
2690 option (which defaults to 0600).
2692 PH/08 Applied small patch from the Sieve maintainer.
2694 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2695 folder from quota calculations, a direct delivery into this folder messed
2696 up the contents of the maildirsize file. This was because the regex was
2697 used only to exclude .Trash (or whatever) when the size of the mailbox
2698 was calculated. There was no check that a delivery was happening into an
2699 excluded directory. This bug has been fixed by ignoring all quota
2700 processing for deliveries into excluded directories.
2702 PH/10 Added the maildirfolder_create_regex option to appendfile.
2708 PH/01 The code for finding all the local interface addresses on a FreeBSD
2709 system running IPv6 was broken. This may well have applied to all BSD
2710 systems, as well as to others that have similar system calls. The broken
2711 code found IPv4 interfaces correctly, but gave incorrect values for the
2712 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2713 that it would not match correctly against @[] and not recognize the IPv6
2716 PH/02 The ipliteral router was not recognizing addresses of the form user@
2717 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2719 PH/03 Added disable_ipv6.
2721 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2722 decoded form, because it is most often used to construct To: headers
2723 lines in autoreplies, and the decoded form may well be syntactically
2724 invalid. However, $reply_address has leading white space removed, and all
2725 newlines turned into spaces so that the autoreply transport does not
2728 PH/05 If group was specified without a user on a router, and no group or user
2729 was specified on a transport, the group from the router was ignored.
2731 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2732 for visible compile-time settings that can be used to change these
2733 numbers, for those that want even more. Backwards compatibility with old
2734 spool files has been maintained. However, going back to a previous Exim
2735 release will lost any variables that are in spool files.
2737 PH/07 Two small changes when running in the test harness: increase delay when
2738 passing a TCP/IP connection to a new process, in case the original
2739 process has to generate a bounce, and remove special handling of
2740 127.0.0.2 (sic), which is no longer necessary.
2742 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2743 be the same on different OS.
2745 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2748 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2749 whether --show-vars was specified or not
2751 JJ/02 exipick: Added support for new ACL variable spool format introduced
2754 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2755 syntactically invalid From: or Reply-to: line, and a filter used this to
2756 generate an autoreply, and therefore failed to obtain an address for the
2757 autoreply, Exim could try to deliver to a non-existent relative file
2758 name, causing unrelated and misleading errors. What now happens is that
2759 it logs this as a hard delivery error, but does not attempt to create a
2762 PH/11 The exinext utility has a -C option for testing purposes, but although
2763 the given file was scanned by exinext itself; it wasn't being passed on
2764 when Exim was called.
2766 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2767 an end-of-file indication when reading a command response.
2769 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2770 compiled. In many other places in Exim, IPv6 addresses are always
2771 recognized, so I have changed this. It also means that IPv4 domain
2772 literals of the form [IPV4:n.n.n.n] are now always recognized.
2774 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2775 used if the router is not running as root, for example, when verifying at
2776 ACL time, or when using -bh. The debugging output from this situation was
2777 non-existent - all you got was a failure to exec. I have made two
2780 (a) Failures to set uid/gid, the current directory, or a process leader
2781 in a subprocess such as that created by queryprogram now generate
2782 suitable debugging ouput when -d is set.
2784 (b) The queryprogram router detects when it is not running as root,
2785 outputs suitable debugging information if -d is set, and then runs
2786 the subprocess without attempting to change uid/gid.
2788 PH/15 Minor change to Makefile for building test_host (undocumented testing
2791 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2792 additional section of a DNS packet that returns MX or SRV records.
2793 Instead, it always explicitly searches for A/AAAA records. This avoids
2794 major problems that occur when a DNS server includes only records of one
2795 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2796 fixed another bug: if SRV records were looked up and the corresponding
2797 address records were *not* found in the additional section, the port
2798 values from the SRV records were lost.
2800 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2801 using the correct key (the original address) when searching the retry
2802 rules in order to find which one to use for generating the retry hint.
2804 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2805 adding the default one. Similarly, if it contains a Reply-To: header, the
2806 errors_reply_to option, if set, is not used.
2808 PH/19 When calculating a retry time, Exim used to measure the "time since
2809 failure" by looking at the "first failed" field in the retry record. Now
2810 it does not use this if it is later than than the arrival time of the
2811 message. Instead it uses the arrival time. This makes for better
2812 behaviour in cases where some deliveries succeed, thus re-setting the
2813 "first failed" field. An example is a quota failure for a huge message
2814 when small messages continue to be delivered. Without this change, the
2815 "time since failure" will always be short, possible causing more frequent
2816 delivery attempts for the huge message than are intended.
2817 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2819 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2820 $1, $2, $3) because the numerical variables can be reset during some
2821 expansion items (e.g. "match"), thereby losing the authentication data.
2823 PH/21 Make -bV show the size of off_t variables so that the test suite can
2824 decide whether to run tests for quotas > 2G.
2826 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2827 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2828 filecount value is greater than 2G or if a quota value is greater than 2G
2829 on a system where the size of off_t is not greater than 4, a panic error
2832 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2833 never match. The debug and -bh output now contains an explicit error
2834 message indicating a malformed IPv4 address or mask.
2836 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2837 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2838 PH/23 above applies.
2840 PH/25 Do not write to syslog when running in the test harness. The only
2841 occasion when this arises is a failure to open the main or panic logs
2842 (for which there is an explicit test).
2844 PH/26 Added the /no_tell option to "control=freeze".
2846 PH/27 If a host name lookup failed very early in a connection, for example, if
2847 the IP address matched host_lookup and the reverse lookup yielded a name
2848 that did not have a forward lookup, an error message of the form "no IP
2849 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2850 could be logged. Now it outputs the IP address instead of "NULL".
2852 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2853 allows the sender and the authenticated sender to be set when
2854 submitting a message from within Exim. Since child_open_exim() is
2855 documented for local_scan(), the new function should be too.
2857 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2858 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2859 results in an empty string is now treated as unset.
2861 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2863 PH/31 Added sender_verify_fail logging option.
2865 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2866 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2867 tidied the source and removed it altogether.
2869 PH/33 When a queue run was abandoned because the load average was too high, a
2870 log line was always written; now it is written only if the queue_run log
2871 selector is set. In addition, the log line for abandonment now contains
2872 information about the queue run such as the pid. This is always present
2873 in "start" and "stop" lines but was omitted from the "abandon" line.
2875 PH/34 Omit spaces between a header name and the colon in the error message that
2876 is given when verify = headers_syntax fails (if there are lots of them,
2877 the message gets confusing).
2879 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2880 names, as there are now some PTR records that contain slashes. This check
2881 is only to protect against broken name servers that fall over on strange
2882 characters, so the fact that it applies to all lookups doesn't matter.
2884 PH/36 Now that the new test suite is complete, we can remove some of the
2885 special code in Exim that was needed for the old test suite. For example,
2886 sorting DNS records because real resolvers return them in an arbitrary
2887 order. The new test suite's fake resolver always returns records in the
2890 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2891 bounces) except when queue_only is set, to avoid logging races between
2892 the different processes.
2894 PH/38 Panic-die if .include specifies a non-absolute path.
2896 PH/39 A tweak to the "H" retry rule from its user.
2898 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2899 a label. They prevented compilation on older perls.
2901 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2902 a warning to be raised on newish perls.
2904 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2905 on queue. Changes to match documented behaviour of showing count of
2906 messages matching specified criteria.
2908 PH/40 Changed the default ident timeout from 30s to 5s.
2910 PH/41 Added support for the use of login_cap features, on those BSD systems
2911 that have them, for controlling the resources used by pipe deliveries.
2913 PH/42 The content-scanning code uses fopen() to create files in which to put
2914 message data. Previously it was not paying any attention to the mode of
2915 the files. Exim runs with umask(0) because the rest of the code creates
2916 files with open(), and sets the required mode explicitly. Thus, these
2917 files were ending up world-writeable. This was not a big issue, because,
2918 being within the spool directory, they were not world-accessible. I have
2919 created a function called modefopen, which takes an additional mode
2920 argument. It sets umask(777), creates the file, chmods it to the required
2921 mode, then resets the umask. All the relevant calls to fopen() in the
2922 content scanning code have been changed to use this function.
2924 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2925 to 24 hours. This avoids potential overflow problems when processing G
2926 and H retry rules. I suspect nobody ever tinkers with this value.
2928 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2930 PH/45 When the plaintext authenticator is running as a client, the server's
2931 challenges are checked to ensure they are valid base64 strings. By
2932 default, the authentication attempt is cancelled if an invalid string is
2933 received. Setting client_ignore_invalid_base64 true ignores these errors.
2934 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2935 they are received. Thus, the responses can be made to depend on the
2936 challenges. If an invalid string is ignored, an empty string is placed in
2939 PH/46 Messages that are created by the autoreply transport now contains a
2940 References: header, in accordance with RFCs 2822 and 3834.
2942 PH/47 Added authenticated_sender_force to the smtp transport.
2944 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2946 PH/49 Installed latest patch from the Sieve maintainer.
2948 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2949 was also set, the appendfile transport was unnecessarily scanning a
2950 directory of message files (e.g. for maildir delivery) to find the count
2951 of files (along with the size), even though it did not need this
2952 information. It now does the scan only if it needs to find either the
2953 size of the count of files.
2955 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2957 PH/52 Two bugs concerned with error handling when the smtp transport is
2960 (i) Exim was not creating retry information for temporary errors given
2961 for individual recipients after the DATA command when the smtp transport
2962 was used in LMTP mode. This meant that they could be retried too
2963 frequently, and not timed out correctly.
2965 (ii) Exim was setting the flag that allows error details to be returned
2966 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2967 recipients that were returned after the DATA command.
2969 PH/53 This is related to PH/52, but is more general: for any failing address,
2970 when detailed error information was permitted to be returned to the
2971 sender, but the error was temporary, then after the final timeout, only
2972 "retry timeout exceeded" was returned. Now it returns the full error as
2973 well as "retry timeout exceeded".
2975 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2976 do this, and (what is worse) MTAs that accept it.
2978 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2979 will now be deprecated.
2981 PH/56 New os.c-cygwin from the Cygwin maintainer.
2983 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2984 formats (previously only available in exim formats via -bpr, -bpru,
2985 and -bpra. Now also available in native and exiqgrep formats)
2987 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2988 with very large, slow to parse queues
2990 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2992 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2994 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2995 responses to authentication challenges, though it was showing the
2996 challenges; (ii) I've removed the CR characters from the debug output for
2999 PH/58 Allow for the insertion of a newline as well as a space when a string
3000 is turned into more than one encoded-word during RFC 2047 encoding. The
3001 Sieve code now uses this.
3003 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3004 data_4xx, lost_connection, tls_required.
3006 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3007 message was being sent as an SMTP response.
3009 PH/61 Add -l and -k options to exicyclog.
3011 PH/62 When verifying, if an address was redirected to one new address, so that
3012 verification continued, and the new address failed or deferred after
3013 having set something in $address_data, the value of $address_data was not
3014 passed back to the ACL. This was different to the case when no
3015 redirection occurred. The value is now passed back in both cases.
3017 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3018 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3019 use login_cap.h, so on its own it isn't the distinguishing feature. The
3020 new name refers directly to the setclassresources() function.
3022 PH/65 Added configuration files for NetBSD3.
3024 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3026 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3027 is preferred over IPv4.
3029 PH/68 The bounce_return_message and bounce_return_body options were not being
3030 honoured for bounces generated during the reception of non-SMTP messages.
3031 In particular, this applied to messages rejected by the ACL. This bug has
3032 been fixed. However, if bounce_return_message is true and bounce_return_
3033 body is false, the headers that are returned for a non-SMTP message
3034 include only those that have been read before the error was detected.
3035 (In the case of an ACL rejection, they have all been read.)
3037 PH/69 The HTML version of the specification is now built in a directory called
3038 spec_html instead of spec.html, because the latter looks like a path with
3039 a MIME-type, and this confuses some software.
3041 PH/70 Catch two compiler warnings in sieve.c.
3043 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3044 function verify_get_ident() calls ip_connect() to connect a socket, but
3045 if the "connect()" function timed out, ip_connect() used to close the
3046 socket. However, verify_get_ident() also closes the socket later, and in
3047 between Exim writes to the log, which may get opened at this point. When
3048 the socket was closed in ip_connect(), the log could get the same file
3049 descriptor number as the socket. This naturally causes chaos. The fix is
3050 not to close the socket in ip_connect(); the socket should be closed by
3051 the function that creates it. There was only one place in the code where
3052 this was missing, in the iplookup router, which I don't think anybody now
3053 uses, but I've fixed it anyway.
3055 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3056 well as to direct DNS lookups. Otherwise the handling of names in host
3057 lists is inconsistent and therefore confusing.
3063 PH/01 Two changes to the default runtime configuration:
3065 (1) Move the checks for relay_from_hosts and authenticated clients from
3066 after to before the (commented out) DNS black list checks.
3068 (2) Add control=submission to the relay_from_hosts and authenticated
3069 clients checks, on the grounds that messages accepted by these
3070 statements are most likely to be submissions.
3072 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3074 (1) Generate an error if the third argument for the ${prvs expansion is
3077 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3080 (3) Reset the variables that are obtained from the first argument of
3081 ${prvscheck and used in the second argument before leaving the code,
3082 because their memory is reclaimed, so using them afterwards may do
3085 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3086 one (it's much easier than Tom thought :-).
3088 (5) Because of (4), we can now allow for the use of $prvscheck_result
3089 inside the third argument.
3091 PH/03 For some reason, the default setting of PATH when running a command from
3092 a pipe transport was just "/usr/bin". I have changed it to
3095 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3096 anything to be listed in the output from -bV.
3098 PH/05 When a filter generated an autoreply, the entire To: header line was
3099 quoted in the delivery log line, like this:
3101 => >A.N.Other <ano@some.domain> <original@ddress> ...
3103 This has been changed so that it extracts the operative address. There
3104 may be more than one such address. If so, they are comma-separated, like
3107 => >ano@some.domain,ona@other.domain <original@ddress> ...
3109 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3110 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3111 not being looked up in the rDNS to get a host name, Exim was showing the
3112 IP address twice in Received: lines, even though the IP addresses were
3113 identical. For example:
3115 Received: from [1.2.3.4] (helo=[1.2.3.4])
3117 However, if the real host name was known, it was omitting the HELO data
3118 if it matched the actual IP address. This has been tidied up so that it
3119 doesn't show the same IP address twice.
3121 PH/07 When both +timestamp and +memory debugging was on, the value given by
3122 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3123 called by the debug printing, thereby overwriting the timestamp buffer.
3124 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3126 PH/08 When the original message was included in an autoreply transport, it
3127 always said "this is a copy of the message, including all the headers",
3128 even if body_only or headers_only was set. It now gives an appropriate
3131 PH/09 Applied a patch from the Sieve maintainer which:
3133 o fixes some comments
3134 o adds the (disabled) notify extension core
3135 o adds some debug output for the result of if/elsif tests
3136 o points to the current vacation draft in the documentation
3137 and documents the missing references header update
3141 o fixes a bug in processing the envelope test (when testing
3142 multiple envelope elements, the last element determinted the
3145 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3146 Electronic Mail") by including:
3148 Auto-submitted: auto-generated
3150 in the messages that it generates (bounce messages and others, such as
3151 warnings). In the case of bounce messages for non-SMTP mesages, there was
3152 also a typo: it was using "Auto_submitted" (underscore instead of
3153 hyphen). Since every message generated by Exim is necessarily in response
3154 to another message, thes have all been changed to:
3156 Auto-Submitted: auto-replied
3158 in accordance with these statements in the RFC:
3160 The auto-replied keyword:
3162 - SHOULD be used on messages sent in direct response to another
3163 message by an automatic process,
3165 - MUST NOT be used on manually-generated messages,
3167 - MAY be used on Delivery Status Notifications (DSNs) and Message
3168 Disposition Notifications (MDNs),
3170 - MUST NOT be used on messages generated by automatic or periodic
3171 processes, except for messages which are automatic responses to
3174 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3175 to the default Received: header definition.
3177 PH/12 Added log selector acl_warn_skipped (default on).
3179 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3180 variables because (a) they are in the wrong storage pool and (b) even if
3181 they were copied, it wouldn't work properly because of the caching.
3183 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3184 checking when decoding. Apparently there are clients that generate
3185 overlong encoded strings. Why am I not surprised?
3187 PH/15 If the first argument of "${if match_address" was not empty, but did not
3188 contain an "@" character, Exim crashed. Now it writes a panic log message
3189 and treats the condition as false.
3191 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3193 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3194 extension "envelope-auth". The code is finished and in agreement with
3195 other implementations, but there is no documentation so far and in fact,
3196 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3197 not changing the active code.
3199 Print executed "if" and "elsif" statements when debugging is used. This
3200 helps a great deal to understand what a filter does.
3202 Document more things not specified clearly in RFC3028. I had all this
3203 sorted out, when out of a sudden new issues came to my mind. Oops."
3205 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3208 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3209 canonical form (as documented). However, after a host name lookup from
3210 the IP address, check_host() was doing a simple string comparison with
3211 addresses acquired from the DNS when checking that the found name did
3212 have the original IP as one of its addresses. Since any found IPv6
3213 addresses are likely to be in abbreviated form, the comparison could
3214 fail. Luckily, there already exists a function for doing the comparison
3215 by converting both addresses to binary, so now that is used instead of
3216 the text comparison.
3218 PH/20 There was another similar case to PH/19, when a complete host name was
3219 given in a host list; looking up its IP address could give an abbreviated
3220 form, whereas the current host's name might or might not be abbreviated.
3221 The same fix has been applied.
3227 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3228 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3231 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3232 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3234 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3236 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3237 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3238 Also, for "Auto-Submitted", treat anything other than "no" as
3239 non-personal, in accordance with RFC 3834. (Previously it treated
3240 anything starting "auto-" as non-personal.)
3242 TF/01 The control=submission/name=... option had a problem with syntax
3243 errors if the name included a slash character. The /name= option
3244 now slurps the rest of the string, so it can include any characters
3245 but it must come last in the list of options (after /sender_retain
3248 PH/05 Some modifications to the interface to the fake nameserver for the new
3256 TK/01 Added the "success_on_redirect" address verification option. See
3257 NewStuff for rationale and an example.
3259 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3261 PH/02 Patch to exigrep to allow it to work on syslog lines.
3263 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3264 fread() to skip over the body file's header line, because in Cygwin the
3265 header line is locked and is inaccessible.
3267 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3268 co-exist for some time) to make it clear that it is the Exim ID that is
3269 referenced, not the Message-ID: header line.
3271 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3272 string_format() function, because snprintf() does not exist on all
3275 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3276 ${stat: expansion item.
3278 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3279 protocol synchronization error", to keep the pedants happy.
3281 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3282 well as for IRIX systems, when gcc is being used. See the host.c source
3285 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3287 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3290 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3291 transport and to the smtp transport in LMTP mode.
3293 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3295 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3296 run a filter in a subprocess. This could lead to confusion in subsequent
3297 lookups in the parent process. There should also be a search_tidyup() at
3298 the end of the subprocess.
3300 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3301 only if the host matched helo_try_verify_hosts, which caused the
3302 verification to occur when the EHLO/HELO command was issued. The ACL just
3303 tested the remembered result. Now, if a previous verification attempt has
3304 not happened, "verify = helo" does it there and then.
3306 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3308 TK/03 Fix log output including CR from clamd.
3310 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3311 exist provoked a memory error which could cause a segfault.
3313 PH/15 Installed PCRE 6.2
3315 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3317 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3318 of the problem. Specifically, suggested +O2 rather than +O1 for the
3321 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3323 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3324 record, the port was not correctly logged when the outgoing_port log
3325 selector was set (it logged the transort's default port).
3327 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3328 fallback_hosts, and "hosts" in the smtp transport.
3330 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3331 host errors such as "Connection refused".
3333 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3334 authentication with radiusclient 0.4.9:
3336 - Error returned from rc_read_config was caught wrongly
3337 - Username/password not passed on to radius server due to wrong length.
3339 The presumption is that some radiusclient API changes for 4.51/PH/17
3340 were not taken care of correctly. The code is still untested by me (my
3341 Linux distribution still has 0.3.2 of radiusclient), but it was
3342 contributed by a Radius user.
3344 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3345 expanding the "port" option of the smtp transport.
3347 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3348 while reading a MIME header. Thanks to Tom Hughes for a patch.
3350 PH/24 Include config.h inside local_scan.h so that configuration settings are
3353 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3354 that in an ACL for RCPT (for example), you can examine exactly what was
3357 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3358 commands, but it was not correctly comparing the address with the actual
3359 client host address. Thus, it would show the EHLO address in Received:
3360 header lines when this was not necessary.
3362 PH/27 Added the % operator to ${eval:}.
3364 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3365 it should be ignoring failures (because with -C, for example, it has lost
3366 privilege). It wasn't ignoring creation failures other than "already
3369 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3372 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3373 because an input line was too long, either on its own, or by virtue of
3374 too many continuations, the temporary file was not being removed, and the
3375 return code was incorrect.
3377 PH/31 Missing "BOOL" in function definition in filtertest.c.
3379 PH/32 Applied Sieve patches from the maintainer.
3381 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3383 PH/33 Added "verify = not_blind".
3385 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3386 Local/Makefile (with some defaults set). These are used in built scripts
3387 such as exicyclog, but they have never been used in the exim_install
3388 script (though there are many overriding facilities there). I have
3389 arranged that the exim_install script now takes note of these two
3392 PH/35 Installed configuration files for Dragonfly.
3394 PH/36 When a locally submitted message by a trusted user did not contain a
3395 From: header, and the sender address was obtained from -f or from an SMTP
3396 MAIL command, and the trusted user did not use -F to supply a sender
3397 name, $originator_name was incorrectly used when constructing a From:
3398 header. Furthermore, $originator_name was used for submission mode
3399 messages from external hosts without From: headers in a similar way,
3400 which is clearly wrong.
3402 PH/37 Added control=suppress_local_fixups.
3404 PH/38 When log_selector = +received_sender was set, and the addition of the
3405 sender made the log line's construction buffer exactly full, or one byte
3406 less than full, an overflow happened when the terminating "\n" was
3409 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3410 when the result of a list match is failure because a DNS lookup failed.
3412 PH/40 RM_COMMAND is now used in the building process.
3414 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3415 the "build-* directories that it finds.
3417 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3418 address in a domain literal was a prefix of an interface address.
3420 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3421 when verifying a sender address, unless rewrite_headers is false.
3423 PH/44 Wrote a long comment about why errors_to addresses are verified as
3424 recipients, not senders.
3426 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3427 the ratelimit ACL was added.
3429 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3431 PH/46 Added extra information about PostgreSQL errors to the error string.
3433 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3434 avoiding the need to install special zones in a real server. This is
3435 backwards compatible; if it can't find the fake resolver, it drops back.
3436 Thus, both old and new test suites can be run.
3438 TF/02 Added util/ratelimit.pl
3440 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3441 clock is set back in time.
3443 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3444 Candler <B.Candler@pobox.com>.
3446 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3447 OK for addresses that are the result of redirections.
3449 PH/48 A number of further additions for the benefit of the new test suite,
3450 including a fake gethostbyname() that interfaces to the fake DNS resolver
3453 TF/06 The fix for widen_domains has also been applied to qualify_single and
3454 search_parents which are the other dnslookup options that can cause
3457 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3460 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3462 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3463 service" if it encountered an MX record with an empty target hostname.
3464 The message is now "an MX or SRV record indicated no SMTP service".
3466 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3467 if the DNS of the sending site is misconfigured. This is quite a
3468 common situation. This change restores the behaviour of treating a
3469 helo verification defer as a failure.
3471 PH/51 If self=fail was set on a router, the bounce message did not include the
3472 actual error message.
3478 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3480 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3481 command itself ended in error, the underlying message about the transport
3482 filter timeout was being overwritten with the pipe command error. Now the
3483 underlying error message should be appended to the second error message.
3485 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3487 PH/02 Reduce the amount of output that "make" produces by default. Full output
3488 can still be requested.
3490 PH/03 The warning log line about a condition test deferring for a "warn" verb
3491 was being output only once per connection, rather than after each
3492 occurrence (because it was using the same function as for successful
3493 "warn" verbs). This seems wrong, so I have changed it.
3495 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3496 it should not have, which might have caused a crash in the right
3497 circumstances, but probably never did.
3499 PH/04 Installed a modified version of Tony Finch's patch to make submission
3500 mode fix the return path as well as the Sender: header line, and to
3501 add a /name= option so that you can make the user's friendly name appear
3504 TF/03 Added the control = fakedefer ACL modifier.
3506 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3507 Mark Lowes for thorough testing.
3509 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3511 TK/03 Merged latest SRS patch from Miles Wilton.
3513 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3514 with the definition in sysexits.h (which is #included earlier).
3515 Fortunately, Exim does not actually use EX_OK. The code used to try to
3516 preserve the sysexits.h value, by assumimg that macro definitions were
3517 scanned for macro replacements. I have been disabused of this notion,
3518 so now the code just undefines EX_OK before #including unistd.h.
3520 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3521 in the smtp transport. When a block could not be written in a single
3522 write() function, the timeout was being re-applied to each part-write.
3523 This seems wrong - if the receiver was accepting one byte at a time it
3524 would take for ever. The timeout is now adjusted when this happens. It
3525 doesn't have to be particularly precise.
3527 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3528 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3530 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3531 without a domain if the check to <postmaster@domain> fails.
3533 SC/01 Eximstats: added -xls and the ability to specify output files
3534 (patch written by Frank Heydlauf).
3536 SC/02 Eximstats: use FileHandles for outputing results.
3538 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3540 SC/04 Eximstats: fixed display of large numbers with -nvr option
3542 SC/05 Eximstats: fixed merging of reports with empty tables.
3544 SC/06 Eximstats: added the -include_original_destination flag
3546 SC/07 Eximstats: removed tabs and trailing whitespace.
3548 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3550 TK/06 MBOX spool code: Add real "From " MBOX separator line
3551 so the .eml file is really in mbox format (even though
3552 most programs do not really care). Patch from Alex Miller.
3554 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3555 The latter is generated from $received_to and is only set if the
3556 message has one envelope recipient. SA can use these headers,
3557 obviously out-of-the-box. Patch from Alex Miller.
3559 PH/08 The ${def test on a variable was returning false if the variable's
3560 value was "0", contrary to what the specification has always said!
3561 The result should be true unless the variable is empty.
3563 PH/09 The syntax error of a character other than { following "${if
3564 def:variable_name" (after optional whitespace) was not being diagnosed.
3565 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3566 accidental colon was present, for example, could give incorrect results.
3568 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3569 result is used (not including appendfile, where other changes are about
3572 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3573 This involved changing a lot of size variables from int to off_t. It
3574 should work with maildirs and everything.
3576 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3577 spamd dying while we are connected to it.
3579 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3582 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3583 function for BDB 4.3.
3585 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3587 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3588 the sources for PCRE have been re-arranged and more files are now
3591 PH/15 The code I had for printing potentially long long variables in PH/11
3592 above was not the best (it lost precision). The length of off_t variables
3593 is now inspected at build time, and an appropriate printing format (%ld
3594 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3595 to be "long long int" or "long int". This is needed for the internal
3596 formatting function string_vformat().
3598 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3599 the configuration file to be ":syslog", then the script "guesses" where
3600 the logs files are, rather than using the compiled in default. In our
3601 case the guess is not the same as the compiled default, so the script
3602 suddenly stopped working when I started to use syslog. The patch checks
3603 to see if log_file_path is "". If so, it attempts to read it from exim
3604 with no configuration file to get the compiled in version, before it
3605 falls back to the previous guessing code."
3607 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3608 implementing BATV in an Exim configuration. See NewStuff for the gory
3611 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3612 Makefile that are specific to HP-UX.
3614 PH/18 If the "use_postmaster" option was set for a recipient callout together
3615 with the "random" option, the postmaster address was used as the MAIL
3616 FROM address for the random test, but not for the subsequent recipient
3617 test. It is now used for both.
3619 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3620 patch removes a few documentation additions to RFC 3028, because the
3621 latest draft now contains them. It adds the new en;ascii-case comparator
3622 and a new error check for 8bit text in MIME parts. Comparator and
3623 require names are now matched exactly. I enabled the subaddress
3624 extension, but it is not well tested yet (read: it works for me)."
3626 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3627 rework some of the code of TK/09 above to avoid the hardwired use of
3628 "%lld" and "long long". Replaced the call to snprintf() with a call to
3631 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3632 records point to non-existent hosts", "retry timeout exceeded", and
3633 "retry time not reached for any host after a long failure period".
3635 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3636 experimental DomainKeys support:
3638 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3639 (2) On an error such as an illegally used "control", the wrong name for
3640 the control was given.
3642 These problems did NOT occur unless DomainKeys support was compiled.
3644 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3646 PH/24 Added ${if match_ip condition.
3648 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3649 databases so that it will be absolutely obvious if a crash occurs in the
3650 DB library. This is a regular occurrence (often caused by mis-matched
3653 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3654 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3655 picked up on a user's system that detects such things. There doesn't seem
3656 to be a gcc warning option for this - only an attribute that has to be
3657 put on the function's prototype. It seems that in Fedora Core 4 they have
3658 set this on a number of new functions. No doubt there will be more in due
3661 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3662 specify a transport. However, if an address that was verified by such a
3663 router was the subject of a callout, Exim crashed because it tried to
3664 read the rcpt_include_affixes from the non-existent transport. Now it
3665 just assumes that the setting of that option is false. This bug was
3666 introduced by 4.51/PH/31.
3668 PH/28 Changed -d+all to exclude +memory, because that information is very
3669 rarely of interest, but it makes the output a lot bigger. People tend to
3670 do -d+all out of habit.
3672 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3673 code in os-type was giving problems when libc.so lives in lib64, like on
3676 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3677 aren't the modern standard, and it seems that some systems' include files
3678 don't always have them. Exim was already checking for some of the newer
3679 ones like T_AAAA, and defining it itself. I've added checks for all the
3680 record types that Exim uses.
3682 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3683 not automatically generating a new one, as it is supposed to. This
3684 prevented TLS from working. If the file did exist, but contained invalid
3685 data, a new version was generated, as expected. It was only the case of a
3686 non-existent file that was broken.
3688 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3689 with a change in libdomainkeys > 0.64.
3691 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3692 from DNS. If the selector record carries the flag, it now has
3693 precedence over the domain-wide flag.
3695 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3697 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3698 the use of a port name that isn't defined in /etc/services) occurred, the
3699 message was deferred as in a normal delivery, and thus remained on the
3700 spool, instead of being failed because of the mua_wrapper setting. This
3701 is now fixed, and I tidied up some of the mua_wrapper messages at the
3704 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3705 the messages in a hash of arrays rather than using individual hashes.
3706 This is a bit cleaner and results in dramatic memory savings, albeit
3707 at a slight CPU cost.
3709 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3710 as requested by Marc Sherman.
3712 SC/10 Eximstats: added histograms for user specified patterns as requested
3715 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3717 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3718 fopen() in the content-scanning modules that did not already have it.
3724 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3725 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3727 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3729 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3731 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3732 to test Sieve filters that use "vacation".
3734 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3735 that changes the way the GnuTLS parameters are stored in the cache file.
3736 The new format can be generated externally. For backward compatibility,
3737 if the data in the cache doesn't make sense, Exim assumes it has read an
3738 old-format file, and it generates new data and writes a new file. This
3739 means that you can't go back to an older release without removing the
3742 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3743 work if there are any delivery delays because "one_time" forces the
3744 parent to be marked "delivered", so its unseen clone is never tried
3745 again. For this reason, Exim now forbids the simultaneous setting of
3748 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3749 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3750 entry if you want to know the details. The fix, however, neglected to
3751 consider the case when local delivery batching is involved. The test for
3752 "previously delivered" was not happening when checking to see if an
3753 address could be batched with a previous (undelivered) one; under
3754 certain circumstances this could lead to multiple deliveries to the same
3757 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3758 in its include files, and this causes problems building Exim.
3760 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3761 header_syntax) but Exim was just ignoring anything given after a slash.
3762 In particular, this caused confusion with an attempt to use "verify =
3763 reverse_host_lookup/defer_ok". An error is now given when options are
3764 supplied for verify items that do not have them. (Maybe reverse_host_
3765 lookup should have a defer_ok option, but that's a different point.)
3767 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3768 defined by RFC 821) to 2048, because there were problems with some AUTH
3769 commands, and RFC 1869 says the size should be increased for extended
3770 SMTP commands that take arguments.
3772 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3775 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3776 "unknown" error; now it says that the functionality isn't in the binary.
3778 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3779 an address' error message when a string expansion fails (syntax or
3780 whatever). Otherwise the password may appear in the log. Following change
3781 PH/42 below, there is no longer a chance of it appearing in a bounce
3784 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3786 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3787 of its addresses was ever tried. (Bugzilla bug #2.)
3789 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3790 the result incorrectly in the debug output. (It correctly added a newline
3791 to what was transported.)
3793 TF/01 Added $received_time.
3795 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3796 commented out examples of how to interface to a virus scanner and to
3797 SpamAssassin. Also added commented examples of av_scanner and
3798 spamd_address settings.
3800 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3801 and controls are allowed in which ACLs. There were a couple of minor
3802 errors. Some of the entries in the conditions table (which is a table of
3803 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3804 negation of where the condition IS allowed.
3806 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3808 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3809 header file does not have a version number, so I've had to invent a new
3810 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3811 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3812 radiusclient), but it was contributed by a Radius user.
3814 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3815 files or directories, for OpenSSL.
3817 PH/19 When an Exim process that is running as root has to create an Exim log
3818 file, it does so in a subprocess that runs as exim:exim so as to get the
3819 ownership right at creation (otherwise, other Exim processes might see
3820 the file with the wrong ownership). There was no test for failure of this
3821 fork() call, which would lead to the process getting stuck as it waited
3822 for a non-existent subprocess. Forks do occasionally fail when resources
3823 run out. I reviewed all the other calls to fork(); they all seem to check
3826 PH/20 When checking for unexpected SMTP input at connect time (before writing
3827 the banner), Exim was not dealing correctly with a non-positive return
3828 from the read() function. If the client had disconnected by this time,
3829 the result was a log entry for a synchronization error with an empty
3830 string after "input=" when read() returned zero. If read() returned -1
3831 (an event I could not check), uninitialized data bytes were printed.
3832 There were reports of junk text (parts of files, etc) appearing after
3835 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3837 PH/22 Added support for macro redefinition, and (re)definition in between
3838 driver and ACL definitions.
3840 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3841 forgetting to use the resulting value; it was using the unexpanded value.
3843 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3844 hadn't been configured. The fix is from Juergen Kreileder, who
3845 understands it better than I do:
3847 "Here's what I see happening with three configured cyrus_sasl
3848 authenticators configured (plain, login, cram-md5):
3850 On startup auth_cyrus_sasl_init() gets called for each of these.
3851 This means three calls to sasl_listmech() without a specified mech_list.
3852 => SASL tests which mechs of all available mechs actually work
3853 => three warnings about OTP not working
3854 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3856 With the patch, sasl_listmech() also gets called three times. But now
3857 SASL's mech_list option is set to the server_mech specified in the the
3858 authenticator. Or in other words, the answer from sasl_listmech()
3859 gets limited to just the mech you're testing for (which is different
3861 => the return list contains just 'plain' or 'login', 'cram-md5' or
3862 nothing depending on the value of ob->server_mech.
3864 I've just tested the patch: Authentication still works fine,
3865 unavailable mechs specified in the exim configuration are still
3866 caught, and the auth.log warnings about OTP are gone."
3868 PH/25 When debugging is enabled, the contents of the command line are added
3869 to the debugging output, even when log_selector=+arguments is not
3872 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3873 answer is "GNU", and only if the return is "GNU/something" is the answer
3876 PH/27 $acl_verify_message is now set immediately after the failure of a
3877 verification in an ACL, and so is available in subsequent modifiers. In
3878 particular, the message can be preserved by coding like this:
3880 warn !verify = sender
3881 set acl_m0 = $acl_verify_message
3883 Previously, $acl_verify_message was set only while expanding "message"
3884 and "log_message" when a very denied access.
3886 PH/28 Modified OS/os.c-Linux with
3888 -#ifndef OS_LOAD_AVERAGE
3889 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3891 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3892 nomenclature these days.)
3894 PH/29 Installed patch from the Sieve maintainer that adds the options
3895 sieve_useraddress and sieve_subaddress to the redirect router.
3897 PH/30 In these circumstances:
3898 . Two addresses routed to the same list of hosts;
3899 . First host does not offer TLS;
3900 . First host accepts first address;
3901 . First host gives temporary error to second address;
3902 . Second host offers TLS and a TLS session is established;
3903 . Second host accepts second address.
3904 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3905 and peerdn, if requested) that were in fact used only for the second
3908 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3909 attention to any local part prefix or suffix that was matched by the
3910 router that accepted the address. It now behaves in the same way as it
3911 does for delivery: the affixes are removed from the local part unless
3912 rcpt_include_affixes is set on the transport.
3914 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3915 timeout during the DATA phase of an incoming message.
3917 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3918 applied a patch sanctioned by the Sieve maintainer.
3920 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3921 the uid or gid is negative. A case of a negative gid caused this to be
3922 noticed. The fix allows for either to be negative.
3924 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3925 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3928 PH/36 The change PH/12 above was broken. Fixed it.
3930 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3931 the grounds that routing the same address twice would always produce the
3932 same answer. This might have been true once, but it is certainly no
3933 longer true now. Routing a child address may depend on the previous
3934 routing that produced that child. Some complicated redirection strategies
3935 went wrong when messages had multiple recipients, and made Exim's
3936 behaviour dependent on the order in which the addresses were given.
3938 I have moved the duplicate checking until after the routing is complete.
3939 Exim scans the addresses that are assigned to local and remote
3940 transports, and removes any duplicates. This means that more work will be
3941 done, as duplicates will always all be routed, but duplicates are
3942 presumably rare, so I don't expect this is of any significance.
3944 For deliveries to pipes, files, and autoreplies, the duplicate checking
3945 still happens during the routing process, since they are not going to be
3948 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3949 It corrects a timeout issue with spamd. This is Ian's comment: "The
3950 background is that sometimes spamd either never reads data from a
3951 connection it has accepted, or it never writes response data. The exiscan
3952 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3953 blindly assumes that writes won't block so it may never time out."
3955 PH/39 Allow G after quota size as well as K and M.
3957 PH/40 The value set for $authenticated_id in an authenticator may not contain
3958 binary zeroes or newlines because the value is written to log lines and
3959 to spool files. There was no check on this. Now the value is run through
3960 the string_printing() function so that such characters are converted to
3961 printable escape sequences.
3963 PH/41 $message_linecount is a new variable that contains the total number of
3964 lines in the message. Compare $body_linecount, which is the count for the
3967 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3968 bounce and delay warning messages, except in certain special cases, which
3971 (a) An SMTP error message from a remote host;
3972 (b) A message specified in a :fail: redirection;
3973 (c) A message specified in a "fail" command in a system filter;
3974 (d) A message specified in a FAIL return from the queryprogram router;
3975 (e) A message specified by the cannot_route_message router option.
3977 In these cases only, Exim does include the error details in bounce and
3978 warning messages. There are also a few cases where bland messages such
3979 as "unrouteable address" or "local delivery error" are given.
3981 PH/43 $value is now also set for the "else" part of a ${run expansion.
3983 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3984 being worked on, but at least Exim now implements the latest version to
3987 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3988 process to complete was treated as a delivery failure, a timeout while
3989 writing the message to the pipe was logged, but erroneously treated as a
3990 successful delivery. Such timeouts include transport filter timeouts. For
3991 consistency with the overall process timeout, these timeouts are now
3992 treated as errors, giving rise to delivery failures by default. However,
3993 there is now a new Boolean option for the pipe transport called
3994 timeout_defer, which, if set TRUE, converts the failures into defers for
3995 both kinds of timeout. A transport filter timeout is now identified in
3998 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3999 systems where "make" and "gmake" are different, calling "gmake" at top
4000 level broke things. I've arranged for the value of $(MAKE) to be passed
4001 from the Makefile to this script so that it can call the same version of
4005 A note about Exim versions 4.44 and 4.50
4006 ----------------------------------------
4008 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4009 changes of various kinds. As a consequence, a big documentation update was
4010 needed. This delayed the release for rather longer than seemed good, especially
4011 in the light of a couple of (minor) security issues. Therefore, the changes
4012 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4013 release. So 4.44 and 4.50 are in effect two different branches that both start
4016 I have left the 4.50 change log unchanged; it contains all the changes since
4017 4.43. The change log for 4.44 is below; many of its items are identical to
4018 those for 4.50. This seems to be the most sensible way to preserve the
4019 historical information.
4025 1. Minor wording change to the doc/README.SIEVE file.
4027 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4028 computation of the current number of files was incorrect.
4030 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4031 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4034 4. Give more explanation in the error message when the command for a transport
4035 filter fails to execute.
4037 5. There are several places where Exim runs a non-Exim command in a
4038 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4039 was being done only for the command run by the queryprogram router. It is
4040 now done for all such subprocesses. The other cases are: ${run, transport
4041 filters, and the commands run by the lmtp and pipe transports.
4043 6. Added CONFIGURE_GROUP build-time option.
4045 7. Some older OS have a limit of 256 on the maximum number of file
4046 descriptors. Exim was using setrlimit() to set 1000 as a large value
4047 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4048 systems. I've change it so that if it can't get 1000, it tries for 256.
4050 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4051 was an oversight, and furthermore, ever since the addition of extra
4052 controls (e.g. 4.43/32), the checks on when to allow different forms of
4053 "control" were broken. There should now be diagnostics for all cases when a
4054 control that does not make sense is encountered.
4056 9. Added the /retain_sender option to "control=submission".
4058 10. $recipients is now available in the predata ACL (oversight).
4060 11. Tidy the search cache before the fork to do a delivery from a message
4061 received from the command line. Otherwise the child will trigger a lookup
4062 failure and thereby defer the delivery if it tries to use (for example) a
4063 cached ldap connection that the parent has called unbind on.
4065 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4066 of $address_data from the recipient verification was clobbered by the
4067 sender verification.
4069 13. The value of address_data from a sender verification is now available in
4070 $sender_address_data in subsequent conditions in the ACL statement.
4072 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4074 15. Added a new option "connect=<time>" to callout options, to set a different
4077 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4078 was its contents. (It was OK if the option was not defined at all.)
4080 17. A "Completed" log line is now written for messages that are removed from
4081 the spool by the -Mrm option.
4083 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4084 information about exactly what failed.
4086 19. Added -dd to debug only the daemon process.
4088 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4089 handles timeouts, both on the server side and network timeouts. Renamed the
4090 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4092 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4093 It is now set to "smtps".
4095 22. $host_address is now set to the target address during the checking of
4096 ignore_target_hosts.
4098 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4099 being passed; this would have caused $sender_host_name to have been used if
4100 matching the list had actually called for a host name (not very likely,
4101 since this list is usually IP addresses). A host name is now passed as
4104 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4105 code that specifies a non-restarting handler (typically sigaction() in
4106 modern systems) in an attempt to fix a rare and obscure crash bug.
4108 25. Narrowed the window for a race in the daemon that could cause it to ignore
4109 SIGCHLD signals. This is not a major problem, because they are used only to
4110 wake it up if nothing else does.
4112 26. A malformed maildirsize file could cause Exim to calculate negative values
4113 for the mailbox size or file count. Odd effects could occur as a result.
4114 The maildirsize information is now recalculated if the size or filecount
4117 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4118 support for a long time. Removed HAVE_SYS_VFS_H.
4120 28. Installed the latest version of exipick from John Jetmore.
4122 29. In an address list, if the pattern was not a regular expression, an empty
4123 subject address (from a bounce message) matched only if the pattern was an
4124 empty string. Non-empty patterns were not even tested. This was the wrong
4125 because it is perfectly reasonable to use an empty address as part of a
4126 database query. An empty address is now tested by patterns that are
4127 lookups. However, all the other forms of pattern expect the subject to
4128 contain a local part and a domain, and therefore, for them, an empty
4129 address still always fails if the pattern is not itself empty.
4131 30. Exim went into a mad DNS loop when attempting to do a callout where the
4132 host was specified on an smtp transport, and looking it up yielded more
4133 than one IP address.
4135 31. Re-factored the code for checking spool and log partition space into a
4136 function that finds that data and another that does the check. The former
4137 is then used to implement four new variables: $spool_space, $log_space,
4138 $spool_inodes, and $log_inodes.
4140 32. The RFC2047 encoding function was originally intended for short strings
4141 such as real names; it was not keeping to the 75-character limit for
4142 encoded words that the RFC imposes. It now respects the limit, and
4143 generates multiple encoded words if necessary. To be on the safe side, I
4144 have increased the buffer size for the ${rfc2047: expansion operator from
4147 33. It is now permitted to omit both strings after an "if" condition; if the
4148 condition is true, the result is "true". As before, when the second string
4149 is omitted, a false condition yields an empty string. This makes it less
4150 cumbersome to write custom ACL and router conditions.
4152 34. Failure to deliver a bounce message always caused it to be frozen, even if
4153 there was an errors_to setting on the router. The errors_to setting is now
4156 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4157 canonical form (fully expanded) before being placed in
4158 $sender_host_address.
4160 36. The table in the code that translates DNS record types into text (T_A to
4161 "A" for instance) was missing entries for NS and CNAME. It is just possible
4162 that this could have caused confusion if both these types were looked up
4163 for the same domain, because the text type is used as part of Exim's
4164 per-process caching. But the chance of anyone hitting this buglet seems
4167 37. The dnsdb lookup has been extended in a number of ways.
4169 (1) There is a new type, "zns", which walks up the domain tree until it
4170 finds some nameserver records. It should be used with care.
4172 (2) There is a new type, "mxh", which is like "mx" except that it returns
4173 just the host names, not the priorities.
4175 (3) It is now possible to give a list of domains (or IP addresses) to be
4176 looked up. The behaviour when one of the lookups defers can be
4177 controlled by a keyword.
4179 (4) It is now possible to specify the separator character for use when
4180 multiple records are returned.
4182 38. The dnslists ACL condition has been extended: it is now possible to supply
4183 a list of IP addresses and/or domains to be looked up in a particular DNS
4186 39. Added log_selector=+queue_time_overall.
4188 40. When running the queue in the test harness, wait just a tad after forking a
4189 delivery process, to get repeatability of debugging output.
4191 41. Include certificate and key file names in error message when GnuTLS fails
4192 to set them up, because the GnuTLS error message doesn't include the name
4193 of the failing file when there is a problem reading it.
4195 42. Allow both -bf and -bF in the same test run.
4197 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4199 44. The "Exiscan patch" is now merged into the mainline Exim source.
4201 45. Sometimes the final signoff response after QUIT could fail to get
4202 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4203 < 0 before doing a fflush(). This bug looks as though it goes back to the
4204 introduction of TLS in release 3.20, but "sometimes" must have been rare
4205 because the tests only now provoked it.
4207 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4208 (this can affect the format of dates).
4210 47. exim_tidydb, when checking for the continued existence of a message for
4211 which it has found a message-specific retry record, was not finding
4212 messages that were in split spool directories. Consequently, it was
4213 deleting retry records that should have stayed in existence.
4215 48. Steve fixed some bugs in eximstats.
4217 49. The SPA authentication driver was not abandoning authentication and moving
4218 on to the next authenticator when an expansion was forced to fail,
4219 contradicting the general specification for all authenticators. Instead it
4220 was generating a temporary error. It now behaves as specified.
4222 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4223 (the order specifies the preference for clients). The order is now AES256,
4224 AES128, 3DES, ARCFOUR128.
4226 51. Small patch to Sieve code - explicitly set From: when generating an
4229 52. Exim crashed if a remote delivery caused a very long error message to be
4230 recorded - for instance if somebody sent an entire SpamAssassin report back
4231 as a large number of 550 error lines. This bug was coincidentally fixed by
4232 increasing the size of one of Exim's internal buffers (big_buffer) that
4233 happened as part of the Exiscan merge. However, to be on the safe side, I
4234 have made the code more robust (and fixed the comments that describe what
4237 53. Now that there can be additional text after "Completed" in log lines (if
4238 the queue_time_overall log selector is set), a one-byte patch to exigrep
4239 was needed to allow it to recognize "Completed" as not the last thing in
4242 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4243 patch that reportedly fixes this has been added. I am not expert enough to
4244 create a test for it. This is what the patch creator wrote:
4246 "I found a little strange behaviour of ldap code when working with
4247 Windows 2003 AD Domain, where users was placed in more than one
4248 Organization Units. When I tried to give exim partial DN, the exit code
4249 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4250 But simultaneously result of request was absolutely normal ldap result,
4251 so I produce this patch..."
4253 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4254 REFERENCE, so I have modified the code to exclude the patch when that macro
4257 55. Some experimental protocols are using DNS PTR records for new purposes. The
4258 keys for these records are domain names, not reversed IP addresses. The
4259 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4260 leaves it alone. Component reversal etc. now happens only for IP addresses.
4263 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4265 57. Double the size of the debug message buffer (to 2048) so that more of very
4266 long debug lines gets shown.
4268 58. The exicyclog utility now does better if the number of log files to keep
4269 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4271 59. Two changes related to the smtp_active_hostname option:
4273 (1) $smtp_active_hostname is now available as a variable.
4274 (2) The default for smtp_banner uses $smtp_active_hostname instead
4275 of $primary_hostname.
4277 60. The host_aton() function is supposed to be passed a string that is known
4278 to be a valid IP address. However, in the case of IPv6 addresses, it was
4279 not checking this. This is a hostage to fortune. Exim now panics and dies
4280 if the condition is not met. A case was found where this could be provoked
4281 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4282 components; fortuitously, this particular loophole had already been fixed
4283 by change 4.50/55 above.
4285 If there are any other similar loopholes, the new check in host_aton()
4286 itself should stop them being exploited. The report I received stated that
4287 data on the command line could provoke the exploit when Exim was running as
4288 exim, but did not say which command line option was involved. All I could
4289 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4290 running as the user.
4293 61. There was a buffer overflow vulnerability in the SPA authentication code
4294 (which came originally from the Samba project). I have added a test to the
4295 spa_base64_to_bits() function which I hope fixes it.
4298 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4299 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4301 63. The daemon start-up calls getloadavg() while still root for those OS that
4302 need the first call to be done as root, but it missed one case: when
4303 deliver_queue_load_max is set with deliver_drop_privilege. This is
4304 necessary for the benefit of the queue runner, because there is no re-exec
4305 when deliver_drop_privilege is set.
4307 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4308 This has been fixed.
4310 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4311 was in use, was not putting the data itself into the right store pool;
4312 consequently, it could be overwritten for a subsequent message in the same
4313 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4316 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4318 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4319 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4320 the function were treating the return as a boolean value, which happened to
4321 work because 0=false and not-0=true, but is not correct code.
4323 68. The host_aton() function was not handling scoped IPv6 addresses (those
4324 with, for example, "%eth0" on the end) correctly.
4326 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4327 negated items (that is, ~something) in unsigned ints. Some compilers
4328 apparently mutter when there is no cast.
4330 70. If an address verification called from an ACL failed, and did not produce a
4331 user-specific message (i.e. there was only a "system" message), nothing was
4332 put in $acl_verify_message. In this situation, it now puts the system
4335 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4336 session; change 4.31/43 added the unwanted input to the log line - except
4337 that it did not do this in the start of session case. It now does.
4339 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4340 This is wrong and can cause the other end to generate a synchronization
4341 error if it is another Exim or anything else that does the synchronization
4342 check. A QUIT command is no longer sent after a timeout.
4344 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4345 during host lookups.
4347 74. The defer_ok option of callout verification was not working if it was used
4348 when verifying addresses in header lines, that is, for this case:
4350 verify = header_sender/callout=defer_ok
4352 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4353 those file descriptors could be used for SMTP connections. If anything
4354 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4355 could be sent to the SMTP client, causing chaos. The daemon now opens
4356 stdin, stdout, and stderr to /dev/null when it puts itself into the
4359 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4360 log by default. The user can override this with suitable Perl magic.
4362 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4363 the log message when discard triggers, was not working for the DATA ACL or
4364 for the non-SMTP ACL.
4366 78. Error message wording change in sieve.c.
4368 79. If smtp_accept_max_per_host was set, the number of connections could be
4369 restricted to fewer than expected, because the daemon was trying to set up
4370 a new connection before checking whether the processes handling previous
4371 connections had finished. The check for completed processes is now done
4372 earlier. On busy systems, this bug wouldn't be noticed because something
4373 else would have woken the daemon, and it would have reaped the completed
4376 80. If a message was submitted locally by a user whose login name contained one
4377 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4378 It caused a spool format error. I have fixed the spool reading code. A
4379 related problem was that the "from" clause in the Received: line became
4380 illegal because of the space(s). It is now covered by ${quote_local_part.
4382 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4385 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4387 83. Patch from Sieve maintainer to fix unterminated string problem in
4388 "vacation" handling.
4390 84. Some minor changes to the Linux configuration files to help with other
4391 OS variants using glibc.
4393 85. One more patch for Sieve to update vacation handling to latest spec.
4396 ----------------------------------------------------
4397 See the note above about the 4.44 and 4.50 releases.
4398 ----------------------------------------------------
4404 1. Change 4.43/35 introduced a bug that caused file counts to be
4405 incorrectly computed when quota_filecount was set in an appendfile
4408 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4409 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4412 3. Give more explanation in the error message when the command for a transport
4413 filter fails to execute.
4415 4. There are several places where Exim runs a non-Exim command in a
4416 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4417 was being done only for the command run by the queryprogram router. It is
4418 now done for all such subprocesses. The other cases are: ${run, transport
4419 filters, and the commands run by the lmtp and pipe transports.
4421 5. Some older OS have a limit of 256 on the maximum number of file
4422 descriptors. Exim was using setrlimit() to set 1000 as a large value
4423 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4424 systems. I've change it so that if it can't get 1000, it tries for 256.
4426 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4427 was an oversight, and furthermore, ever since the addition of extra
4428 controls (e.g. 4.43/32), the checks on when to allow different forms of
4429 "control" were broken. There should now be diagnostics for all cases when a
4430 control that does not make sense is encountered.
4432 7. $recipients is now available in the predata ACL (oversight).
4434 8. Tidy the search cache before the fork to do a delivery from a message
4435 received from the command line. Otherwise the child will trigger a lookup
4436 failure and thereby defer the delivery if it tries to use (for example) a
4437 cached ldap connection that the parent has called unbind on.
4439 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4440 of $address_data from the recipient verification was clobbered by the
4441 sender verification.
4443 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4444 was its contents. (It was OK if the option was not defined at all.)
4446 11. A "Completed" log line is now written for messages that are removed from
4447 the spool by the -Mrm option.
4449 12. $host_address is now set to the target address during the checking of
4450 ignore_target_hosts.
4452 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4453 being passed; this would have caused $sender_host_name to have been used if
4454 matching the list had actually called for a host name (not very likely,
4455 since this list is usually IP addresses). A host name is now passed as
4458 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4459 code that specifies a non-restarting handler (typically sigaction() in
4460 modern systems) in an attempt to fix a rare and obscure crash bug.
4462 15. Narrowed the window for a race in the daemon that could cause it to ignore
4463 SIGCHLD signals. This is not a major problem, because they are used only to
4464 wake it up if nothing else does.
4466 16. A malformed maildirsize file could cause Exim to calculate negative values
4467 for the mailbox size or file count. Odd effects could occur as a result.
4468 The maildirsize information is now recalculated if the size or filecount
4471 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4472 support for a long time. Removed HAVE_SYS_VFS_H.
4474 18. Updated exipick to current release from John Jetmore.
4476 19. Allow an empty sender to be matched against a lookup in an address list.
4477 Previously the only cases considered were a regular expression, or an
4480 20. Exim went into a mad DNS lookup loop when doing a callout where the
4481 host was specified on the transport, if the DNS lookup yielded more than
4484 21. The RFC2047 encoding function was originally intended for short strings
4485 such as real names; it was not keeping to the 75-character limit for
4486 encoded words that the RFC imposes. It now respects the limit, and
4487 generates multiple encoded words if necessary. To be on the safe side, I
4488 have increased the buffer size for the ${rfc2047: expansion operator from
4491 22. Failure to deliver a bounce message always caused it to be frozen, even if
4492 there was an errors_to setting on the router. The errors_to setting is now
4495 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4496 canonical form (fully expanded) before being placed in
4497 $sender_host_address.
4499 24. Updated eximstats to version 1.33
4501 25. Include certificate and key file names in error message when GnuTLS fails
4502 to set them up, because the GnuTLS error message doesn't include the name
4503 of the failing file when there is a problem reading it.
4505 26. Expand error message when OpenSSL has problems setting up cert/key files.
4508 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4509 (this can affect the format of dates).
4511 28. exim_tidydb, when checking for the continued existence of a message for
4512 which it has found a message-specific retry record, was not finding
4513 messages that were in split spool directories. Consequently, it was
4514 deleting retry records that should have stayed in existence.
4516 29. eximstats updated to version 1.35
4517 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4518 1.35 - bugfix such that pie charts by volume are generated correctly
4520 30. The SPA authentication driver was not abandoning authentication and moving
4521 on to the next authenticator when an expansion was forced to fail,
4522 contradicting the general specification for all authenticators. Instead it
4523 was generating a temporary error. It now behaves as specified.
4525 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4526 (the order specifies the preference for clients). The order is now AES256,
4527 AES128, 3DES, ARCFOUR128.
4529 31. Small patch to Sieve code - explicitly set From: when generating an
4532 32. Exim crashed if a remote delivery caused a very long error message to be
4533 recorded - for instance if somebody sent an entire SpamAssassin report back
4534 as a large number of 550 error lines. This bug was coincidentally fixed by
4535 increasing the size of one of Exim's internal buffers (big_buffer) that
4536 happened as part of the Exiscan merge. However, to be on the safe side, I
4537 have made the code more robust (and fixed the comments that describe what
4540 33. Some experimental protocols are using DNS PTR records for new purposes. The
4541 keys for these records are domain names, not reversed IP addresses. The
4542 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4543 leaves it alone. Component reversal etc. now happens only for IP addresses.
4546 34. The host_aton() function is supposed to be passed a string that is known
4547 to be a valid IP address. However, in the case of IPv6 addresses, it was
4548 not checking this. This is a hostage to fortune. Exim now panics and dies
4549 if the condition is not met. A case was found where this could be provoked
4550 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4551 components; fortuitously, this particular loophole had already been fixed
4552 by change 4.50/55 or 4.44/33 above.
4554 If there are any other similar loopholes, the new check in host_aton()
4555 itself should stop them being exploited. The report I received stated that
4556 data on the command line could provoke the exploit when Exim was running as
4557 exim, but did not say which command line option was involved. All I could
4558 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4559 running as the user.
4562 35. There was a buffer overflow vulnerability in the SPA authentication code
4563 (which came originally from the Samba project). I have added a test to the
4564 spa_base64_to_bits() function which I hope fixes it.
4567 36. The daemon start-up calls getloadavg() while still root for those OS that
4568 need the first call to be done as root, but it missed one case: when
4569 deliver_queue_load_max is set with deliver_drop_privilege. This is
4570 necessary for the benefit of the queue runner, because there is no re-exec
4571 when deliver_drop_privilege is set.
4573 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4574 was in use, was not putting the data itself into the right store pool;
4575 consequently, it could be overwritten for a subsequent message in the same
4576 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4579 38. Sometimes the final signoff response after QUIT could fail to get
4580 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4581 < 0 before doing a fflush(). This bug looks as though it goes back to the
4582 introduction of TLS in release 3.20, but "sometimes" must have been rare
4583 because the tests only now provoked it.
4589 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4590 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4591 Now it returns an integer. A number of calls were still expecting a T/F
4592 return. Fortuitously, in all cases, the tests worked in OK situations,
4593 which is the norm. However, things would have gone wrong on any write
4594 failures on the smtp file descriptor. This function is used when sending
4595 messages over SMTP and also when doing verify callouts.
4597 2. When Exim is called to do synchronous delivery of a locally submitted
4598 message (the -odf or -odi options), it no longer closes stderr before doing
4601 3. Implemented the mua_wrapper option.
4603 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4605 5. Implemented the functions header_remove(), header_testname(),
4606 header_add_at_position(), and receive_remove_recipient(), and exported them
4609 6. If an ACL "warn" statement specified the addition of headers, Exim already
4610 inserted X-ACL-Warn: at the start if there was no header name. However, it
4611 was not making this test for the second and subsequent header lines if
4612 there were newlines in the string. This meant that an invalid header could
4613 be inserted if Exim was badly configured.
4615 7. Allow an ACL "warn" statement to add header lines at the start or after all
4616 the Received: headers, as well as at the end.
4618 8. Added the rcpt_4xx retry error code.
4620 9. Added postmaster_mailfrom=xxx to callout verification option.
4622 10. Added mailfrom=xxxx to the callout verification option, for verify=
4625 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4626 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4628 12. Inserted some casts to stop certain compilers warning when using pointer
4629 differences as field lengths or precisions in printf-type calls (mostly
4630 affecting debugging statements).
4632 13. Added optional readline() support for -be (dynamically loaded).
4634 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4635 same clock tick as a message's arrival, so that its received time was the
4636 same as the "first fail" time on the retry record, and that message
4637 remained on the queue past the ultimate address timeout, every queue runner
4638 would try a delivery (because it was past the ultimate address timeout) but
4639 after another failure, the ultimate address timeout, which should have then
4640 bounced the address, did not kick in. This was a "< instead of <=" error;
4641 in most cases the first failure would have been in the next clock tick
4642 after the received time, and all would be well.
4644 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4645 being recognized when the domain list was tested by the match_domain
4646 condition in an expansion string.
4648 16. Added the ${str2b64: operator.
4650 17. Exim was always calling setrlimit() to set a large limit for the number of
4651 processes, without checking whether the existing limit was already
4652 adequate. (It did check for the limit on file descriptors.) Furthermore,
4653 errors from getrlimit() and setrlimit() were being ignored. Now they are
4654 logged to the main and panic logs, but Exim does carry on, to try to do its
4655 job under whatever limits there are.
4657 18. Imported PCRE 5.0.
4659 19. Trivial typo in log message " temporarily refused connection" (the leading
4662 20. If the log selector return_path_on_delivery was set and an address was
4663 redirected to /dev/null, the delivery process crashed because it assumed
4664 that a return path would always be set for a "successful" delivery. In this
4665 case, the whole delivery is bypassed as an optimization, and therefore no
4668 21. Internal re-arrangement: the function for sending a challenge and reading
4669 a response while authentication was assuming a zero-terminated challenge
4670 string. It's now changed to take a pointer and a length, to allow for
4671 binary data in such strings.
4673 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4675 23. Exim was not respecting finduser_retries when seeking the login of the
4676 uid under which it was called; it was always trying 10 times. (The default
4677 setting of finduser_retries is zero.) Also, it was sleeping after the final
4678 failure, which is pointless.
4680 24. Implemented tls_on_connect_ports.
4682 25. Implemented acl_smtp_predata.
4684 26. If the domain in control=submission is set empty, Exim assumes that the
4685 authenticated id is a complete email address when it generates From: or
4686 Sender: header lines.
4688 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4689 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4690 chown and chgrp in /bin and hostname in /usr/bin.
4692 28. Exim was keeping the "process log" file open after each use, just as it
4693 does for the main log. This opens the possibility of it remaining open for
4694 long periods when the USR1 signal hits a daemon. Occasional processlog
4695 errors were reported, that could have been caused by this. Anyway, it seems
4696 much more sensible not to leave this file open at all, so that is what now
4699 29. The long-running daemon process does not normally write to the log once it
4700 has entered its main loop, and it closes the log before doing so. This is
4701 so that log files can straightforwardly be renamed and moved. However,
4702 there are a couple of unusual error situations where the daemon does write
4703 log entries, and I had neglected to close the log afterwards.
4705 30. The text of an SMTP error response that was received during a remote
4706 delivery was being truncated at 512 bytes. This is too short for some of
4707 the long messages that one sometimes sees. I've increased the limit to
4710 31. It is now possible to make retry rules that apply only when a message has a
4711 specific sender, in particular, an empty sender.
4713 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4714 it possible to be selective about when SMTP synchronization is enforced.
4716 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4718 32. Implemented hosts_connection_nolog.
4720 33. Added an ACL for QUIT.
4722 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4725 35. Added mailbox_size and mailbox_filecount to appendfile.
4727 36. Added control = no_multiline_responses to ACLs.
4729 37. There was a bug in the logic of the code that waits for the clock to tick
4730 in the case where the clock went backwards by a substantial amount such
4731 that the microsecond fraction of "now" was more than the microsecond
4732 fraction of "then" (but the whole seconds number was less).
4734 38. Added support for the libradius Radius client library this is found on
4735 FreeBSD (previously only the radiusclient library was supported).
4741 1. When certain lookups returned multiple values in the form name=value, the
4742 quoting of the values was not always being done properly. Specifically:
4743 (a) If the value started with a double quote, but contained no whitespace,
4745 (b) If the value contained whitespace other than a space character (i.e.
4746 tabs or newlines or carriage returns) it was not quoted.
4747 This fix has been applied to the mysql and pgsql lookups by writing a
4748 separate quoting function and calling it from the lookup code. The fix
4749 should probably also be applied to nisplus, ibase and oracle lookups, but
4750 since I cannot test any of those, I have not disturbed their existing code.
4752 2. A hit in the callout cache for a specific address caused a log line with no
4753 reason for rejecting RCPT. Now it says "Previous (cached) callout
4754 verification failure".
4756 3. There was an off-by-one bug in the queryprogram router. An over-long
4757 return line was truncated at 256 instead of 255 characters, thereby
4758 overflowing its buffer with the terminating zero. As well as fixing this, I
4759 have increased the buffer size to 1024 (and made a note to document this).
4761 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4762 when Exim is waiting for an SMTP response from a remote server, Exim
4763 restarts its select() call on the socket, thereby resetting its timeout.
4764 This is not a problem when such interrupts are rare. Somebody set up a cron
4765 job to run exiwhat every 2 minutes, which is less than the normal select()
4766 timeout (5 or 10 minutes). This meant that the select() timeout never
4767 kicked in because it was always reset. I have fixed this by comparing the
4768 time when an interrupt arrives with the time at the start of the first call
4769 to select(). If more time than the timeout has elapsed, the interrupt is
4770 treated as a timeout.
4772 5. Some internal re-factoring in preparation for the addition of Sieve
4773 extensions (by MH). In particular, the "personal" test is moved to a
4774 separate function, and given an option for scanning Cc: and Bcc: (which is
4775 not set for Exim filters).
4777 6. When Exim created an email address using the login of the caller as the
4778 local part (e.g. when creating a From: or Sender: header line), it was not
4779 quoting the local part when it contained special characters such as @.
4781 7. Installed new OpenBSD configuration files.
4783 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4784 try to make them clearer.
4786 9. Callout options, other than the timeout value, were being ignored when
4787 verifying sender addresses in header lines. For example, when using
4789 verify = header_sender/callout=no_cache
4791 the cache was (incorrectly) being used.
4793 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4794 only the Cygwin environment.
4796 11. When return_path_on_delivery was set as a log selector, if different remote
4797 addresses in the same message used different return paths and parallel
4798 remote delivery occurred, the wrong values would sometimes be logged.
4799 (Whenever a remote delivery process finished, the return path value from
4800 the most recently started remote delivery process was logged.)
4802 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4803 lines when AUTH and/or TLS are in use. This is the "received protocol"
4804 field. Exim used to use "asmtp" for authenticated SMTP, without any
4805 indication (in the protocol name) for TLS use. Now it follows the RFC and
4806 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4807 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4808 names appear in log lines as well as in Received: header lines.
4810 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4811 extensions, and comparison tests, and to fix some bugs.
4813 14. Changes to the "personal" filter test:
4815 (1) The test was buggy in that it was just doing the equivalent of
4816 "contains" tests on header lines. For example, if a user's address was
4817 anne@some.where, the "personal" test would incorrectly be true for
4819 To: susanne@some.where
4821 This test is now done by extracting each address from the header in turn,
4822 and checking the entire address. Other tests that are part of "personal"
4823 are now done using regular expressions (for example, to check local parts
4824 of addresses in From: header lines).
4826 (2) The list of non-personal local parts in From: addresses has been
4827 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4828 taken from the Sieve specification recommendations.
4830 (3) If the message contains any header line starting with "List-" it is
4831 treated as non-personal.
4833 (4) The test for "circular" in the Subject: header line has been removed
4834 because it now seems ill-conceived.
4836 15. Minor typos in src/EDITME comments corrected.
4838 16. Installed latest exipick from John Jetmore.
4840 17. If headers_add on a router specified a text string that was too long for
4841 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4842 of string_sprintf() is now avoided.
4844 18. $message_body_size was not set (it was always zero) when running the DATA
4845 ACL and the local_scan() function.
4847 19. For the "mail" command in an Exim filter, no default was being set for
4848 the once_repeat time, causing a random time value to be used if "once" was
4849 specified. (If the value happened to be <= 0, no repeat happened.) The
4850 default is now 0s, meaning "never repeat". The "vacation" command was OK
4851 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4852 (I found it when inspecting the code).
4854 20. There is now an overall timeout for performing a callout verification. It
4855 defaults to 4 times the callout timeout, which applies to individual SMTP
4856 commands during the callout. The overall timeout applies when there is more
4857 than one host that can be tried. The timeout is checked before trying the
4858 next host. This prevents very long delays if there are a large number of
4859 hosts and all are timing out (e.g. when the network connections are timing
4860 out). The value of the overall timeout can be changed by specifying an
4861 additional sub-option for "callout", called "maxwait". For example:
4863 verify = sender/callout=5s,maxwait=20s
4865 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4866 to the end before writing, but this should make it even safer).
4868 22. Exim was forgetting that it had advertised PIPELINING for the second and
4869 subsequent messages on an SMTP connection. It was also not resetting its
4870 memory on STARTTLS and an internal HELO.
4872 23. When Exim logs an SMTP synchronization error within a session, it now
4873 records whether PIPELINING has been advertised or not.
4875 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4876 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4877 rather than long int.
4879 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4881 26. Added the never_mail option to autoreply.
4887 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4888 crash if the getsockname() call failed; this can happen if a connection is
4889 closed very soon after it is established. The problem was simply in the
4890 order in which certain operations were done, causing Exim to try to write
4891 to the SMTP stream before it had set up the file descriptor. The bug has
4892 been fixed by making things happen in the correct order.
4898 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4899 before the connection was closed, thus losing the rejection response.
4901 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4902 some early Solaris releases, but causes trouble in current releases where
4903 socklen_t is defined.
4905 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4908 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4911 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4912 that releasing the top of it at the end releases what was used for sub-
4913 expansions (unless the block got too big). However, discard this block if
4914 the first thing is a variable or header, so that we can use its block when
4915 it is dynamic (useful for very large $message_headers, for example).
4917 6. Lookups now cache *every* query, not just the most recent. A new, separate
4918 store pool is used for this. It can be recovered when all lookup caches are
4919 flushed. Lookups now release memory at the end of their result strings.
4920 This has involved some general refactoring of the lookup sources.
4922 7. Some code has been added to the store_xxx() functions to reduce the amount
4923 of flapping under certain conditions.
4925 8. log_incoming_interface used to affect only the <= reception log lines. Now
4926 it causes the local interface and port to be added to several more SMTP log
4927 lines, for example "SMTP connection from", and rejection lines.
4929 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4931 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4933 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4934 could be overwritten at the end of the current message (or the start of a
4935 new message if it was set in a HELO ACL). The value is now preserved for
4936 the duration of the SMTP connection.
4938 12. If a transport had a headers_rewrite setting, and a matching header line
4939 contained an unqualified address, that address was qualified, even if it
4940 did not match any rewriting rules. The underlying bug was that the values
4941 of the flags that permit the existence of unqualified sender and recipient
4942 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4943 non-local messages, and by -bnq for local messages) were not being
4944 preserved with the message after it was received.
4946 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4947 "next input=" as part of the text comprising the host identity instead of
4948 the correct text. The code was using the same buffer for two different
4949 strings. However, depending on which order the printing function evaluated
4950 its arguments, the bug did not always show up. Under Linux, for example, my
4951 test suite worked just fine.
4953 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4954 doesn't work with some older versions of Perl. It has been changed to "my",
4955 which in any case is probably the better facility to use.
4957 15. A really picky compiler found some instances of statements for creating
4958 error messages that either had too many or two few arguments for the format
4961 16. The size of the buffer for calls to the DNS resolver has been increased
4962 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4963 for addresses that have a lot of PTR records. This alleviates a problem; it
4964 does not fully solve it.
4966 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4967 buffer now truncates the list and logs the incident, which is the same
4968 action as happens when Exim is looking up a host name and its aliases.
4969 Previously in this situation something unpredictable would happen;
4970 sometimes it was "internal error: store_reset failed".
4972 18. If a server dropped the connection unexpectedly when an Exim client was
4973 using GnuTLS and trying to read a response, the client delivery process
4974 crashed while trying to generate an error log message.
4976 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4977 string, for example:
4979 warn message = H1: something\nH2: something
4981 the text was added as a single header line from Exim's point of view
4982 though it ended up OK in the delivered message. However, searching for the
4983 second and subsequent header lines using $h_h2: did not work. This has been
4984 fixed. Similarly, if a system filter added multiple headers in this way,
4985 the routers could not see them.
4987 20. Expanded the error message when iplsearch is called with an invalid key to
4988 suggest using net-iplsearch in a host list.
4990 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4991 ACLs are no longer actually imposed (and a message to that effect is
4994 22. If a "gecos" field in a passwd entry contained escaped characters, in
4995 particular, if it contained a \" sequence, Exim got it wrong when building
4996 a From: or a Sender: header from that name. A second bug also caused
4997 incorrect handling when an unquoted " was present following a character
4998 that needed quoting.
5000 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5001 was not being matched caselessly.
5003 24. Arranged for all hyphens in the exim.8 source to be escaped with
5006 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5007 properly. Recipient callout cache records were still being keyed to include
5008 the sender, even when use_sender was set false. This led to far more
5009 callouts that were necessary. The sender is no longer included in the key
5010 when use_sender is false.
5012 26. Added "control = submission" modifier to ACLs.
5014 27. Added the ${base62d: operator to decode base 62 numbers.
5016 28. dnsdb lookups can now access SRV records.
5018 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5019 the configuration file.
5021 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5022 (-v) mode. This makes the output for a verbose queue run more intelligible.
5024 31. Added a use_postmaster feature to recipient callouts.
5026 32. Added the $body_zerocount variable, containing the number of binary zero
5027 bytes in the message body.
5029 33. The time of last modification of the "new" subdirectory is now used as the
5030 "mailbox time last read" when there is a quota error for a maildir
5033 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5035 35. Added +ignore_unknown as a special item in host lists.
5037 36. Code for decoding IPv6 addresses in host lists is now included, even if
5038 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5039 address was recognized as an IP address, but was then not correctly decoded
5040 into binary, causing unexpected and incorrect effects when compared with
5047 1. Very minor rewording of debugging text in manualroute to say "list of
5048 hosts" instead of "hostlist".
5050 2. If verify=header_syntax was set, and a header line with an unqualified
5051 address (no domain) and a large number of spaces between the end of the
5052 name and the colon was received, the reception process suffered a buffer
5053 overflow, and (when I tested it) crashed. This was caused by some obsolete
5054 code that should have been removed. The fix is to remove it!
5056 3. When running in the test harness, delay a bit after writing a bounce
5057 message to get a bit more predictability in the log output.
5059 4. Added a call to search_tidyup() just before forking a reception process. In
5060 theory, someone could use a lookup in the expansion of smtp_accept_max_
5061 per_host which, without the tidyup, could leave open a database connection.
5063 5. Added the variables $recipient_data and $sender_data which get set from a
5064 lookup success in an ACL "recipients" or "senders" condition, or a router
5065 "senders" option, similar to $domain_data and $local_part_data.
5067 6. Moved the writing of debug_print from before to after the "senders" test
5070 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5071 problems for message scanning, either using a data ACL, or using
5072 local_scan() because the Received: header was not generated till after they
5073 were called (in order to set the time as the time of reception completion).
5074 I have revised the way this works. The header is now generated after the
5075 body is received, but before the ACL or local_scan() are called. After they
5076 are run, the timestamp in the header is updated.
5082 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5083 before starting a queue runner without re-exec. This happened only when
5084 deliver_drop_privilege was set or when the Exim user was set to root. The
5085 effect of the bug was that timeouts during subsequent deliveries caused
5086 crashes instead of being properly handled. The handler is now left at its
5087 default (and expected) setting.
5089 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5090 message, again when deliver_drop_privilege is set or Exim is run as root.
5091 The bug described in (1) was not present in this case, but the tidying up
5092 of the other signals was missing. I have made the two cases consistent.
5094 3. The ignore_target_hosts setting on a manualroute router was being ignored
5095 for hosts that were looked up using the /MX notation.
5097 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5100 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5101 operated on the sender address. After changing the $sender_address to <>
5102 for the sender address verify, Exim was re-instated it as the original
5103 (before rewriting) address, but remembering that it had rewritten it, so it
5104 wasn't rewriting it again. This bug also had the effect of breaking the
5105 sender address verification caching when the sender address was rewritten.
5107 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5108 This has been changed so that if the ip literal address matches
5109 ignore_target_hosts, the router declines.
5111 7. Added expansion conditions match_domain, match_address, and match_local_
5112 part (NOT match_host).
5114 8. The placeholder for the Received: header didn't have a length field set.
5116 9. Added code to Exim itself and to exim_lock to test for a specific race
5117 condition that could lead to file corruption when using MBX delivery. The
5118 issue is with the lockfile that is created in /tmp. If this file is removed
5119 after a process has opened it but before that process has acquired a lock,
5120 there is the potential for a second process to recreate the file and also
5121 acquire a lock. This could lead to two Exim processes writing to the file
5122 at the same time. The added code performs the same test as UW imapd; it
5123 checks after acquiring the lock that its file descriptor still refers to
5124 the same named file.
5126 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5127 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5130 11. Added the smtp_active_hostname option. If used, this will typically be made
5131 to depend on the incoming interface address. Because $interface_address is
5132 not set up until the daemon has forked a reception process, error responses
5133 that can happen earlier (such as "too many connections") no longer contain
5136 12. If an expansion in a condition on a "warn" statement fails because a lookup
5137 defers, the "warn" statement is abandoned, and the next ACL statement is
5138 processed. Previously this caused the whole ACL to be aborted.
5140 13. Added the iplsearch lookup type.
5142 14. Added ident_timeout as a log selector.
5144 15. Added tls_certificate_verified as a log selector.
5146 16. Added a global option tls_require_ciphers (compare the smtp transport
5147 option of the same name). This controls incoming TLS connections.
5149 17. I finally figured out how to make tls_require_ciphers do a similar thing
5150 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5151 before starting the TLS session.
5153 18. Tabs are now shown as \t in -bP output.
5155 19. If the log selector return_path_on_delivery was set, Exim crashed when
5156 bouncing a message because it had too many Received: header lines.
5158 20. If two routers both had headers_remove settings, and the first one included
5159 a superfluous trailing colon, the final name in the first list and the
5160 first name in the second list were incorrectly joined into one item (with a
5161 colon in the middle).
5167 1. Added -C and -D options to the exinext utility, mainly to make it easier
5168 to include in the automated testing, but these could be helpful when
5169 multiple configurations are in use.
5171 2. The exinext utility was not formatting the output nicely when there was
5172 an alternate port involved in the retry record key, nor when there was a
5173 message id as well (for retries that were specific to a specific message
5174 and a specific host). It was also confused by IPv6 addresses, because of
5175 the additional colons they contain. I have fixed the IPv4 problem, and
5176 patched it up to do a reasonable job for IPv6.
5178 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5179 delivery, the log line now contains "pipelined" if PIPELINING was used.
5181 4. An SMTP transport process used to panic and die if the bind() call to set
5182 an explicit outgoing interface failed. This has been changed; it is now
5183 treated in the same way as a connect() failure.
5185 5. A reference to $sender_host_name in the part of a conditional expansion
5186 that was being skipped was still causing a DNS lookup. This no longer
5189 6. The def: expansion condition was not recognizing references to header lines
5190 that used bh_ and bheader_.
5192 7. Added the _cache feature to named lists.
5194 8. The code for checking quota_filecount in the appendfile transport was
5195 allowing one more file than it should have been.
5197 9. For compatibility with Sendmail, the command line option
5205 and sets the incoming protocol and host name (for trusted callers). The
5206 host name and its colon can be omitted when only the protocol is to be set.
5207 Note the Exim already has two private options, -pd and -ps, that refer to
5208 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5209 "s", but I don't think that's a major issue.
5211 10. A number of refactoring changes to the code, none of which should affect
5214 (a) The number of logging options was getting close to filling up the
5215 32-bit word that was used as a bit map. I have split them into two classes:
5216 those that are passed in the argument to log_write(), and those that are
5217 only ever tested independently outside of that function. These are now in
5218 separate 32-bit words, so there is plenty of room for expansion again.
5219 There is no change in the user interface or the logging behaviour.
5221 (b) When building, for example, log lines, the code previously used a
5222 macro that called string_cat() twice, in order to add two strings. This is
5223 not really sufficiently general. Furthermore, there was one instance where
5224 it was actually wrong because one of the argument was used twice, and in
5225 one call a function was used. (As it happened, calling the function twice
5226 did not affect the overall behaviour.) The macro has been replaced by a
5227 function that can join an arbitrary number of extra strings onto a growing
5230 (c) The code for expansion conditions now uses a table and a binary chop
5231 instead of a serial search (which was left over from when there were very
5232 few conditions). Also, it now recognizes conditions like "pam" even when
5233 the relevant support is not compiled in: a suitably worded error message is
5234 given if an attempt is made to use such a condition.
5236 11. Added ${time_interval:xxxxx}.
5238 12. A bug was causing one of the ddress fields not to be passed back correctly
5239 from remote delivery subprocesses. The field in question was not being
5240 subsequently used, so this caused to problems in practice.
5242 13. Added new log selectors queue_time and deliver_time.
5244 14. Might have fixed a bug in maildirsizefile handling that threw up
5245 "unexpected character" debug warnings, and recalculated the data
5246 unnecessarily. In any case, I expanded the warning message to give more
5249 15. Added the message "Restricted characters in address" to the statements in
5250 the default ACL that block characters like @ and % in local parts.
5252 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5253 Three changes have been made:
5255 (a) There was a serious bug; a negative response to MAIL caused the whole
5256 recipient domain to be cached as invalid, thereby blocking all messages
5257 to all local parts at the same domain, from all senders. This bug has
5258 been fixed. The domain is no longer cached after a negative response to
5259 MAIL if the sender used is not empty.
5261 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5264 (c) A new callout option, "use_sender" has been added for people who want
5265 the modified behaviour.
5271 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5274 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5275 indeed breaks things for older releases.
5277 3. Added additional logging to the case where there is a problem reading data
5278 from a filter that is running in a subprocess using a pipe, in order to
5279 try to track down a specific problem.
5281 4. Testing facility fudge: when running in the test harness and attempting
5282 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5283 getting "No route to host". Convert this to a timeout.
5285 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5288 6. Some OS don't have socklen_t but use size_t instead. This affects the
5289 fifth argument of getsockopt() amongst other things. This is now
5290 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5291 can be set for individual OS. I have set it for SunOS5, OSF1, and
5292 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5293 some earlier ones do not.
5295 7. Change 4.30/15 was not doing the test caselessly.
5297 8. The standard form for an IPv6 address literal was being rejected by address
5298 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5299 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5300 this, as well as the form without the "IPv6" on the front (but only when
5301 address literals are enabled, of course).
5303 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5305 10. Exim crashed if a message with an empty sender address specified by -f
5306 encountered a router with an errors_to setting. This could be provoked only
5307 by a command such as
5311 where an empty string was supplied; "<>" did not hit this bug.
5313 11. Installed PCRE release 4.5.
5315 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5316 remained set. It is now erased.
5318 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5319 times from message ids (which are base 36 rather than the normal 62).
5321 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5322 were being counted as actual protocol errors, and logged if the log
5323 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5324 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5325 and DATA following a set of rejected RCPTs do not count as protocol errors.
5326 In other words, Exim assumes they were pipelined, though this may not
5327 actually be the case. Of course, in all cases the client gets an
5328 appropriate error code.
5330 15. If a lookup fails in an ACL condition, a message about the failure may
5331 be available; it is used if testing the ACL cannot continue, because most
5332 such messages specify what the cause of the deferral is. However, some
5333 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5334 that caused an old message to be retained and used if a later statement
5335 caused a defer, replacing the real cause of the deferral.
5337 16. If an IP address had so many PTR records that the DNS lookup buffer
5338 was not large enough to hold them, Exim could crash while trying to process
5339 the truncated data. It now detects and logs this case.
5341 17. Further to 4.21/58, another change has been made: if (and only if) the
5342 first line of a message (the first header line) ends with CRLF, a bare LF
5343 in a subsequent header line has a space inserted after it, so as not to
5344 terminate the header.
5346 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5347 unnecessarily, used atoi() instead of strtol(), and didn't check the
5348 termination when getting file sizes from file names by regex.
5350 19. Completely re-implemented the support for maildirsize files, in the light
5351 of a number of problems with the previous contributed implementation
5352 (4.30/29). In particular:
5354 . If the quota is zero, the maildirsize file is maintained, but no quota is
5357 . If the maildir directory does not exist, it is created before any attempt
5358 to write a maildirsize file.
5360 . The quota value in the file is just a cache; if the quota is changed in
5361 the transport, the new value overrides.
5363 . A regular expression is available for excluding directories from the
5366 20. The autoreply transport checks the characters in options that define the
5367 message's headers; it allows continued headers, but it was checking with
5368 isspace() after an embedded newline instead of explicitly looking for a
5371 21. If all the "regular" hosts to which an address was routed had passed their
5372 expiry times, and had not reached their retry times, the address was
5373 bounced, even if fallback hosts were defined. Now Exim should go on to try
5376 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5377 equivalent code in the SMTP transport. Some hosts send humungous responses
5378 to HELO/EHLO, more than 1024 it seems.
5380 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5381 gives compiler warnings in some environments. I've now done it "properly",
5384 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5385 (because of problems with the built-in one) was declared to return uschar *
5386 instead of char *, causing compiler failure.
5388 25. Fixed a file descriptor leak when processing alias/forward files.
5390 26. Fixed a minor format string issue in dbfn.c.
5392 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5394 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5395 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5398 29. When a sender address is verified, it is cached, to save repeating the test
5399 when there is more than one recipient in a message. However, when the
5400 verification involves a callout, it is possible for different callout
5401 options to be set for different recipients. It is too complicated to keep
5402 track of this in the cache, so now Exim always runs a verification when a
5403 callout is required, relying on the callout cache for the optimization.
5404 The overhead is duplication of the address routing, but this should not be
5407 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5408 to be verified with callout=postmaster, and the main callout worked but the
5409 postmaster check failed, the verification correctly failed. However, if a
5410 subsequent RCPT command asked for sender verification *without* the
5411 postmaster check, incorrect caching caused this verification also to fail,
5414 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5415 it was not caching the DNS options (qualify_single, search_parents) that
5416 were used when the lookup failed. A subsequent lookup with different
5417 options therefore always gave the same answer, though there were cases
5418 where it should not have. (Example: a "domains = !$mx_any" option on a
5419 dnslookup router: the "domains" option is always processed without any
5420 widening, but the router might have qualify_single set.) Now Exim uses the
5421 cached value only when the same options are set.
5423 32. Added John Jetmore's "exipick" utility to the distribution.
5425 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5426 than a timeout (e.g. a certificate is required, and is not provided), an
5427 Exim server now closes the connection immediately. Previously it waited for
5428 the client to close - but if the client is SSL, it seems that they each
5429 wait for each other, leading to a delay before one of them times out.
5431 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5432 maintained 0.8.x compatibility because I don't think many are using it, and
5433 it is clearly obsolete.
5435 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5438 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5439 client certificate was expired. A simple patch fixes this, though I don't
5440 understand the full logic of why the verify callback is called multiple
5443 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5444 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5445 which causes problems with some clients (such as the Certicom SSL Plus
5446 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5447 disables the coutermeasure allowing Eudora to connect."
5449 38. Exim was not checking that a write() to a log file succeeded. This could
5450 lead to Bad Things if a log got too big, in particular if it hit a file
5451 size limit. Exim now panics and dies if it cannot write to a log file, just
5452 as it does if it cannot open a log file.
5454 39. Modified OS/Makefile-Linux so that it now contains
5456 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5458 The two -D definitions ensure that Exim is compiled with large file
5459 support, which makes it possible to handle log files that are bigger than
5462 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5463 instance) a domain was checked against a named list that involved a lookup,
5464 causing $domain_data to be set, then another domain was checked against the
5465 same list, then the first domain was re-checked, the value of $domain_data
5466 after the final check could be wrong. In particular, if the second check
5467 failed, it could be set empty. This bug probably also applied to
5470 41. The strip_trailing_dot option was not being applied to the address given
5471 with the -f command-line option.
5473 42. The code for reading a message's header from the spool was incrementing
5474 $received_count, but never initializing it. This meant that the value was
5475 incorrect (doubled) while delivering a message in the same process in which
5476 it was received. In the most common configuration of Exim, this never
5477 happens - a fresh exec is done - but it can happen when
5478 deliver_drop_privilege is set.
5480 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5481 it now includes up to 150 characters of the unexpected data in the log
5484 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5485 and building data strings. The size of both of these buffers was 10 000
5486 bytes - far larger than anybody would *ever* want, thought I. Needless to
5487 say, somebody hit the limit. I have increased the maximum line length to
5488 20 000 and the maximum data length of concatenated lines to 100 000. I have
5489 also fixed two bugs, because there was no checking on these buffers. Tsk,
5490 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5491 buffer is too small.
5493 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5494 lsearch lookups. Now it does.
5496 46. When parsing a route_list item in a manualroute router, a fixed-length
5497 buffer was used for the list of hosts. I made this 1024 bytes long,
5498 thinking that nobody would ever have a list of hosts that long. Wrong.
5499 Somebody had a whole pile of complicated expansion conditions, and the
5500 string was silently truncated, leading to an expansion error. It turns out
5501 that it is easier to change to an unlimited length (owing to other changes
5502 that have happened since this code was originally written) than to build
5503 structure for giving a limitation error. The length of the item that
5504 expands into the list of hosts is now unlimited.
5506 47. The lsearch lookup could not handle data where the length of text line was
5507 more than 4095 characters. Such lines were truncated, leading to shortened
5508 data being returned. It should now handle lines of any length.
5510 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5511 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5514 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5515 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5516 no longer happens in comment lines. A list of replacements is now placed
5517 at the head of all of the source files, except those whose only change is
5518 to replace PERL_COMMAND in the very first #! line.
5520 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5521 messages on the queue, with a bottom-up merge sort, using code contributed
5522 by Michael Haardt. This should make operations like -bp somewhat faster on
5523 large queues. It won't affect queue runners, except when queue_run_in_order
5526 51. Installed eximstats 1.31 in the distribution.
5528 52. Added support for SRV lookups to the dnslookup router.
5530 53. If an ACL referred to $message_body or $message_body_end, the value was not
5531 reset for any messages that followed in the same SMTP session.
5533 54. The store-handling optimization for building very long strings was not
5534 differentiating between the different store pools. I don't think this
5535 actually made any difference in practice, but I've tidied it.
5537 55. While running the routers to verify a sender address, $sender_address
5538 was still set to the sender address. This is wrong, because when routing to
5539 send a bounce to the sender, it would be empty. Therefore, I have changed
5540 it so that, while verifying a sender address, $sender_address is set to <>.
5541 (There is no change to what happens when verifying a recipient address.)
5543 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5544 target A or AAAA records (if not already returned) without resetting the
5545 qualify_single or search_parents options of the DNS resolver. These are
5546 inappropriate in this case because the targets of MX and SRV records must
5547 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5548 target that, when qualified, matched something in the local domain. These
5549 two options are now turned off when doing these lookups.
5551 57. It seems that at least some releases of Reiserfs (which does not have the
5552 concept of a fixed number of inodes) returns zero and not -1 for the
5553 number of available inodes. This interacted badly with check_spool_inodes,
5554 which assumed that -1 was the "no such thing" setting. What I have done is
5555 to check that the total number of inodes is greater than zero before doing
5556 the test of how many are available.
5558 58. When a "warn" ACL statement has a log_message modifier, the message is
5559 remembered, and not repeated. This is to avoid a lot of repetition when a
5560 message has many recipients that cause the same warning to be written.
5561 Howewer, Exim was preserving the list of already written lines for an
5562 entire SMTP session, which doesn't seem right. The memory is now reset if a
5563 new message is started.
5565 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5566 debugging output unless log_rewrite was also set.
5568 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5569 of (int)(handle) when we know that handle contains (void *)(-1).
5571 61. The Exim daemon panic-logs an error return when it closes the incoming
5572 connection. However "connection reset by peer" seems to be common, and
5573 isn't really an error worthy of noting specially, so that particular error
5576 62. When Exim is trying to find all the local interfaces, it used to panic and
5577 die if the ioctl to get the interface flags failed. However, it seems that
5578 on at least one OS (Solaris 9) it is possible to have an interface that is
5579 included in the list of interfaces, but for which you get a failure error
5580 for this call. This happens when the interface is not "plumbed" into a
5581 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5582 failure of the "get flags" call assumes that the interface is down.
5584 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5585 makes life easier for people who are doing arithmetic on fields extracted
5586 from dates, where you often get leading zeros that should not be
5587 interpreted as octal.
5589 64. Added qualify_domain to the redirect router, to override the global
5592 65. If a pathologically long header line contained very many addresses (the
5593 report of this problem mentioned 10 000) and each of them was rewritten,
5594 Exim could use up a very large amount of memory. (It kept on making new
5595 copies of the header line as it rewrote, and never released the old ones.)
5596 At the expense of a bit more processing, the header rewriting function has
5597 been changed so that it no longer eats memory in this way.
5599 66. The generation of the Received: header has been moved from the time that a
5600 message starts to be received, to the time that it finishes. The timestamp
5601 in the Received: header should now be very close to that of the <= log
5602 line. There are two side-effects of this change:
5604 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5605 logged header lines no longer include the local Received: line, because
5606 it has not yet been created. The same applies to a copy of the message
5607 that is returned to a non-SMTP sender when a message is rejected.
5609 (b) When a filter file is tested using -bf, no additional Received: header
5610 is added to the test message. After some thought, I decided that this
5613 This change does not affect the value of $received_for. It is still set
5614 after address rewriting, but before local_scan() is called.
5616 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5618 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5619 gave an unhelpful panic error message, and a defer error. I have managed to
5620 change this behaviour so that it now rejects any supplied certificate,
5621 which seems right, as the list of acceptable certificates is empty.
5623 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5624 gave an unhelpful defer error. I have not managed to make this reject any
5625 supplied certificates, but the error message it gives is "no certificate
5626 supplied", which is not helpful.
5628 70. exigrep's output now also includes lines that are not associated with any
5629 message, but which match the given pattern. Implemented by a patch from
5630 Martin Sluka, which also tidied up the Perl a bit.
5632 71. Recipient callout verification, like sender verification, was using <> in
5633 the MAIL FROM command. This isn't really the right thing, since the actual
5634 sender may affect whether the remote host accepts the recipient or not. I
5635 have changed it to use the actual sender in the callout; this means that
5636 the cache record is now keyed on a recipient/sender pair, not just the
5637 recipient address. There doesn't seem to be a real danger of callout loops,
5638 since a callout by the remote host to check the sender would use <>.
5639 [SEE ABOVE: changed after hitting problems.]
5641 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5642 temporary errors. However, in the case of such a code being given after
5643 the end of a data transmission (i.e. after ".") Exim was failing to write
5644 a retry record for the message. (Yes, there was some broken host that was
5645 actually sending 8xx at this point.)
5647 73. An unknown lookup type in a host list could cause Exim to panic-die when
5648 the list was checked. (An example that provoked this was putting <; in the
5649 middle of a list instead of at the start.) If this happened during a DATA
5650 ACL check, a -D file could be left lying around. This kind of configuration
5651 error no longer causes Exim to die; instead it causes a defer errror. The
5652 incident is still logged to the main and panic logs.
5654 74. Buglet left over from Exim 3 conversion. The message "too many messages
5655 in one connection" was written to the rejectlog but not the mainlog, except
5656 when address rewriting (yes!) was being logged.
5658 75. Added write_rejectlog option.
5660 76. When a system filter was run not as root (that is, when system_filter_user
5661 was set), the values of the $n variables were not being returned to the
5662 main process; thus, they were not subsequently available in the $sn
5665 77. Added +return_path_on_delivery log selector.
5667 78. A connection timeout was being treated differently from recipients deferred
5668 when testing hosts_max_try with a message that was older than the host's
5669 retry timeout. (The host should not be counted, thus allowing all hosts to
5670 be tried at least once before bouncing.) This may have been the cause of an
5671 occasionally reported bug whereby a message would remain on the queue
5672 longer than the retry timeout, but would be bounced if a delivery was
5673 forced. I say "may" because I never totally pinned down the problem;
5674 setting up timeout/retry tests is difficult. See also the next item.
5676 79. The ultimate address timeout was not being applied to errors that involved
5677 a combination of host plus message (for example, a timeout on a MAIL
5678 command). When an address resolved to a number of possible hosts, and they
5679 were not all tried for each delivery (e.g. because of hosts_max_try), a
5680 message could remain on the queue longer than the retry timeout.
5682 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5685 81. Fixed an obscure SMTP outgoing bug which required at least the following
5686 conditions: (a) there was another message waiting for the same server;
5687 (b) the server returned 5xx to all RCPT commands in the first message so
5688 that the message was not completed; (c) the server dropped the connection
5689 or gave a negative response to the RSET that Exim sends to abort the
5690 transaction. The observed case was a dropped connection after DATA that had
5691 been sent in pipelining mode. That is, the server had advertised PIPELINING
5692 but was not implementing it correctly. The effect of the bug was incorrect
5693 behaviour, such as trying another host, and this could lead to a crash.
5699 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5700 and daemon.c were passed as pointers to ints; they should have been
5701 pointers to socklen_t variables (which are typically unsigned ints).
5703 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5706 3. Fixed a really odd bug that affected only the testing scheme; patching a
5707 certain fixed string in the binary changed the value of another string that
5708 happened to be identical to the end of the original first string.
5710 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5711 name", it returns that address as the IP address. On some operating
5712 systems (e.g. Solaris), it also passes back the IP address string as the
5713 "host name". However, on others (e.g. Linux), it passes back an empty
5714 string. Exim wasn't checking for this, and was changing the host name to an
5715 empty string, assuming it had been canonicized.
5717 5. Although rare, it is permitted to have more than one PTR record for a given
5718 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5719 all the names associated with an address, because they do in Solaris.
5720 However, it seems that they do not in Linux for data that comes from the
5721 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5722 I found this out when I moved to a new Linux workstation and tried to run
5723 the Exim test suite.
5725 To get round this problem I have changed the code so that it now does its
5726 own call to the DNS to look up PTR records when searching for a host name.
5727 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5728 addresses that are only in /etc/hosts are still found.
5730 This behaviour is, however, controlled by an option called host_lookup_
5731 order, which defaults to "bydns:byaddr". If people want to use the other
5732 order, or indeed, just use one or the other means of lookup, they can
5733 specify it in this variable.
5735 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5736 some operating systems, this comes back from gethostbyaddr() as an empty
5737 string, and this is what Exim used to test for. However, it seems that in
5738 other systems, "." is yielded. Exim now tests for this case too.
5740 7. The values of check_spool_space and check_log_space are now held internally
5741 as a number of kilobytes instead of an absolute number of bytes. If a
5742 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5743 kilobyte. This means that much larger values can be stored.
5745 8. Exim monitor: an attempt to get the action menu when not actually pointing
5746 at a message produces an empty menu entitled "No message selected". This
5747 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5748 no entries in it ("Shell widget menu has zero width and/or height"). So I
5749 have added a single, blank menu entry in this case.
5751 9. Added ${quote_local_part.
5753 10. MIME decoding is now applied to the contents of Subject: header lines when
5756 11. Now that a reference to $sender_host_address automatically causes a reverse
5757 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5758 host lookup before query-style lookups in lists that might use this
5759 variable. This has therefore been abolished, and the "net-" prefix is no
5760 longer necessary for query-style lookups.
5762 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5763 have been a typo for LFLAGS, so it has been changed.
5765 13. The install script calls Exim with "-C /dev/null" in order to find the
5766 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5767 to be output. Howeve, since Exim outputs its version number before the
5768 error, it didn't break the script. It just looked ugly. I fixed this by
5769 always allowing "-C /dev/null" if the caller is root.
5771 14. Ignore overlarge ACL variable number when reading spool file - insurance
5772 against a later release with more variables having written the file.
5774 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5775 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5776 this, as well as the form without the "IPv6" on the front.
5778 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5779 OS/Makefile-Darwin file.
5781 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5782 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5784 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5785 variable. However, some OS use ints for inodes. I've added cast to long int
5786 to get rid of the compiler warning.
5788 19. I had forgotten to lock out "/../" in configuration file names when
5789 ALT_CONFIG_PREFIX was set.
5791 20. Routers used for verification do not need to specify transports. However,
5792 if such a router generated a host list, and callout was configured, Exim
5793 crashed, because it could not find a port number from the (non-existent)
5794 transport. It now assumes port 25 in this circumstance.
5796 21. Added the -t option to exigrep.
5798 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5799 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5800 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5801 them is set, LOOKUP_LSEARCH is forced.
5803 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5805 24. Added sender and host information to the "rejected by local_scan()" log
5806 line; previously there was no indication of these.
5808 25. Added .include_if_exists.
5810 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5811 Linux. It turns out that not all file systems support this. Apparently some
5812 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5813 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5814 not supported on the file descriptor, is now ignored when Exim is trying to
5815 sync a directory. This applies only to Linux.
5817 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5819 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5820 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5823 29. Added support for maildirsize files from supplied patch (modified a bit).
5825 30. The use of :fail: followed by an empty string could lead Exim to respond to
5826 sender verification failures with (e.g.):
5828 550 Verification failed for <xxx>
5829 550 Sender verify failed
5831 where the first response line was missing the '-' that indicates it is not
5832 the final line of the response.
5834 31. The loop for finding the name of the user that called Exim had a hardwired
5835 limit of 10; it now uses the value of finduser_retries, which is used for
5836 all other user lookups.
5838 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5841 33. Exim was neglecting to zero errno before one call of strtol() when
5842 expanding a string and expecting an integer value. On some systems this
5843 resulted in spurious "integer overflow" errors. Also, it was casting the
5844 result into an int without checking.
5846 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5847 did not work. The code looks as if it has *never* worked, though it appears
5848 to have been documented since at least releast 1.62. I have made it work.
5850 35. The "timeout_DNS" error in retry rules, also documented since at least
5851 1.62, also never worked. As it isn't clear exactly what this means, and
5852 clearly it isn't a major issue, I have abolished the feature by treating it
5853 as "timeout", and writing a warning to the main and panic logs.
5855 36. The display of retry rules for -brt wasn't always showing the error code
5858 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5859 timeout_connect_A, timeout_connect_MX.
5861 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5862 to the empty sender.
5864 39. The daemon was not analysing the content of -oX till after it had closed
5865 stderr and disconnected from the controlling terminal. This meant that any
5866 syntax errors were only noted on the panic log, and the return code from
5867 the command was 0. By re-arranging the code a little, I've made the
5868 decoding happen first, so such errors now appear on stderr, and the return
5869 code is 1. However, the actual setting up of the sockets still happens in
5870 the disconnected process, so errors there are still only recorded on the
5873 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5874 connections (as happens on some IP stacks) was logged at start up time as
5875 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5876 it from "IPv6 and IPv4", which means that two separate sockets are being
5879 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5880 says whether AF_INET or AF_INET6 was passed as an argument.
5882 42. Exiwhat output was messed up when time zones were included in log
5885 43. Exiwhat now gives more information about the daemon's listening ports,
5886 and whether -tls-on-connect was used.
5888 44. The "port" option of the smtp transport is now expanded.
5890 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5891 silently ignored. Now an error message is written to the main and panic
5894 46. There's a new ACL modifier called "logwrite" which writes to a log file
5895 as soon as it is encountered.
5897 47. Added $local_user_uid and $local_user_gid at routing time.
5899 48. Exim crashed when trying to verify a sender address that was being
5902 49. Exim was recognizing only a space character after ".include". It now also
5903 recognizes a tab character.
5905 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5906 extracting the relevant information from the specification. The man page no
5907 longer contains scrambled data for the -d option, and I've added a section
5908 at the front about calling Exim under different names.
5910 51. Added "extra_headers" argument to the "mail" command in filter files.
5912 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5915 53. Installed eximstats 1.29.
5917 54. Added transport_filter_timeout as a generic transport option.
5919 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5920 Cc: header lines. This was required by RFC 822, but it not required by RFC
5923 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5924 incoming messages that did not have them. Now it does so only if the
5925 message originates locally, that is, if there is no associated remote host
5926 address. When Resent- header lines are present, this applies to the Resent-
5927 lines rather than the non-Resent- lines.
5929 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5930 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5932 58. Messages for configuration errors now include the name of the main
5933 configuration file - useful now that there may be more than one file in a
5934 list (.included file names were always shown).
5936 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5937 for those rare installations that do not start the daemon as root or run it
5938 setuid root. I've cut out the call to initgroups() if the daemon is not
5941 60. The Exim user and group can now be bound into the binary as text strings
5942 that are looked up at the start of Exim's processing.
5944 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5946 62. Added $mailstore_basename variable.
5948 63. Installed patch to sieve.c from Michael Haardt.
5950 64. When Exim failed to open the panic log after failing to open the main log,
5951 the original message it was trying to log was written to stderr and debug
5952 output, but if they were not available (the usual case in production), it
5953 was lost. Now it is written to syslog before the two lines that record the
5954 failures to open the logs.
5956 65. Users' Exim filters run in subprocesses under the user's uid. It is
5957 possible for a "deliver" command or an alias in a "personal" command to
5958 provoke an address rewrite. If logging of address rewriting is configured,
5959 this fails because the process is not running as root or exim. There may be
5960 a better way of dealing with this, but for the moment (because 4.30 needs
5961 to be released), I have disabled address rewrite logging when running a
5962 filter in a non-root, non-exim process.
5968 1. The buildconfig auxiliary program wasn't quoting the value set for
5969 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5970 not defined. This bug was masked in 4.22 by the effect that was fixed in
5973 2. Some messages that were rejected after a message id was allocated were
5974 shown as "incomplete" by exigrep. It no longer does this for messages that
5975 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5977 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5978 have allow_domain_literals set, the ID did not get logged in the <= line.
5979 Domain literals are now always recognized in Message-ID: header lines.
5981 4. The first argument for a ${extract expansion item is the key name or field
5982 number. Leading and trailing spaces in this item were not being ignored,
5983 causing some misleading effects.
5985 5. When deliver_drop_privilege was set, single queue runner processes started
5986 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5987 same command in the process it spins off) were not dropping privilege.
5989 6. When the daemon running as "exim" started a queue runner, it always
5990 re-executed Exim in the spun-off process. This is a waste of effort when
5991 deliver_drop_privilege is set. The new process now just calls the
5992 queue-runner function directly.
5998 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6001 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6002 ignored. Though the use of -oP was forcing the writing of a pid file, it
6003 was always written to the default place.
6005 3. If the message "no IP address found for host xxxx" is generated during
6006 incoming verification, it is now followed by identification of the incoming
6007 connection (so you can more easily find what provoked it).
6009 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6011 5. Added some features to "harden" Exim a bit more against certain attacks:
6013 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6014 be put in Local/Makefile. This is like the never_users runtime option,
6015 but it cannot be overridden. The default setting is "root".
6017 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6018 prefix string with which any file named in a -C command line option
6021 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6022 is retained for -C and -D only if the caller of Exim is root. Without
6023 it, the exim user may also use -C and -D and retain privilege.
6025 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6026 command line option is disabled.
6028 6. Macro names set by the -D option must start with an upper case letter, just
6029 like macro names defined in the configuration file.
6031 7. Added "dereference=" facility to LDAP.
6033 8. Two instances of the typo "uknown" in the source files are fixed.
6035 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6036 the Configure-Makefile script screwed up while processing it.
6038 10. Incorporated PCRE 4.4.
6040 11. The SMTP synchronization check was not operating right at the start of an
6041 SMTP session. For example, it could not catch a HELO sent before the client
6042 waited for the greeting. There is now a check for outstanding input at the
6043 point when the greeting is written. Because of the duplex, asynchronous
6044 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6045 way, but not yet received, when the check is performed.
6047 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6048 on TCP/IP sockets, because this apparently causes some broken clients to
6051 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6052 unchanged) from the Cygwin maintainer.
6054 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6055 was supported instead of testing for mbx. Effectively a typo.
6057 15. The spa authenticator server code was not checking that the input it
6058 received was valid base64.
6060 16. The debug output line for the "set" modifier in ACLs was not showing the
6061 name of the variable that was being set.
6063 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6065 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6066 lookup on its own. Something else was needed to trigger the lookup. For
6067 example, a match in host_lookup or the need for a host name in a host list.
6068 Now, if $sender_host_name is referenced and the host name has not yet been
6069 looked up, a lookup is performed. If the lookup fails, the variable remains
6070 empty, and $host_lookup_failed is set to "1".
6072 19. Added "eqi" as a case-independent comparison operator.
6074 20. The saslauthd authentication condition could segfault if neither service
6075 nor realm was specified.
6077 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6078 error message that was logged was misleading, and incoming SMTP
6079 connections were dropped. The message is now more accurate, and temporary
6080 errors are given to SMTP connections.
6082 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6083 (except RSET) with a 421 error, until QUIT is received. However, it was
6084 failing to send a response to QUIT.
6086 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6087 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6088 ACL is now run after the helo_try_verify_hosts code.
6090 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6091 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6092 case-independent, so other case variants are also recognized). Apparently
6093 some systems use these upper case variants.
6095 25. If more than two messages were waiting for the same host, and a transport
6096 filter was specified for the transport, Exim sent two messages over the
6097 same TCP/IP connection, and then failed with "socket operation on non-
6098 socket" when it tried to send the third.
6100 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6102 27. The extern definition of crypt16() in expand.c was not being excluded when
6103 the OS had its own crypt16() function.
6105 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6106 as a preferred synonym for return_size_limit, both as an option and as an
6109 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6111 30. Changed the default configuration ACL to relax the local part checking rule
6112 for addresses that are not in any local domains. For these addresses,
6113 slashes and pipe symbols are allowed within local parts, but the sequence
6114 /../ is explicitly forbidden.
6116 31. SPA server authentication was not clearing the challenge buffer before
6119 32. log_message in a "warn" ACL statement was writing to the reject log as
6120 well as to the main log, which contradicts the documentation and doesn't
6121 seem right (because no rejection is happening). So I have stopped it.
6123 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6124 However, I am unable to do any testing of this.
6126 34. Fixed an infelicity in the appendfile transport. When checking directories
6127 for a mailbox, to see if any needed to be created, it was accidentally
6128 using path names with one or more superfluous leading slashes; tracing
6129 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6131 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6132 contents are added to the log line that is written for every discarded
6133 recipient. (Previously a log_message setting was ignored.)
6135 36. The ${quote: operator now quotes the string if it is empty.
6137 37. The install script runs exim in order to find its version number. If for
6138 some reason other than non-existence or emptiness, which it checks, it
6139 could not run './exim', it was installing it with an empty version number,
6140 i.e. as "exim-". This error state is now caught, and the installation is
6143 38. An argument was missing from the function that creates an error message
6144 when Exim fails to connect to the socket for saslauthd authentication.
6145 This could cause Exim to crash, or give a corrupted message.
6147 39. Added isip, isip4, and isip6 to ${if conditions.
6149 40. The ACL variables $acl_xx are now saved with the message, and can be
6150 accessed later in routers, transports, and filters.
6152 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6153 strings in the file are not string-expanded.
6155 42. If a MAIL command specified a SIZE value that was too large to fit into an
6156 int variable, the check against message_size_limit failed. Such values are
6157 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6158 day this will have to be increased, but I don't think I want to be around
6159 when emails are that large.
6166 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6167 iconv() is not standard in FreeBSD.
6169 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6170 IPv6 enabled. The observed symptom was a segmentation fault on return from
6171 the function os_common_find_running_interfaces() in src/os.c.
6173 3. In the check_special_case() function in daemon.c I had used "errno" as an
6174 argument name, which causes warnings on some systems. This was basically a
6175 typo, since it was named "eno" in the comments!
6177 4. The code that waits for the clock to tick (at a resolution of some fraction
6178 of a second) so as to ensure message-id uniqueness was always waiting for
6179 at least one whole tick, when it could have waited for less. [This is
6180 almost certainly not relevant at current processor speeds, where it is
6181 unlikely to ever wait at all. But we try to future-proof.]
6183 5. The function that sleeps for a time interval that includes fractions of a
6184 second contained a race. It did not block SIGALRM between setting the
6185 timer, and suspending (a couple of lines later). If the interval was short
6186 and the sigsuspend() was delayed until after it had expired, the suspension
6187 never ended. On busy systems this could lead to processes getting stuck for
6190 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6191 process, before it forks any delivery processes. The open lookup caching
6192 mechanism meant that the open file or database connection was passed into
6193 the delivery process. The problem was that delivery processes always tidy
6194 up cached lookup data. This could cause a problem for the next delivery
6195 process started by the queue runner, because the external queue runner
6196 process does not know about the closure. So the next delivery process
6197 still has data in the lookup cache. In the case of a file lookup, there was
6198 no problem because closing a file descriptor in a subprocess doesn't affect
6199 the parent. However, if the lookup was caching a connection to a database,
6200 the connection was closed, and the second delivery process was likely to
6201 see errors such as "PGSQL: query failed: server closed the connection
6202 unexpectedly". The problem has been fixed by closing all cached lookups
6203 in a queue runner before running a delivery process.
6205 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6206 seem to have the "const" qualifier which it has on other OS. I've
6209 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6210 *of the same type* (client or server) with the same public name that an
6211 error should be diagnosed.
6213 9. When Exim looked up a host name for an IP address, but failed to find the
6214 original IP address when looking up the host name (a safety check), it
6215 output the message "<ip address> does not match any IP for NULL", which was
6216 confusing, to say the least. The bug was that the host name should have
6217 appeared instead of "NULL".
6219 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6220 user that is built into the binary, and the -C or -D options is used, root
6221 privilege is dropped before the configuration file is read. In addition,
6222 logging is switched to stderr instead of the normal log files. If the
6223 configuration then re-defines the Exim user, the unprivileged environment
6224 is probably not what is expected, so Exim logs a panic warning message (but
6227 However, if deliver_drop_privilege is set, the unprivileged state may well
6228 be exactly what is intended, so the warning has been cut out in that case,
6229 and Exim is allowed to try to write to its normal log files.
6235 1. smtp_return_error_details was not giving details for temporary sender
6236 or receiver verification errors.
6238 2. Diagnose a configuration error if two authenticators have the same public
6241 3. Exim used not to create the message log file for a message until the first
6242 delivery attempt. This could be confusing when incoming messages were held
6243 for policy or load reasons. The message log file is now created at the time
6244 the message is received, and an initial "Received" line is written to it.
6246 4. The automatically generated man page for command line options had a minor
6247 bug that caused no ill effects; however, a more serious problem was that
6248 the procedure for building the man page automatically didn't always
6249 operate. Consequently, release 4.20 contains an out-of-date version. This
6250 shouldn't happen again.
6252 5. When building Exim with embedded Perl support, the script that builds the
6253 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6254 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6256 6. The freeze_tell option was not being used for messages that were frozen on
6257 arrival, either by an ACL or by local_scan().
6259 7. Added the smtp_incomplete_transaction log selector.
6261 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6262 was accepting AUTH without a new EHLO.
6264 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6265 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6268 10. Logging of TCP/IP connections (when configured) now happens in the main
6269 daemon process instead of the child process, so that the TCP/IP connection
6270 count is more accurate (but it can never be perfect).
6272 11. The use of "drop" in a nested ACL was not being handled correctly in the
6273 outer ACL. Now, if condition failure induced by the nested "drop" causes
6274 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6275 or "require"), the connection is dropped.
6277 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6278 that yield "discard" can now be used with an "accept" or a "discard" verb,
6279 but an error is generated for any others (because I can't see a useful way
6280 to define what should happen).
6282 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6283 are now processed in the same way as lines in the Exim configuration file.
6284 In particular, continuation lines are supported.
6286 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6288 15. Added -ti meaning -t -i.
6290 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6291 domains, and warn by logging if others are found.
6293 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6294 returned from GIFCONF when Exim is trying to find the list of interfaces on
6295 a host. The code in os.c has been modified to copy each ifreq to an aligned
6296 structure in all cases.
6298 Also, in some cases, the returned ifreq's were being copied to a 'struct
6299 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6300 means the last couple of bytes of an IPv6 address could be chopped if the
6301 ifreq contained only a normal sockaddr (14 bytes storage).
6303 18. Named domain lists were not supported in the hosts_treat_as_local option.
6304 An entry such as +xxxx was not recognized, and was treated as a literal
6307 19. Ensure that header lines added by a DATA ACL are included in the reject log
6308 if the ACL subsequently rejects the message.
6310 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6311 MD5 (which is deprecated).
6313 21. When testing a filter file using -bf, Exim was writing a message when it
6314 took the sender from a "From " line in the message, but it was not doing so
6315 when it took $return_path from a Return-Path: header line. It now does.
6317 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6318 with a valid header line field name (a series of printing characters
6319 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6321 23. Changed "disc" in the source to "disk" to conform to the documentation and
6322 the book and for uniformity.
6324 24. Ignore Sendmail's -Ooption=value command line item.
6326 25. When execve() failed while trying to run a command in a pipe transport,
6327 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6328 could be confused with a return value of 69 from the command itself. This
6329 has been changed to 127, the value the shell returns if it is asked to run
6330 a non-existent command. The wording for the related log line suggests a
6331 non-existent command as the problem.
6333 26. If received_header_text expands to an empty string, do not add a Received:
6334 header line to the message. (Well, it adds a token one on the spool, but
6335 marks it "old" so that it doesn't get used or transmitted.)
6337 27. Installed eximstats 1.28 (addition of -nt option).
6339 28. There was no check for failure on the call to getsockname() in the daemon
6340 code. This can fail if there is a shortage of resources on the system, with
6341 ENOMEM, for example. A temporary error is now given on failure.
6343 29. Contrary to the C standard, it seems that in some environments, the
6344 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6345 program. Exim now does this explicitly; it affects the formatting of
6346 timestamps using strftime().
6348 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6349 complaints. I've now initialized all the variables, to avoid this.
6351 32. Header lines added by a system filter were not being "seen" during
6352 transport-time rewrites.
6354 33. The info_callback() function passed to OpenSSL is set up with type void
6355 (*)(SSL *, int, int), as described somewhere. However, when calling the
6356 function (actually a macro) that sets it up, the type void(*)() is
6357 expected. I've put in a cast to prevent warnings from picky compilers.
6359 34. If a DNS black list lookup found a CNAME record, but there were no A
6360 records associated with the domain it pointed at, Exim crashed.
6362 35. If a DNS black list lookup returned more than one A record, Exim ignored
6363 all but the first. It now scans all returned addresses if a particular IP
6364 value is being sought. In this situation, the contents of the
6365 $dnslist_value variable are a list of all the addresses, separated by a
6368 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6369 to accept a host name and all its aliases if the forward lookup for any of
6370 them yielded the IP address of the incoming connection. Now it accepts only
6371 those names whose forward lookup yields the correct IP address. Any other
6372 names are discarded. This closes a loophole whereby a rogue DNS
6373 administrator could create reverse DNS records to break through a
6374 wildcarded host restriction in an ACL.
6376 37. If a user filter or a system filter that ran in a subprocess used any of
6377 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6378 the wrong values were passed to the pipe command ($thisaddress had the
6379 value of $0, $0 had the value of $1, etc). This bug was introduced by
6380 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6382 38. Improved the line breaking for long SMTP error messages from ACLs.
6383 Previously, if there was no break point between 40 and 75 characters, Exim
6384 left the rest of the message alone. Two changes have been made: (a) I've
6385 reduced the minimum length to 35 characters; (b) if it can't find a break
6386 point between 35 and 75 characters, it looks ahead and uses the first one
6387 that it finds. This may give the occasional overlong line, but at least the
6388 remaining text gets split now.
6390 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6391 file descriptors might be low, and that setting 1000 would always raise it.
6392 It turns out that in some environments, the limit is already over 1000 and
6393 that lowering it causes trouble. So now Exim takes care not to decrease it.
6395 40. When delivering a message, the value of $return_path is set to $sender_
6396 address at the start of routing (routers may change the value). By an
6397 oversight, this default was not being set up when an address was tested by
6398 -bt or -bv, which affected the outcome if any router or filter referred to
6401 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6402 log only when "message" or "log_message" are set. However, if one of the
6403 conditions was an address verification, or a call to a nested ACL, the
6404 messages generated by the underlying test were being passed through. This
6405 no longer happens. The underlying message is available in $acl_verify_
6406 message for both "message" and "log_message" expansions, so it can be
6407 passed through if needed.
6409 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6410 new expansion $bh_ to give the encoded byte string without charset
6411 translation. Translation happens only if iconv() is available; HAVE_ICONV
6412 indicates this at build time. HEADERS_CHARSET gives the charset to
6413 translate to; headers_charset can change it in the configuration, and
6414 "headers charset" can change it in an individual filter file.
6416 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6417 that creates RFC 2047 encoded "words" labels them as that charset instead
6418 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6419 expansion operator; (ii) when Exim creates a From: line for a local
6420 message; (iii) when a header line is rewritten to include a "phrase" part.
6422 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6423 buggy, causing it to skip the first lines of messages whose message ID
6424 ended in 'D'. This would not have bitten before Exim release 4.14, because
6425 message IDs were unlikely to end in 'D' before then. The effect was to have
6426 incorrect size information for certain domains.
6428 45. #include "config.h" was missing at the start of the crypt16.c module. This
6429 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6432 46. If there was a timeout during a "random" callout check, Exim treated it as
6433 a failure of the random address, and carried on sending RSET and the real
6434 address. If the delay was just some slowness somewhere, the response to the
6435 original RCPT would be taken as a response to RSET and so on, causing
6436 mayhem of various kinds.
6438 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6439 when I implemented it. It didn't allow for the fact that some option values
6440 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6441 the right test for positive values.
6443 48. Domain names in DNS records are case-independent. Exim always looks them up
6444 in lower case. Some resolvers return domain names in exactly the case they
6445 appear in the zone file, that is, they may contain uppercase letters. Not
6446 all resolvers do this - some return always lower case. Exim was treating a
6447 change of case by a resolver as a change of domain, similar to a widening
6448 of a domain abbreviation. This triggered its re-routing code and so it was
6449 trying to route what was effectively the same domain again. This normally
6450 caused routing to fail (because the router wouldn't handle the domain
6451 twice). Now Exim checks for this case specially, and just changes the
6452 casing of the domain that it ultimately uses when it transmits the message
6455 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6458 50. If a filter generated a file delivery with a non-absolute name (possible if
6459 no home directory exists for the router), the forbid_file option was not
6462 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6463 the existing equality matching.
6465 52. Exim was using ints instead of ino_t variables in some places where it was
6466 dealing with inode numbers.
6468 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6469 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6470 called TMPDIR, and if it finds it is different, it changes its value.
6472 54. The smtp_printf() function is now made available to local_scan() so
6473 additional output lines can be written before returning. There is also an
6474 smtp_fflush() function to enable the detection of a dropped connection.
6475 The variables smtp_input and smtp_batched_input are exported to
6478 55. Changed the default runtime configuration: the message "Unknown user"
6479 has been removed from the ACL, and instead placed on the localuser router,
6480 using the cannot_route_message feature. This means that any verification
6481 failures that generate their own messages won't get overridden. Similarly,
6482 the "Unrouteable address" message that was in the ACL for unverifiable
6483 relay addresses has also been removed.
6485 56. Added hosts_avoid_esmtp to the smtp transport.
6487 57. The exicyclog script was not checking for the esoteric option
6488 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6489 will work only if exicyclog is run under the appropriate euid.
6491 58. Following a discussion on the list, the rules by which Exim recognises line
6492 endings on incoming messages have been changed. The -dropcr and drop_cr
6493 options are now no-ops, retained only for backwards compatibility. The
6494 following line terminators are recognized: LF CRLF CR. However, special
6495 processing applies to CR:
6497 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6498 nor a local message in the state where . is a terminator.
6500 (ii) If a bare CR is encountered in a header line, an extra space is added
6501 after the line terminator so as not to end the header. The reasoning
6502 behind this is that bare CRs in header lines are most likely either
6503 to be mistakes, or people trying to play silly games.
6505 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6506 was being incorrectly given as 18 bytes larger than it should have been.
6507 This is a VOB (very old bug).
6509 60. This may never have affected anything current, but just in case it has:
6510 When the local host is found other than at the start of a list of hosts,
6511 the local host, those with the same MX, and any that follow, are discarded.
6512 When the list in question was part of a longer list of hosts, the following
6513 hosts (not currently being processed) were also being discarded. This no
6514 longer happens. I'm not sure if this situation could ever has previously
6517 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6520 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6521 header. This is something that is recommended in a new Internet Draft, and
6522 is something that is documented as being done by Sendmail. There are two
6523 possible values. For messages generated by the autoreply transport, Exim
6526 Auto-Submitted: auto-replied
6528 whereas for all other generated messages (e.g. bounces) it adds
6530 Auto-Submitted: auto-generated
6532 63. The "personal" condition in filters now includes a test for the
6533 Auto-Submitted: header. If it contains the string "auto-" the message it
6534 not considered personal.
6536 64. Added rcpt_include_affixes as a generic transport option.
6538 65. Added queue_only_override (default true).
6540 66. Added the syslog_duplication option.
6542 67. If what should have been the first header line of a message consisted of
6543 a space followed by a colon, Exim was mis-interpreting it as a header line.
6544 It isn't of course - it is syntactically invalid and should therefore be
6545 treated as the start of the message body. The misbehaviour could have
6546 caused a number of strange effects, including loss of data in subsequent
6547 header lines, and spool format errors.
6549 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6550 client host had authenticated. This control can now be exercised by an ACL
6551 for more flexibility.
6553 69. By default, callouts do not happen when testing with -bh. There is now a
6554 variant, -bhc, which does actually run the callout code, including
6555 consulting and updating the callout cache.
6557 70. Added support for saslauthd authentication, courtesy of Alexander
6560 71. If statvfs() failed on the spool or log directories while checking their
6561 size for availability, Exim confusingly gave the error "space shortage".
6562 Furthermore, in debugging mode it crashed with a floating point exception.
6563 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6564 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6565 serious problem, Exim now writes to the main and panic logs when this
6566 happens, with details of the failure. It then refuses to accept the
6567 incoming message, giving the message "spool directory problem" or "log
6568 directory problem" with a 421 code for SMTP messages.
6570 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6571 0, 1, and 2 exist, because some OS complain for execs without them (see
6572 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6573 descriptors. However, the code omitted to check that the open succeeded,
6574 causing mysterious errors if for some reason the permissions on /dev/null
6575 got screwed. Now Exim writes a message to the main and panic logs, and
6576 bombs out if it can't open /dev/null.
6578 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6579 interact so that it is all more flexible. It is supposed to remain
6580 backwards compatible. Also added extra_local_interfaces.
6582 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6583 to bomb out with an assertion failure - to the client this appears as a
6584 connection drop. This problem occurs in the part of the code that was taken
6585 from the Samba project. Fortunately, the assertion is in a very simple
6586 function, so I have fixed this by reproducing the function inline in the
6587 one place where it is called, and arranging for authentication to fail
6588 instead of killing the process with assert().
6590 75. The SPA client code was not working when the server requested OEM rather
6591 than Unicode encoding.
6593 76. Added code to make require_files with a specific uid setting more usable in
6594 the case where statting the file as root fails - usually a non-root-mounted
6595 NFS file system. When this happens and the failure is EACCES, Exim now
6596 forks a subprocess and does the per-uid checking as the relevant uid.
6598 77. Added process_log_path.
6600 78. If log_file_path was not explicitly set, a setting of check_log_space or
6601 check_log_inodes was ignored.
6603 79. If a space check for the spool or log partitions fails, the incident is now
6604 logged. Of course, in the latter case the data may get lost...
6606 80. Added the %p formatting code to string_format() so that it can be used to
6607 print addresses in debug_print(). Adjusted all the address printing in the
6608 debugging in store.c to use %p rather than %d.
6610 81. There was a concern that a line of code in smtp_in.c could overflow a
6611 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6612 initially expressed, the concern was not well-founded, because trailing
6613 spaces are removed early. However, if the trailing spaces were followed by
6614 a NULL, they did not get removed, so the overflow was possible. Two fixes
6617 (a) I re-wrote the offending code in a cleaner fashion.
6618 (b) If an incoming SMTP command contains a NULL character, it is rejected
6621 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6622 runs initgroups(), so that if the Exim user is in any additional groups,
6623 they will be used during message reception.
6629 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob