1 Change log file for Exim from version 4.21
2 -------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
10 supports it and a size is available (ie. the sending peer gave us one).
12 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
13 years of being deprecated). The replacements are the ACLs
14 acl_smtp_mime and acl_not_smtp_mime.
16 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
17 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
18 or even an in-clear connection were permitted. Now, if the host lookup
19 was dnssec and dane was requested then the host is only used if the
20 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
21 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
22 if one fails this test.
23 This means that a poorly-configured remote DNS will make it incommunicado;
24 but it protects against a DNS-interception attack on it.
26 JH/04 Bug 1810: make continued-use of an open smtp transport connection
27 non-noisy when a race steals the message being considered.
29 JH/05 If main configuration option tls_certificate is unset, generate a
30 self-signed certificate for inbound TLS connections.
32 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
33 in rewrites and routers.
35 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
36 and logged a warning sing 4.83; now they are a configuration file error.
38 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
39 (lacking @domain). Apply the same qualification processing as RCPT.
41 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
43 JH/10 Support ${sha256:} applied to a string (as well as the previous
46 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
47 a cutthrough deliver is pending, as we always want to make a connection.
48 This also avoids re-routing the message when later placing the cutthrough
49 connection after a verify cache hit.
50 Do not update it with the verify result either.
52 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
53 when routing results in more than one destination address.
55 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
56 signing (which inhibits the cutthrough capability). Previously only
57 the presence of an option was tested; now an expansion evaluating as
58 empty is permissible (obviously it should depend only on data available
59 when the cutthrough connection is made).
61 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
62 the relevant preceding SMTP command did not note the pipelining mode.
64 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
65 Previously they were not counted.
67 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
68 as one having no matching records. Previously we deferred the message
69 that needed the lookup.
71 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
72 distinguished as "(=".
74 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
75 for missing MX records. Previously it only worked for missing A records.
77 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
79 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
80 after the data-go-ahead and data-ack. Patch from Jason Betts.
82 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
83 even for a "none" policy. Patch from Tony Meyer.
85 JH/22 Fix continued use of a connection for further deliveries. If a port was
86 specified by a router, it must also match for the delivery to be
89 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
90 When one of the recipients of a message was unsuitable for the connection
91 (has no matching addresses), we lost track of needing to mark it
92 deferred. As a result mail would be lost.
94 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
96 JH/25 Decoding ACL controls is now done using a binary search; the source code
97 takes up less space and should be simpler to maintain. Merge the ACL
98 condition decode tables also, with similar effect.
100 JH/26 Fix problem with one_time used on a redirect router which returned the
101 parent address unchanged. A retry would see the parent address marked as
102 delivered, so not attempt the (identical) child. As a result mail would
105 JH/27 Fix a possible security hole, wherein a process operating with the Exim
106 UID can gain a root shell. Credit to http://www.halfdog.net/ for
107 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
110 JH/28 Enable {spool,log} filesystem space and inode checks as default.
111 Main config options check_{log,spool}_{inodes,space} are now
112 100 inodes, 10MB unless set otherwise in the configuration.
114 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
115 Previously it only applied to the main-section connection policy
118 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
120 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
121 by me. Added RFC7919 DH primes as an alternative.
123 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
124 are not in the system include path.
126 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
127 GnuTLS, when a session startup failed (eg because the client disconnected)
128 Exim did stdio operations after fclose. This was exposed by a recent
129 change which nulled out the file handle after the fclose.
131 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
132 signed directly by the cert-signing cert, rather than an intermediate
133 OCSP-signing cert. This is the model used by LetsEncrypt.
135 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
137 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
138 an incoming connection.
140 HS/02 Bug 1802: Do not half-close the connection after sending a request
143 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
144 fallback to "prime256v1".
148 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
149 and 3.4.4 - once the server is enabled to respond to an OCSP request
150 it does even when not requested, resulting in a stapling non-aware
151 client dropping the TLS connection.
153 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
154 support variable-length bit vectors. No functional change.
156 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
157 The I= interface field on outgoing lines is now after the H= remote
158 host field, same as incoming lines. There is a separate
159 outgoing_interface log selector which allows you to disable the
162 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
163 If not running log_selector +smtp_connection the mainlog would be held
164 open indefinitely after a "too many connections" event, including to a
165 deleted file after a log rotate. Leave the per net connection logging
166 leaving it open for efficiency as that will be quickly detected by the
167 check on the next write.
169 HS/01 Bug 1671: Fix post transport crash.
170 Processing the wait-<transport> messages could crash the delivery
171 process if the message IDs didn't exist for some reason. When
172 using 'split_spool_directory=yes' the construction of the spool
173 file name failed already, exposing the same netto behaviour.
175 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
176 mime_regex ACL conditions.
178 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
179 to DSN fail messages (bounces): remote IP, remote greeting, remote response
180 to HELO, local diagnostic string.
182 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
183 log line to debug. Even when configured with a tls authenticator many
184 client connections are expected to not authenticate in this way, so
185 an authenticate fail is not an error.
187 HS/02 Add the Exim version string to the process info. This way exiwhat
188 gives some more detail about the running daemon.
190 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
191 matter for fast-change records such as DNSBLs.
193 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
194 retry record, even if constant. There may be multiple transports with
195 different interface settings and the retry behaviour needs to be kept
198 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
200 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
202 JH/10 Bug 840: fix log_defer_output option of pipe transport
204 JH/11 Bug 830: use same host for all RCPTS of a message, even under
205 hosts_randomize. This matters a lot when combined with mua_wrapper.
207 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
208 ${quote_pgsql:<string>} operator.
210 JH/13 Bug 1708: avoid misaligned access in cached lookup.
212 JH/14 Change header file name for freeradius-client. Relevant if compiling
213 with Radius support; from the Gentoo tree and checked under Fedora.
215 JH/15 Bug 1712: Introduce $prdr_requested flag variable
217 JH/16 Bug 1714: Permit an empty string as expansion result for transport
218 option transport_filter, meaning no filtering.
220 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
222 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
223 defaults to "*" (all hosts). The variable is now available when not built
224 with TLS, default unset, mainly to enable keeping the testuite sane.
225 If a server certificate is not supplied (via tls_certificate) an error is
226 logged, and clients will find TLS connections fail on startup. Presumably
227 they will retry in-clear.
228 Packagers of Exim are strongly encouraged to create a server certificate
229 at installation time.
231 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
232 with the $config_file variable.
234 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
235 in transport context, after the attempt, and per-recipient. The latter type
236 is per host attempted. The event data is the error message, and the errno
237 information encodes the lookup type (A vs. MX) used for the (first) host,
238 and the trailing two digits of the smtp 4xx reponse.
240 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
241 to write to mainlog (or rejectlog, paniclog) in the window between file
242 creation and permissions/ownership being changed. Particularly affects
243 installations where exicyclog is run as root, rather than exim user;
244 result is that the running daemon panics and dies.
246 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
248 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
249 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
250 "pri" and "weight". Note that the previous implicit priority given by the
251 list order is no longer honoured.
253 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
256 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
257 by defining SUPPORT_SOCKS.
259 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
260 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
261 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
262 variables are renamed to proxy_{local,external}_{address,port}.
264 JH/27 Move Internationalisation support from Experimental to mainline, enabled
265 for a build by defining SUPPORT_I18N
267 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
268 of the query string, and make ${quote_redis:} do that quoting.
270 JH/29 Move Events support from Experimental to mainline, enabled by default
271 and removable for a build by defining DISABLE_EVENT.
273 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
275 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
276 cached by the daemon.
278 JH/32 Move Redis support from Experimental to mainline, enabled for a build
279 by defining LOOKUP_REDIS. The libhiredis library is required.
281 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
282 keys are given for lookup.
284 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
285 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
286 only supported when built with TLS support. The PolarSSL SHA routines
287 are still used when the TLS library is too old for convenient support.
289 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
290 openssl_options), for security. OpenSSL forces this from version 1.1.0
291 server-side so match that on older versions.
293 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
294 allocation for $value could be released as the expansion processing
295 concluded, but leaving the global pointer active for it.
297 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
298 and to use the domains and local_parts ACL conditions.
300 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
301 incorrectly not doubled on cutthrough transmission, hence seen as a
302 body-termination at the receiving system - resulting in truncated mails.
303 Commonly the sender saw a TCP-level error, and retransmitted the message
304 via the normal store-and-forward channel. This could result in duplicates
305 received - but deduplicating mailstores were liable to retain only the
306 initial truncated version.
308 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
310 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
312 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
313 we're in there, support oversigning also; bug 1309.
315 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
317 HS/04 Add support for keep_environment and add_environment options.
319 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
320 either intentional arithmetic overflow during PRNG, or testing config-
323 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
324 delivery resulted in actual delivery. Cancel cutthrough before DATA
327 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
328 rejects a recipient: pass the reject to the originator.
330 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
331 Many were false-positives and ignorable, but it's worth fixing the
334 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
335 for the new environment-manipulation done at startup. Move the routines
336 from being local to tls.c to being global via the os.c file.
338 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
339 an extract embedded as result-arg for a map, the first arg for extract
340 is unavailable so we cannot tell if this is a numbered or keyed
341 extraction. Accept either.
347 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
350 JH/02 The smtp transport option "multi_domain" is now expanded.
352 JH/03 The smtp transport now requests PRDR by default, if the server offers
355 JH/04 Certificate name checking on server certificates, when exim is a client,
356 is now done by default. The transport option tls_verify_cert_hostnames
357 can be used to disable this per-host. The build option
358 EXPERIMENTAL_CERTNAMES is withdrawn.
360 JH/05 The value of the tls_verify_certificates smtp transport and main options
361 default to the word "system" to access the system default CA bundle.
362 For GnuTLS, only version 3.0.20 or later.
364 JH/06 Verification of the server certificate for a TLS connection is now tried
365 (but not required) by default. The verification status is now logged by
366 default, for both outbound TLS and client-certificate supplying inbound
369 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
372 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
373 Status Notification (bounce) messages are now MIME format per RFC 3464.
374 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
375 under the control of the dsn_advertise_hosts option, and routers may
376 have a dsn_lasthop option.
378 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
379 default, modifiable by a malware= option. The list separator for
380 the options can now be changed in the usual way. Bug 68.
382 JH/10 The smtp_receive_timeout main option is now expanded before use.
384 JH/11 The incoming_interface log option now also enables logging of the
385 local interface on delivery outgoing connections.
387 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
388 if the interface and destination host and port all match.
390 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
393 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
394 Patch from Andrew Lewis.
396 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
397 now supports optional time-restrictions, weighting, and priority
398 modifiers per server. Patch originally by <rommer@active.by>.
400 JH/16 The spamd_address main option now supports a mixed list of local
401 and remote servers. Remote servers can be IPv6 addresses, and
402 specify a port-range.
404 JH/17 Bug 68: The spamd_address main option now supports an optional
405 timeout value per server.
407 JH/18 Bug 1581: Router and transport options headers_add/remove can
408 now have the list separator specified.
410 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
413 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
416 JH/21 Support for the A6 type of dns record is withdrawn.
418 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
419 rather than the verbs used.
421 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
422 from 255 to 1024 chars.
424 JH/24 Verification callouts now attempt to use TLS by default.
426 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
427 are generic router options now. The defaults didn't change.
429 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
430 Original patch from Alexander Shikoff, worked over by JH.
432 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
433 files if ZCAT_COMMAND is not executable.
435 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
437 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
439 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
440 Normally benign, it bites when the pair was led to by a CNAME;
441 modern usage is to not canoicalize the domain to a CNAME target
442 (and we were inconsistent anyway for A-only vs AAAA+A).
444 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
446 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
447 when evaluating $sender_host_dnssec.
449 JH/31 Check the HELO verification lookup for DNSSEC, adding new
450 $sender_helo_dnssec variable.
452 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
454 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
456 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
458 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
459 documented as working, but never had. Support all but $spam_report.
461 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
462 added for tls authenticator.
464 HS/03 Add perl_taintmode main config option
469 TL/01 When running the test suite, the README says that variables such as
470 no_msglog_check are global and can be placed anywhere in a specific
471 test's script, however it was observed that placement needed to be near
472 the beginning for it to behave that way. Changed the runtest perl
473 script to read through the entire script once to detect and set these
474 variables, reset to the beginning of the script, and then run through
475 the script parsing/test process like normal.
477 TL/02 The BSD's have an arc4random API. One of the functions to induce
478 adding randomness was arc4random_stir(), but it has been removed in
479 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
480 function when detected.
482 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
483 cause callback expansion.
485 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
486 syntax errors in an expansion can be treated as a string instead of
487 logging or causing an error, due to the internal use of bool_lax
488 instead of bool when processing it.
490 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
491 server certificates when making smtp deliveries.
493 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
495 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
497 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
499 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
500 Merged patch from Sebastian Wiedenroth.
502 JH/05 Fix results-pipe from transport process. Several recipients, combined
503 with certificate use, exposed issues where response data items split
504 over buffer boundaries were not parsed properly. This eventually
505 resulted in duplicates being sent. This issue only became common enough
506 to notice due to the introduction of conection certificate information,
507 the item size being so much larger. Found and fixed by Wolfgang Breyha.
509 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
510 size buffer was used, resulting in syntax errors when an expansion
513 JH/07 Add support for directories of certificates when compiled with a GnuTLS
514 version 3.3.6 or later.
516 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
517 is EXPERIMENTAL_EVENT, the main-configuration and transport options
518 both become "event_action", the variables become $event_name, $event_data
519 and $event_defer_errno. There is a new variable $verify_mode, usable in
520 routers, transports and related events. The tls:cert event is now also
521 raised for inbound connections, if the main configuration event_action
524 TL/06 In test suite, disable OCSP for old versions of openssl which contained
525 early OCSP support, but no stapling (appears to be less than 1.0.0).
527 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
528 server certificate names available under the smtp transport option
529 "tls_verify_cert_hostname" now do not permit multi-component wildcard
532 JH/10 Time-related extraction expansions from certificates now use the main
533 option "timezone" setting for output formatting, and are consistent
534 between OpenSSL and GnuTLS compilations. Bug 1541.
536 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
537 encoded parameter in the incoming message. Bug 1558.
539 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
540 include certificate info, eximon was claiming there were spoolfile
543 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
545 JH/14 Log delivery-related information more consistently, using the sequence
546 "H=<name> [<ip>]" wherever possible.
548 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
549 are problematic for Debian distribution, omit them from the release
552 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
554 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
557 JH/17 Fix a null-indirection in certextract expansions when a nondefault
558 output list separator was used.
563 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
564 checkers that were complaining about end of non-void function with no
567 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
568 This was a regression intruduced in 4.83 by another bugfix.
570 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
572 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
573 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
579 TF/01 Correctly close the server side of TLS when forking for delivery.
581 When a message was received over SMTP with TLS, Exim failed to clear up
582 the incoming connection properly after forking off the child process to
583 deliver the message. In some situations the subsequent outgoing
584 delivery connection happened to have the same fd number as the incoming
585 connection previously had. Exim would try to use TLS and fail, logging
586 a "Bad file descriptor" error.
588 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
589 utilities have not been installed.
591 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
592 temporary space as the ACL may create new global variables.
594 TL/01 LDAP support uses per connection or global context settings, depending
595 upon the detected version of the libraries at build time.
597 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
598 to extract and use the src ip:port in logging and expansions as if it
599 were a direct connection from the outside internet. PPv2 support was
600 updated based on HAProxy spec change in May 2014.
602 JH/02 Add ${listextract {number}{list}{success}{fail}}.
604 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
605 Properly escape header and check for NULL return.
607 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
610 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
612 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
613 characters in header names, implemented as a verify condition.
614 Contributed by Michael Fischer v. Mollard.
616 TL/05 Rename SPF condition results err_perm and err_temp to standardized
617 results permerror and temperror. Previous values are deprecated but
618 still accepted. In a future release, err_perm and err_temp will be
619 completely removed, which will be a backward incompatibility if the
620 ACL tests for either of these two old results. Patch contributed by
621 user bes-internal on the mailing list.
623 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
625 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
626 selectors, in both main and reject logs.
628 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
631 JH/07 Add malware type "sock" for talking to simple daemon.
633 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
635 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
636 routers/transports under cutthrough routing.
638 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
639 numbers. Touch up "bool" conditional to keep the same definition.
641 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
643 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
645 JH/12 Expand items in router/transport headers_add or headers_remove lists
646 individually rather than the list as a whole. Bug 1452.
648 Required for reasonable handling of multiple headers_ options when
649 they may be empty; requires that headers_remove items with embedded
650 colons must have them doubled (or the list-separator changed).
652 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
653 view the policy declared in the DMARC record. Currently, $dmarc_status
654 is a combined value of both the record presence and the result of the
657 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
659 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
660 dnslookup router and the smtp transport (applying to the forward
663 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
664 of ldap servers used for a specific lookup. Patch provided by Heiko
667 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
668 New variable $lookup_dnssec_authenticated for observability.
670 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
671 Patch submitted by Lars Timman.
673 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
675 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
676 Requires trusted mode and valid format message id, aborts otherwise.
677 Patch contributed by Heiko Schlichting.
679 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
680 certextract with support for various fields. Bug 1358.
682 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
683 is requested by default, modifiable by smtp transport option
686 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
687 operate on certificate variables to give certificate fingerprints
688 Also new ${sha256:cert_variable}.
690 JH/23 The PRDR feature is moved from being Experimental into the mainline.
692 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
693 Christian Aistleitner.
695 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
697 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
698 file. Patch from Wolfgang Breyha.
700 JH/25 Expand the coverage of the delivery $host and $host_address to
701 client authenticators run in verify callout. Bug 1476.
703 JH/26 Port service names are now accepted for tls_on_connect_ports, to
704 align with daemon_smtp_ports. Bug 72.
706 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
707 support and error reporting did not work properly.
709 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
710 and is readable. Patch from Andrew Colin Kissa.
712 TL/14 Enhance documentation of ${run expansion and how it parses the
713 commandline after expansion, particularly in the case when an
714 unquoted variable expansion results in an empty value.
716 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
718 PP/02 Fix internal collision of T_APL on systems which support RFC3123
719 by renaming away from it. Addresses GH issue 15, reported by
722 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
724 TL/15 SECURITY: prevent double expansion in math comparison functions
725 (can expand unsanitized data). Not remotely exploitable.
732 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
734 PP/02 Make -n do something, by making it not do something.
735 When combined with -bP, the name of an option is not output.
737 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
740 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
741 $sender_host_name and config options to manage this, and basic check
744 PP/05 DSCP support for outbound connections and control modifier for inbound.
746 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
747 (Only plugin which currently uses this is kerberos4, which nobody should
748 be using, but we should make it available and other future plugins might
749 conceivably use it, even though it would break NAT; stuff *should* be
750 using channel bindings instead).
752 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
753 name; added for Sendmail compatibility; requires admin caller.
754 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
755 just ignore it); requires trusted caller.
756 Also parse but ignore: -Ac -Am -X<logfile>
759 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
761 TL/02 Add +smtp_confirmation as a default logging option.
763 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
764 Patch by Magnus Holmgren from 2007-02-20.
766 TL/04 Bugzilla 1281 - Spec typo.
767 Bugzilla 1283 - Spec typo.
768 Bugzilla 1290 - Spec grammar fixes.
770 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
772 TL/06 Add Experimental DMARC support using libopendmarc libraries.
774 TL/07 Fix an out of order global option causing a segfault. Reported to dev
775 mailing list by by Dmitry Isaikin.
777 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
779 JH/02 Support "G" suffix to numbers in ${if comparisons.
781 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
783 NM/01 Bugzilla 1197 - Spec typo
784 Bugzilla 1196 - Spec examples corrections
786 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
788 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
789 gnutls_enable_pkcs11, but renamed to more accurately indicate its
792 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
793 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
795 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
796 "acl {{name}{arg}...}", and optional args on acl condition
799 JH/05 Permit multiple router/transport headers_add/remove lines.
801 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
803 JH/07 Avoid using a waiting database for a single-message-only transport.
804 Performance patch from Paul Fisher. Bugzilla 1262.
806 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
809 JH/09 Add $headers_added variable, with content from use of ACL modifier
810 add_header (but not yet added to the message). Bugzilla 199.
812 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
813 Pulled from Bugzilla 817 by Wolfgang Breyha.
815 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
817 (nb: this is the same fix as in Exim 4.80.1)
819 JH/11 Add A= logging on delivery lines, and a client_set_id option on
822 JH/12 Add optional authenticated_sender logging to A= and a log_selector
825 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
827 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
828 advertise SMTP AUTH mechanism to us, instead of a generic
829 protocol violation error. Also, make Exim more robust to bad
830 data from the Dovecot auth socket.
832 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
834 When a queue runner is handling a message, Exim first routes the
835 recipient addresses, during which it prunes them based on the retry
836 hints database. After that it attempts to deliver the message to
837 any remaining recipients. It then updates the hints database using
840 So if a recipient address works intermittently, it can get repeatedly
841 deferred at routing time. The retry hints record remains fresh so the
842 address never reaches the final cutoff time.
844 This is a fairly common occurrence when a user is bumping up against
845 their storage quota. Exim had some logic in its local delivery code
846 to deal with this. However it did not apply to per-recipient defers
847 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
849 This change adds a proper retry rule check during routing so that the
850 final cutoff time is checked against the message's age. We only do
851 this check if there is an address retry record and there is not a
852 domain retry record; this implies that previous attempts to handle
853 the address had the retry_use_local_parts option turned on. We use
854 this as an approximation for the destination being like a local
855 delivery, as in LMTP.
857 I suspect this new check makes the old local delivery cutoff check
858 redundant, but I have not verified this so I left the code in place.
860 TF/02 Correct gecos expansion when From: is a prefix of the username.
862 Test 0254 submits a message to Exim with the header
866 When I ran the test suite under the user fanf2, Exim expanded
867 the header to contain my full name, whereas it should have added
868 a Resent-Sender: header. It erroneously treats any prefix of the
869 username as equal to the username.
871 This change corrects that bug.
873 GF/01 DCC debug and logging tidyup
874 Error conditions log to paniclog rather than rejectlog.
875 Debug lines prefixed by "DCC: " to remove any ambiguity.
877 TF/03 Avoid unnecessary rebuilds of lookup-related code.
879 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
880 Bug spotted by Jeremy Harris; was flawed since initial commit.
881 Would have resulted in OCSP responses post-SNI triggering an Exim
882 NULL dereference and crash.
884 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
886 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
887 Bug detection, analysis and fix by Samuel Thibault.
888 Bugzilla 1331, Debian bug #698092.
890 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
892 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
893 Server implementation by Todd Lyons, client by JH.
894 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
895 config variable "prdr_enable" controls whether the server
896 advertises the facility. If the client requests PRDR a new
897 acl_data_smtp_prdr ACL is called once for each recipient, after
898 the body content is received and before the acl_smtp_data ACL.
899 The client is controlled by bolth of: a hosts_try_prdr option
900 on the smtp transport, and the server advertisement.
901 Default client logging of deliveries and rejections involving
902 PRDR are flagged with the string "PRDR".
904 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
905 fclose(). Diagnosis by Todd Lyons.
907 PP/17 Update configure.default to handle IPv6 localhost better.
908 Patch by Alain Williams (plus minor tweaks).
911 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
912 This is now consistent with GnuTLS, and is now documented: the
913 previous undocumented portable approach to treating the option as
914 unset was to force an expansion failure. That still works, and
915 an empty string is now equivalent.
917 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
918 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
919 not performing validation itself.
921 PP/20 Added force_command boolean option to pipe transport.
922 Patch from Nick Koston, of cPanel Inc.
924 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
927 TF/04 Added udpsend ACL modifer and hexquote expansion operator
929 PP/21 Fix eximon continuous updating with timestamped log-files.
930 Broken in a format-string cleanup in 4.80, missed when I repaired the
931 other false fix of the same issue.
932 Report and fix from Heiko Schlichting.
935 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
936 Report from Prashanth Katuri.
938 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
939 It's SecureTransport, so affects any MacOS clients which use the
940 system-integrated TLS libraries, including email clients.
942 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
943 using a MIME ACL for non-SMTP local injection.
944 Report and assistance in diagnosis by Warren Baker.
946 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
948 JH/16 Fix comparisons for 64b. Bugzilla 1385.
950 TL/09 Add expansion variable $authenticated_fail_id to keep track of
951 last id that failed so it may be referenced in subsequent ACL's.
953 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
956 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
957 ldap library initialization, allowing self-signed CA's to be
958 used. Also properly sets require_cert option later in code by
959 using NULL (global ldap config) instead of ldap handle (per
960 session). Bug diagnosis and testing by alxgomz.
962 TL/12 Enhanced documentation in the ratelimit.pl script provided in
963 the src/util/ subdirectory.
965 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
966 renamed to Transport Post Delivery Action by Jeremy Harris, as
969 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
970 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
971 redis_servers = needs to be configured which will be used by the redis
972 lookup. Patch from Warren Baker, of The Packet Hub.
974 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
976 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
977 hostname or reverse DNS when processing a host list. Used suggestions
978 from multiple comments on this bug.
980 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
982 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
983 Missed a few lines, added it to make the runtest require no keyboard
986 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
987 contains upper case chars. Make router use caseful_local_part.
989 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
990 support when GnuTLS has been built with p11-kit.
996 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
998 This, or similar/improved, will also be change PP/11 of 4.82.
1004 PP/01 Handle short writes when writing local log-files.
1005 In practice, only affects FreeBSD (8 onwards).
1006 Bugzilla 1053, with thanks to Dmitry Isaikin.
1008 NM/01 Bugzilla 949 - Documentation tweak
1010 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1013 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1015 PP/02 Implemented gsasl authenticator.
1017 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1019 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1020 `pkg-config foo` for cflags/libs.
1022 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1023 with rest of GSASL and with heimdal_gssapi.
1025 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1026 `pkg-config foo` for cflags/libs for the TLS implementation.
1028 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1029 properties get this fed in as external SSF. A number of robustness
1030 and debugging improvements to the cyrus_sasl authenticator.
1032 PP/08 cyrus_sasl server now expands the server_realm option.
1034 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1035 Patch by Jeremy Harris.
1037 PP/10 Added dbmjz lookup type.
1039 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1041 PP/12 MAIL args handles TAB as well as SP, for better interop with
1042 non-compliant senders.
1043 Analysis and variant patch by Todd Lyons.
1045 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1046 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1047 Patch from Dirk Mueller <dmueller@suse.com>
1049 PP/13 tls_peerdn now print-escaped for spool files.
1050 Observed some $tls_peerdn in wild which contained \n, which resulted
1051 in spool file corruption.
1053 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1054 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1055 or write after TLS renegotiation, which otherwise led to messages
1058 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1059 as a tracking header (ie: a signed header comes before the signature).
1060 Patch from Wolfgang Breyha.
1062 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1063 comma-sep list; embedded commas doubled.
1065 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1067 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1069 Report and patch from Dmitry Banschikov.
1071 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1072 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1073 failures. We appear to now support TLS1.1+ with Exim.
1075 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1076 lets Exim select keys and certificates based upon TLS SNI from client.
1077 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1078 before an outbound SMTP session. New log_selector, +tls_sni.
1080 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1081 NULL dereference. Report and patch from Alun Jones.
1083 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1084 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1085 Not seeing resolver debug output on NetBSD, but suspect this is a
1086 resolver implementation change.
1088 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1089 Left warnings. Added "eximon gdb" invocation mode.
1091 PP/21 Defaulting "accept_8bitmime" to true, not false.
1093 PP/22 Added -bw for inetd wait mode support.
1095 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1096 locate the relevant includes and libraries. Made this the default.
1098 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1099 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1101 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1102 This may cause build issues on older platforms.
1104 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1105 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1106 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1107 Added SNI support via GnuTLS too.
1108 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1110 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1112 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1113 Applied second patch from Janne, implementing suggestion to default
1114 multiple-strings-in-record handling to match SPF spec.
1116 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1118 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1119 read-only, out of scope).
1120 Patch from Wolfgang Breyha, report from Stuart Northfield.
1122 PP/29 Fix three issues highlighted by clang analyser static analysis.
1123 Only crash-plausible issue would require the Cambridge-specific
1124 iplookup router and a misconfiguration.
1125 Report from Marcin Mirosław.
1127 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1129 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1130 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1131 As part of this, removing so much warning spew let me fix some minor
1132 real issues in debug logging.
1134 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1135 assignment on my part. Fixed.
1137 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1138 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1139 Janne Snabb (who went above and beyond: thank you).
1141 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1142 string otherwise requires a connection and a bunch more work and it's
1143 relatively easy to get wrong. Should also expose TLS library linkage
1146 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1147 64-bit ${eval} (JH/03).
1149 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1150 GNU libc to support some of the 64-bit stuff, should not lead to
1151 conflicts. Defined before os.h is pulled in, so if a given platform
1152 needs to override this, it can.
1154 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1155 protection layer was required, which is not implemented.
1156 Bugzilla 1254, patch from Wolfgang Breyha.
1158 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1159 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1160 tls_dhparam take prime identifiers. Also unbreak combination of
1161 OpenSSL+DH_params+TLSSNI.
1163 PP/39 Disable SSLv2 by default in OpenSSL support.
1169 PP/01 Solaris build fix for Oracle's LDAP libraries.
1170 Bugzilla 1109, patch from Stephen Usher.
1172 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1174 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1177 TF/02 Fix a couple more cases where we did not log the error message
1178 when unlink() failed. See also change 4.74-TF/03.
1180 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1181 lock up or crash if it happened to be inside a call to libc when it
1182 got a SIGUSR1 from exiwhat.
1184 The SIGUSR1 handler appends the current process status to the process
1185 log which is later printed by exiwhat. It used to use the general
1186 purpose logging code to do this, but several functions it calls are
1187 not safe for signals.
1189 The new output code in the SIGUSR1 handler is specific to the process
1190 log, and simple enough that it's easy to inspect for signal safety.
1191 Removing some special cases also simplifies the general logging code.
1192 Removing the spurious timestamps from the process log simplifies
1195 TF/04 Improved ratelimit ACL condition.
1197 The /noupdate option has been deprecated in favour of /readonly which
1198 has clearer semantics. The /leaky, /strict, and /readonly update modes
1199 are mutually exclusive. The update mode is no longer included in the
1200 database key; it just determines when the database is updated. (This
1201 means that when you upgrde Exim will forget old rate measurements.)
1203 Exim now checks that the per_* options are used with an update mode that
1204 makes sense for the current ACL. For example, when Exim is processing a
1205 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1206 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1207 must specify per_mail/readonly. If you omit the update mode it defaults to
1208 /leaky where that makes sense (as before) or /readonly where required.
1210 The /noupdate option is now undocumented but still supported for
1211 backwards compatibility. It is equivalent to /readonly except that in
1212 ACLs where /readonly is required you may specify /leaky/noupdate or
1213 /strict/noupdate which are treated the same as /readonly.
1215 A useful new feature is the /count= option. This is a generalization
1216 of the per_byte option, so that you can measure the throughput of other
1217 aggregate values. For example, the per_byte option is now equivalent
1218 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1220 The per_rcpt option has been generalized using the /count= mechanism
1221 (though it's more complicated than the per_byte equivalence). When it is
1222 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1223 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1224 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1225 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1226 Note that using per_rcpt with a non-readonly update mode in more than
1227 one ACL will cause the recipients to be double-counted. (The per_mail
1228 and per_byte options don't have this problem.)
1230 The handling of very low rates has changed slightly. If the computed rate
1231 is less than the event's count (usually one) then this event is the first
1232 after a long gap. In this case the rate is set to the same as this event's
1233 count, so that the first message of a spam run is counted properly.
1235 The major new feature is a mechanism for counting the rate of unique
1236 events. The new per_addr option counts the number of different
1237 recipients that someone has sent messages to in the last time period. It
1238 behaves like per_rcpt if all the recipient addresses are different, but
1239 duplicate recipient addresses do not increase the measured rate. Like
1240 the /count= option this is a general mechanism, so the per_addr option
1241 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1242 example, measure the rate that a client uses different sender addresses
1243 with the options per_mail/unique=$sender_address. There are further
1244 details in the main documentation.
1246 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1248 TF/06 Removed a few PCRE remnants.
1250 TF/07 Automatically extract Exim's version number from tags in the git
1251 repository when doing development or release builds.
1253 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1254 Bugzilla 879. Patch from Paul Fisher.
1256 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1257 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1260 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1262 PP/05 Variable $av_failed, true if the AV scanner deferred.
1263 Bugzilla 1078. Patch from John Horne.
1265 PP/06 Stop make process more reliably on build failure.
1266 Bugzilla 1087. Patch from Heiko Schlittermann.
1268 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1269 Bugzilla 1089. Patch from Heiko Schlittermann.
1271 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1272 Bugzilla 1131. Patch from Holger Weiß.
1274 PP/09 Handle IPv6 addresses with SPF.
1275 Bugzilla 860. Patch from Wolfgang Breyha.
1277 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1279 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1282 PP/11 match_* no longer expand right-hand-side by default.
1283 New compile-time build option, EXPAND_LISTMATCH_RHS.
1284 New expansion conditions, "inlist", "inlisti".
1286 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1288 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1290 PP/14 fix log_write() format string regression from TF/03.
1291 Bugzilla 1152. Patch from Dmitry Isaikin.
1297 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1299 PP/02 Harmonised TLS library version reporting; only show if debugging.
1300 Layout now matches that introduced for other libraries in 4.74 PP/03.
1302 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1304 PP/04 New "dns_use_edns0" global option.
1306 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1309 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1310 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1312 TK/01 Updated PolarSSL code to 0.14.2.
1313 Bugzilla 1097. Patch from Andreas Metzler.
1315 PP/07 Catch divide-by-zero in ${eval:...}.
1316 Fixes bugzilla 1102.
1318 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1321 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1322 format-string attack -- SECURITY: remote arbitrary code execution.
1324 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1325 time unintentionally subject to list matching rules, letting the header
1326 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1327 arbitrary string expansion). This allowed for information disclosure.
1329 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1330 INT_MIN/-1 -- value coerced to INT_MAX.
1336 NM/01 Workround for PCRE version dependency in version reporting
1339 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1340 This fixes portability to compilers other than gcc, notably
1341 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1343 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1344 makefiles for portability to HP-UX and POSIX correctness.
1346 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1347 Also via indented variable definition in the Makefile.
1348 (Debugging by Oliver Heesakkers).
1350 PP/02 Restore caching of spamd results with expanded spamd_address.
1351 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1353 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1354 Improves build reliability. Fix from: Frank Elsner
1356 NM/02 Fix wide character breakage in the rfc2047 coding
1357 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1359 NM/03 Allow underscore in dnslist lookups
1360 Fixes bug 1026. Patch from Graeme Fowler
1362 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1363 Code patches from Adam Ciarcinski of NetBSD.
1365 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1368 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1369 is logged, to avoid truncation. Patch from John Horne.
1371 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1372 Patch from Jakob Hirsch.
1374 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1375 SQL string expansion failure details.
1376 Patch from Andrey Oktyabrski.
1378 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1379 Patch from Simon Arlott.
1381 PP/09 New lookups functionality failed to compile on old gcc which rejects
1382 extern declarations in function scope.
1383 Patch from Oliver Fleischmann
1385 PP/10 Use sig_atomic_t for flags set from signal handlers.
1386 Check getgroups() return and improve debugging.
1387 Fixed developed for diagnosis in bug 927 (which turned out to be
1390 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1391 Patch from Mark Zealey.
1393 PP/12 Bugzilla 1056: Improved spamd server selection.
1394 Patch from Mark Zealey.
1396 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1397 Based on patch from Heiko Schlittermann.
1399 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1400 Patch from Uwe Doering, sign-off by Michael Haardt.
1402 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1403 variable declaration deep within a block. Bug and patch from
1406 PP/15 lookups-Makefile IRIX compatibilty coercion.
1408 PP/16 Make DISABLE_DKIM build knob functional.
1410 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1411 Patch by Simon Arlott
1413 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1414 variable argument macros. Our copy now differs from upstream.
1420 TF/01 Failure to get a lock on a hints database can have serious
1421 consequences so log it to the panic log.
1423 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1424 controlled using the smtp_confirmation log selector.
1426 TF/03 Include the error message when we fail to unlink a spool file.
1428 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1429 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1430 for maintaining out-of-tree patches for some time.
1432 PP/01 Bugzilla 139: Documentation and portability issues.
1433 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1434 Handle per-OS dynamic-module compilation flags.
1436 PP/02 Let /dev/null have normal permissions.
1437 The 4.73 fixes were a little too stringent and complained about the
1438 permissions on /dev/null. Exempt it from some checks.
1439 Reported by Andreas M. Kirchwitz.
1441 PP/03 Report version information for many libraries, including
1442 Exim version information for dynamically loaded libraries. Created
1443 version.h, now support a version extension string for distributors
1444 who patch heavily. Dynamic module ABI change.
1446 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1447 privilege escalation vulnerability whereby the Exim run-time user
1448 can cause root to append content of the attacker's choosing to
1451 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1454 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1455 If dropping privileges for untrusted macros, we disabled normal logging
1456 on the basis that it would fail; for the Exim run-time user, this is not
1457 the case, and it resulted in successful deliveries going unlogged.
1458 Fixed. Reported by Andreas Metzler.
1464 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1465 only prepend for the Resent-* case. Fixes regression introduced in
1466 Exim 4.70 by NM/22 for Bugzilla 607.
1468 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1469 increasing numbers of administrators be bitten by this.
1471 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1473 PP/03 Bugzilla 994: added openssl_options main configuration option.
1475 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1477 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1479 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1481 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1482 without a peer certificate, leading to a segfault because of an
1483 assumption that peers always have certificates. Be a little more
1484 paranoid. Problem reported by Martin Tscholak.
1486 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1487 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1488 NB: ClamAV planning to remove STREAM in "middle of 2010".
1489 CL also introduces -bmalware, various -d+acl logging additions and
1490 more caution in buffer sizes.
1492 PP/09 Implemented reverse_ip expansion operator.
1494 PP/10 Bugzilla 937: provide a "debug" ACL control.
1496 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1498 PP/12 Bugzilla 973: Implement --version.
1500 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1502 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1504 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1506 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1507 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1508 ignore trailing whitespace.
1510 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1512 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1515 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1516 Notification from Dr Andrew Aitchison.
1518 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1519 ExtendedDetectionInfo response format.
1520 Notification from John Horne.
1522 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1525 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1526 XSL and documented dependency on system catalogs, with examples of how
1529 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1532 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1533 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1534 configuration file which is writeable by the Exim user or group.
1536 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1537 of configuration files to cover files specified with the -C option if
1538 they are going to be used with root privileges, not just the default
1541 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1542 option (effectively making it always true).
1544 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1545 files to be used while preserving root privileges.
1547 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1548 that rogue child processes cannot use them.
1550 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1551 run-time user, instead of root.
1553 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1554 Exim run-time user without dropping privileges.
1556 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1557 result string, instead of calling string_vformat() twice with the same
1560 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1561 for other users. Others should always drop root privileges if they use
1562 -C on the command line, even for a whitelisted configure file.
1564 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1566 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1572 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1573 $data_path, and $header_path variables; fixed documentation bugs and
1576 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1577 exipick to access non-standard spools, including the "frozen" queue
1580 NM/01 Bugzilla 965: Support mysql stored procedures.
1581 Patch from Alain Williams
1583 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1585 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1586 Patch from Andreas Metzler
1588 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1589 Patch from Kirill Miazine
1591 NM/05 Bugzilla 671: Added umask to procmail example.
1593 JJ/03 installed exipick 20100323.0, fixing doc bug
1595 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1596 directory. Notification and patch from Dan Rosenberg.
1598 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1600 TK/02 Improve log output when DKIM signing operation fails.
1602 MH/01 Treat the transport option dkim_domain as a colon separated
1603 list, not as a single string, and sign the message with each element,
1604 omitting multiple occurences of the same signer.
1606 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1607 Bugzilla 985, 986. Patch by Simon Arlott
1609 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1610 Patch by Simon Arlott
1612 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1613 MBX locking. Notification from Dan Rosenberg.
1619 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1621 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1623 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1625 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1627 NM/04 Bugzilla 915: Flag broken perl installation during build.
1633 TK/01 Added patch by Johannes Berg that expands the main option
1634 "spamd_address" if it starts with a dollar sign.
1636 TK/02 Write list of recipients to X-Envelope-Sender header when building
1637 the mbox-format spool file for content scanning (suggested by Jakob
1640 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1641 (http://www.dcc-servers.net/) support via dccifd. Activated by
1642 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1644 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1645 by Mark Daniel Reidel <mr@df.eu>.
1647 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1648 When building exim an external PCRE library is now needed -
1649 PCRE is a system library on the majority of modern systems.
1650 See entry on PCRE_LIBS in EDITME file.
1652 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1653 conversation. Added nologin parameter to request.
1654 Patch contributed by Kirill Miazine.
1656 TF/01 Do not log submission mode rewrites if they do not change the address.
1658 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1660 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1661 log files in place. Contributed by Roberto Lima.
1663 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1665 TF/03 Bugzilla 615: When checking the local_parts router precondition
1666 after a local_part_suffix or local_part_prefix option, Exim now
1667 does not use the address's named list lookup cache, since this
1668 contains cached lookups for the whole local part.
1670 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1671 Robert Millan. Documentation is in experimental-spec.txt.
1673 TF/04 Bugzilla 668: Fix parallel build (make -j).
1675 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1677 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1678 Patch provided by Jan Srzednicki.
1680 TF/05 Leading white space used to be stripped from $spam_report which
1681 wrecked the formatting. Now it is preserved.
1683 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1684 that they are available at delivery time.
1686 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1688 TF/08 TLS error reporting now respects the incoming_interface and
1689 incoming_port log selectors.
1691 TF/09 Produce a more useful error message if an SMTP transport's hosts
1692 setting expands to an empty string.
1694 NM/06 Bugzilla 744: EXPN did not work under TLS.
1695 Patch provided by Phil Pennock.
1697 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1698 Patch provided by Richard Godbee.
1700 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1701 acl_smtp_notquit, added index entry.
1703 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1704 Patch provided by Eugene Bujak.
1706 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1707 accept(). Patch provided by Maxim Dounin.
1709 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1710 Patch provided by Phil Pennock.
1712 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1714 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1715 Patch provided by Brad "anomie" Jorsch.
1717 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1718 Patch provided by Dean Brooks.
1720 TK/05 Add native DKIM support (does not depend on external libraries).
1722 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1723 Patch provided by Graeme Fowler.
1725 NM/16 Bugzilla 851: Documentation example syntax fix.
1727 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1729 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1732 NM/19 Bugzilla 745: TLS version reporting.
1733 Patch provided by Phil Pennock.
1735 NM/20 Bugzilla 167: bool: condition support.
1736 Patch provided by Phil Pennock.
1738 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1739 clients. Patch provided by Phil Pennock.
1741 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1742 Patch provided by Brad "anomie" Jorsch.
1744 NM/23 Bugzilla 687: Fix misparses in eximstats.
1745 Patch provided by Heiko Schlittermann.
1747 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1748 Patch provided by Heiko Schlittermann.
1750 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1751 plus update to original patch.
1753 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1755 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1756 Patch provided by David Brownlee.
1758 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1760 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1762 NM/30 Bugzilla 888: TLS documentation bugfixes.
1764 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1766 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1767 Unlike the original bugzilla I have changed all shell scripts in src tree.
1769 NM/33 Bugzilla 898: Transport filter timeout fix.
1770 Patch by Todd Rinaldo.
1772 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1773 Patch by Serge Demonchaux.
1775 NM/35 Bugzilla 39: Base64 decode bug fixes.
1776 Patch by Jakob Hirsch.
1778 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1780 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1782 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1784 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1790 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1791 ALT-N's libdkim that I have put here:
1792 http://duncanthrax.net/exim-experimental/
1794 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1795 were called 'true' and it seems that C99 defines that as a reserved
1796 keyword to be used with 'bool' variable types. That means you could
1797 not include C99-style headers which use bools without triggering
1798 build errors in sieve.c.
1800 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1801 as mailq or other aliases. Changed the --help handling significantly
1802 to do whats expected. exim_usage() emits usage/help information.
1804 SC/01 Added the -bylocaldomain option to eximstats.
1806 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1808 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1810 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1816 PH/01 Another patch from the Sieve maintainer.
1818 PH/02 When an IPv6 address is converted to a string for single-key lookup
1819 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1820 dots are used instead of colons so that keys in lsearch files need not
1821 contain colons. This was done some time before quoting was made available
1822 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1823 (notated using the quote facility) so as to distinguish them from IPv4
1824 keys. This meant that lookups for IP addresses in host lists did not work
1825 for iplsearch lookups.
1827 This has been fixed by arranging for IPv6 addresses to be expressed with
1828 colons if the lookup type is iplsearch. This is not incompatible, because
1829 previously such lookups could never work.
1831 The situation is now rather anomolous, since one *can* have colons in
1832 ordinary lsearch keys. However, making the change in all cases is
1833 incompatible and would probably break a number of configurations.
1835 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1838 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1839 conversion specification without a maximum field width, thereby enabling
1840 a rogue spamd server to cause a buffer overflow. While nobody in their
1841 right mind would setup Exim to query an untrusted spamd server, an
1842 attacker that gains access to a server running spamd could potentially
1843 exploit this vulnerability to run arbitrary code as the Exim user.
1845 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1846 $primary_hostname instead of what libspf2 thinks the hosts name is.
1848 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1849 a directory entry by the name of the lookup key. Previously, if a
1850 symlink pointed to a non-existing file or a file in a directory that
1851 Exim lacked permissions to read, a lookup for a key matching that
1852 symlink would fail. Now it is enough that a matching directory entry
1853 exists, symlink or not. (Bugzilla 503.)
1855 PH/03 The body_linecount and body_zerocount variables are now exported in the
1858 PH/04 Added the $dnslist_matched variable.
1860 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1861 This means they are set thereafter only if the connection becomes
1864 PH/06 Added the client_condition to authenticators so that some can be skipped
1865 by clients under certain conditions.
1867 PH/07 The error message for a badly-placed control=no_multiline_responses left
1868 "_responses" off the end of the name.
1870 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1872 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1873 (without spaces) instead of just copying the configuration text.
1875 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1877 PH/11 Added $max_received_linelength.
1879 PH/12 Added +ignore_defer and +include_defer to host lists.
1881 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1882 way in which PCRE > 7.0 is built.
1884 PH/14 Implemented queue_only_load_latch.
1886 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1887 MAIL command. The effect was to mangle the value on 64-bit systems.
1889 PH/16 Another patch from the Sieve maintainer.
1891 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1893 PH/18 If a system quota error occurred while trying to create the file for
1894 a maildir delivery, the message "Mailbox is full" was not appended to the
1895 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1896 only to a quota excession during the actual writing of the file.
1898 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1899 characters?) which causes problems in log lines. The DN values are now
1900 passed through string_printing() before being added to log lines.
1902 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1903 and InterBase are left for another time.)
1905 PH/21 Added message_body_newlines option.
1907 PH/22 Guard against possible overflow in moan_check_errorcopy().
1909 PH/23 POSIX allows open() to be a macro; guard against that.
1911 PH/24 If the recipient of an error message contained an @ in the local part
1912 (suitably quoted, of course), incorrect values were put in $domain and
1913 $local_part during the evaluation of errors_copy.
1919 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1920 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1923 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1924 issue a MAIL command.
1926 PH/02 In an ACL statement such as
1928 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1930 if a client was not listed at all, or was listed with a value other than
1931 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1932 the condition was not true (as it should be), so access was not denied.
1933 The bug was that the ! inversion was incorrectly passed on to the second
1934 item. This has been fixed.
1936 PH/03 Added additional dnslists conditions == and =& which are different from
1937 = and & when the dns lookup returns more than one IP address.
1939 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1940 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1942 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1943 FSYNC, which compiles an option called disable_fsync that allows for
1944 bypassing fsync(). The documentation is heavily laced with warnings.
1946 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1948 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1949 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1950 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1951 including adding "make clean"; (3) Added -fPIC when compiling the test
1952 dynamically loaded module, to get rid of a warning.
1954 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1955 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1956 = 0s. The bug is otherwise harmless.
1958 PH/07 There was a bug in the dovecot authenticator such that the value of
1959 $auth1 could be overwritten, and so not correctly preserved, after a
1960 successful authentication. This usually meant that the value preserved by
1961 the server_setid option was incorrect.
1963 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1965 PH/09 Installed PCRE release 7.0.
1967 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1968 run for batched SMTP input. It is now run at the start of every message
1969 in the batch. While fixing this I discovered that the process information
1970 (output by running exiwhat) was not always getting set for -bs and -bS
1971 input. This is fixed, and it now also says "batched" for BSMTP.
1973 PH/11 Added control=no_pipelining.
1975 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1976 patch, slightly modified), and move the expansion of helo_data till after
1977 the connection is made in the smtp transport (so it can use these
1980 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1982 PH/14 Added log_selector = +pid.
1984 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1986 PH/16 Add ${if forany and ${if forall.
1988 PH/17 Added dsn_from option to vary the From: line in DSNs.
1990 PH/18 Flush SMTP output before performing a callout, unless control =
1991 no_callout_flush is set.
1993 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1994 was true (the default) a successful delivery failed to delete the retry
1995 item, thus causing premature timeout of the address. The bug is now
1998 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2000 PH/21 Long custom messages for fakedefer and fakereject are now split up
2001 into multiline reponses in the same way that messages for "deny" and
2002 other ACL rejections are.
2004 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2005 with slight modification.
2007 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2008 draft, changing the syntax and factoring some duplicate code".
2010 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2011 for deliveries of the second and subsequent messages over the same SMTP
2014 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2015 ${reduce, with only minor "tidies".
2017 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2019 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2020 expansion side effects.
2022 PH/27 When a message times out after an over-quota error from an Exim-imposed
2023 quota, the bounce message says "mailbox is full". This message was not
2024 being given when it was a system quota that was exceeded. It now should
2027 MH/03 Made $recipients available in local_scan(). local_scan() already has
2028 better access to the recipient list through recipients_list[], but
2029 $recipients can be useful in postmaster-provided expansion strings.
2031 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2032 in the case of a MAIL command with additional options following the
2033 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2034 were accidentally chopped off.
2036 PH/29 SMTP synchronization checks are implemented when a command is read -
2037 there is a check that no more input is waiting when there shouldn't be
2038 any. However, for some commands, a delay in an ACL can mean that it is
2039 some time before the response is written. In this time, more input might
2040 arrive, invalidly. So now there are extra checks after an ACL has run for
2041 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2042 pipelining has not been advertised.
2044 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2046 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2047 returned in the SMTP response when smtp_return_error_details was false.
2048 This has been fixed.
2050 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2051 instead of the C I/O that was originally supplied, because problems were
2052 reported on Solaris.
2054 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2055 Exim which did not show up earlier: it was assuming that a call to
2056 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2057 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2058 was a macro that became an assignment, so it seemed to work. This has
2059 changed to a proper function call with a void return, hence the compile
2060 error. Exim's code has been fixed.
2062 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2065 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2067 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2069 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2070 because it turns out to be too big or there is a timeout) there is no
2071 "Completed" line in the log. When some messages of this type were
2072 selected by exigrep, they were listed as "not completed". Others were
2073 picked up by some special patterns. I have improved the selection
2074 criteria to be more general.
2076 PH/38 The host_find_failed option in the manualroute router can now be set
2077 to "ignore", to completely ignore a host whose IP address cannot be
2078 found. If all hosts are ignored, the behaviour is controlled by the new
2079 host_all_ignored option.
2081 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2082 homing or because of multiple MX records with /mx) generated more than
2083 one IP address, and the following item turned out to be the local host,
2084 all the secondary addresses of the first item were incorrectly removed
2085 from the list, along with the local host and any following hosts (which
2086 is what is supposed to happen).
2088 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2089 whoever called Exim into the -H file. In the case of the daemon it was
2090 behaving confusingly. When first started, it used values for whoever
2091 started the daemon, but after a SIGHUP it used the Exim user (because it
2092 calls itself on a restart). I have changed the code so that it now always
2095 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2096 message are rejected with the same error (e.g. no authentication or bad
2097 sender address), and a DATA command is nevertheless sent (as can happen
2098 with PIPELINING or a stupid MUA), the error message that was given to the
2099 RCPT commands is included in the rejection of the DATA command. This is
2100 intended to be helpful for MUAs that show only the final error to their
2103 PH/42 Another patch from the Sieve maintainer.
2105 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2106 Eximstats - Fixed some broken HTML links and added missing column headers
2108 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2109 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2111 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2113 PH/43 Yet another patch from the Sieve maintainer.
2115 PH/44 I found a way to check for a TCP/IP connection going away before sending
2116 the response to the final '.' that terminates a message, but only in the
2117 case where the client has not sent further data following the '.'
2118 (unfortunately, this is allowed). However, in many cases there won't be
2119 any further data because there won't be any more messages to send. A call
2120 to select() can be used: if it shows that the input is "ready", there is
2121 either input waiting, or the socket has been closed. An attempt to read
2122 the next input character can distinguish the two cases. Previously, Exim
2123 would have sent an OK response which the client would never have see.
2124 This could lead to message repetition. This fix should cure that, at
2125 least in a lot of common cases.
2127 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2128 advertised in response to EHLO.
2134 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2135 fixed by 4.65/MH/01 (is this a record?) are fixed:
2137 (i) An empty string was always treated as zero by the numeric comparison
2138 operators. This behaviour has been restored.
2140 (ii) It is documented that the numeric comparison operators always treat
2141 their arguments as decimal numbers. This was broken in that numbers
2142 starting with 0 were being interpreted as octal.
2144 While fixing these problems I realized that there was another issue that
2145 hadn't been noticed. Values of message_size_limit (both the global option
2146 and the transport option) were treated as octal if they started with 0.
2147 The documentation was vague. These values are now always treated as
2148 decimal, and I will make that clear in the documentation.
2154 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2155 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2158 MH/01 Don't check that the operands of numeric comparison operators are
2159 integers when their expansion is in "skipping" mode (fixes bug
2160 introduced by 4.64-PH/07).
2162 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2163 child addresses, Exim now panics and dies. Previously, because the count
2164 is held in a short int, deliveries were likely to be lost. As such a
2165 large number of recipients for a single message is ridiculous
2166 (performance will be very, very poor), I have chosen to impose a limit
2167 rather than extend the field.
2173 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2174 leftover -K file (the existence of which was triggered by #402).
2175 While we were at it, introduced process PID as part of the -K
2176 filename. This should rule out race conditions when creating
2179 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2180 processing considerably. Previous code took too long for large mails,
2181 triggering a timeout which in turn triggers #401.
2183 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2184 in the DK code in transports.c. sendfile() is not really portable,
2185 hence the _LINUX specificness.
2187 TF/01 In the add_headers option to the mail command in an Exim filter,
2188 there was a bug that Exim would claim a syntax error in any
2189 header after the first one which had an odd number of characters
2192 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2193 callout verification, Exim cached a "reject" for the entire domain. This
2194 is correct for most verifications, but it is not correct for a recipient
2195 verification with use_sender or use_postmaster set, because in that case
2196 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2197 case of MAIL FROM:<> rejection from other early rejections (e.g.
2198 rejection of HELO). When verifying a recipient using a non-null MAIL
2199 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2200 Whatever the result of the callout, the value of the domain cache is
2201 left unchanged (for any other kind of callout, getting as far as trying
2202 RCPT means that the domain itself is ok).
2204 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2207 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2208 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2209 session, the daemon ignores ECONNECT errors and logs others; it now
2210 ignores EPIPE as well.
2212 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2213 (quoted-printable decoding).
2215 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2216 later the small subsequent patch to fix an introduced bug.
2218 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2220 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2222 PH/08 An error is now given if message_size_limit is specified negative.
2224 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2225 to be given (somewhat) arbitrary names.
2227 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2230 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2231 miscellaneous code fixes
2233 PH/10 Added the log_reject_target ACL modifier to specify where to log
2236 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2237 hostname. This is wrong, because it relates to the incoming message (and
2238 probably the interface on which it is arriving) and not to the outgoing
2239 callout (which could be using a different interface). This has been
2240 changed to use the value of the helo_data option from the smtp transport
2241 instead - this is what is used when a message is actually being sent. If
2242 there is no remote transport (possible with a router that sets up host
2243 addresses), $smtp_active_hostname is used.
2245 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2246 tweaks were necessary in order to get it to work (see also 21 below):
2247 (a) The code assumed that strncpy() returns a negative number on buffer
2248 overflow, which isn't the case. Replaced with Exim's string_format()
2250 (b) There were several signed/unsigned issues. I just did the minimum
2251 hacking in of casts. There is scope for a larger refactoring.
2252 (c) The code used strcasecmp() which is not a standard C function.
2253 Replaced with Exim's strcmpic() function.
2254 (d) The code set only $1; it now sets $auth1 as well.
2255 (e) A simple test gave the error "authentication client didn't specify
2256 service in request". It would seem that Dovecot has changed its
2257 interface. Fortunately there's a specification; I followed it and
2258 changed what the client sends and it appears to be working now.
2260 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2263 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2264 address A is aliased to B and C, where B exists and C does not. Without
2265 -v the output is "A verified" because verification stops after a
2266 successful redirection if more than one address is generated. However,
2267 with -v the child addresses are also verified. Exim was outputting "A
2268 failed to verify" and then showing the successful verification for C,
2269 with its parentage. It now outputs "B failed to verify", showing B's
2270 parentage before showing the successful verification of C.
2272 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2273 look up a TXT record in a specific list after matching in a combined
2276 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2277 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2278 they consult the DNS. I had assumed they would set it the way they
2279 wanted; and indeed my experiments on Linux seem to show that in some
2280 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2281 To be on the safe side, however, I have now made the interface to
2282 host_find_byname() similar to host_find_bydns(), with an argument
2283 containing the DNS resolver options. The host_find_byname() function now
2284 sets these options at its start, just as host_find_bydns() does. The smtp
2285 transport options dns_qualify_single and dns_search_parents are passed to
2286 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2287 of host_find_byname() use the default settings of RES_DEFNAMES
2288 (qualify_single) but not RES_DNSRCH (search_parents).
2290 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2291 spool_read_header() do less string testing, by means of a preliminary
2292 switch on the second character of optional "-foo" lines. (This is
2293 overdue, caused by the large number of possibilities that now exist.
2294 Originally there were few.) While I was there, I also converted the
2295 str(n)cmp tests so they don't re-test the leading "-" and the first
2296 character, in the hope this might squeeze out yet more improvement.
2298 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2299 flag allowing group syntax was set by the header_syntax check but not
2300 turned off, possible causing trouble later; (2) The flag was not being
2301 set at all for the header_verify test, causing "group"-style headers to
2302 be rejected. I have now set it in this case, and also caused header_
2303 verify to ignore an empty address taken from a group. While doing this, I
2304 came across some other cases where the code for allowing group syntax
2305 while scanning a header line wasn't quite right (mostly, not resetting
2306 the flag correctly in the right place). These bugs could have caused
2307 trouble for malformed header lines. I hope it is now all correct.
2309 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2310 with the "reply" argument non-NULL. The code, however (which originally
2311 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2312 but it didn't always do it. This confused somebody who was copying the
2313 code for some other use. I have removed all the tests.
2315 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2316 feature that was used to support insecure browsers during the U.S. crypto
2317 embargo. It requires special client support, and Exim is probably the
2318 only MTA that supported it -- and would never use it because real RSA is
2319 always available. This code has been removed, because it had the bad
2320 effect of slowing Exim down by computing (never used) parameters for the
2321 RSA_EXPORT functionality.
2323 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2324 authenticator to fail if there's a tab character in the incoming data
2325 (there should never be unless someone is messing about, as it's supposed
2326 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2327 if the connection is using TLS or if the remote IP is the same as the
2328 local IP, and the "valid-client-cert option" if a client certificate has
2331 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2332 authenticators. This can be used for authorization after authentication
2333 succeeds. (In the case of plaintext, it servers for both authentication
2336 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2337 if any retry times were supplied.
2339 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2340 connection, where there is no client IP address to check. In this
2341 situation, the verify now always succeeds.
2343 PH/25 Applied John Jetmore's -Mset patch.
2345 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2347 PH/27 In a string expansion for a processed (not raw) header when multiple
2348 headers of the same name were present, leading whitespace was being
2349 removed from all of them, but trailing whitespace was being removed only
2350 from the last one. Now trailing whitespace is removed from each header
2351 before concatenation. Completely empty headers in a concatenation (as
2352 before) are ignored.
2354 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2355 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2357 PH/29 [Removed. This was a change that I later backed out, and forgot to
2358 correct the ChangeLog entry (that I had efficiently created) before
2359 committing the later change.]
2361 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2362 address errors (4xx response to RCPT) over the same connection as other
2363 messages routed to the same hosts. Such deliveries are always "forced",
2364 so retry times are not inspected. This resulted in far too many retries
2365 for the affected addresses. The effect occurred only when there were more
2366 hosts than the hosts_max_try setting in the smtp transport when it had
2367 the 4xx errors. Those hosts that it had tried were not added to the list
2368 of hosts for which the message was waiting, so if all were tried, there
2369 was no problem. Two fixes have been applied:
2371 (i) If there are any address or message errors in an SMTP delivery, none
2372 of the hosts (tried or untried) are now added to the list of hosts
2373 for which the message is waiting, so the message should not be a
2374 candidate for sending over the same connection that was used for a
2375 successful delivery of some other message. This seems entirely
2376 reasonable: after all the message is NOT "waiting for some host".
2377 This is so "obvious" that I'm not sure why it wasn't done
2378 previously. Hope I haven't missed anything, but it can't do any
2379 harm, as the worst effect is to miss an optimization.
2381 (ii) If, despite (i), such a delivery is accidentally attempted, the
2382 routing retry time is respected, so at least it doesn't keep
2383 hammering the server.
2385 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2386 in ${readsocket because some servers need this prod.
2388 PH/32 Added some extra debug output when updating a wait-xxx database.
2390 PH/33 The hint "could be header name not terminated by colon", which has been
2391 given for certain expansion errors for a long time, was not being given
2392 for the ${if def:h_colon_omitted{... case.
2394 PH/34 The spec says: "With one important exception, whenever a domain list is
2395 being scanned, $domain contains the subject domain." There was at least
2396 one case where this was not true.
2398 PH/35 The error "getsockname() failed: connection reset by peer" was being
2399 written to the panic log as well as the main log, but it isn't really
2400 panic-worthy as it just means the connection died rather early on. I have
2401 removed the panic log writing for the ECONNRESET error when getsockname()
2404 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2405 runs only) independently of the message's sender address. This meant
2406 that, if the 4xx error was in fact related to the sender, a different
2407 message to the same recipient with a different sender could confuse
2408 things. In particualar, this can happen when sending to a greylisting
2409 server, but other circumstances could also provoke similar problems.
2410 I have changed the default so that the retry time for these errors is now
2411 based a combination of the sender and recipient addresses. This change
2412 can be overridden by setting address_retry_include_sender=false in the
2415 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2416 remote server are returned as part of bounce messages. This was not
2417 happening for LMTP over a pipe (the lmtp transport), but now it is the
2418 same for both kinds of LMTP.
2420 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2421 in header lines that were in fact CNAMEs. This is no longer the case.
2423 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2424 and queue runs started by the daemon processed all messages. This has
2425 been fixed so that -R and -S can now usefully be given with -q<time>.
2427 PH/40 Import PCRE release 6.7 (fixes some bugs).
2429 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2431 PH/42 Give an error if -q is specified more than once.
2433 PH/43 Renamed the variables $interface_address and $interface_port as
2434 $received_ip_address and $received_port, to make it clear that these
2435 values apply to message reception, and not to the outgoing interface when
2436 a message is delivered. (The old names remain recognized, of course.)
2438 PH/44 There was no timeout on the connect() call when using a Unix domain
2439 socket in the ${readsocket expansion. There now is.
2441 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2442 be meaningful with "accept".
2444 SC/01 Eximstats V1.43
2445 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2447 SC/02 Eximstats V1.44
2448 Use a glob alias rather than an array ref in the generated
2449 parser. This improves both readability and performance.
2451 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2452 Collect SpamAssassin and rejection statistics.
2453 Don't display local sender or destination tables unless
2454 there is data to show.
2455 Added average volumes into the top table text output.
2457 SC/04 Eximstats V1.46
2458 Collect data on the number of addresses (recipients)
2459 as well as the number of messages.
2461 SC/05 Eximstats V1.47
2462 Added 'Message too big' to the list of mail rejection
2463 reasons (thanks to Marco Gaiarin).
2465 SC/06 Eximstats V1.48
2466 Mainlog lines which have GMT offsets and are too short to
2467 have a flag are now skipped.
2469 SC/07 Eximstats V1.49 (Alain Williams)
2470 Added the -emptyok flag.
2472 SC/08 Eximstats V1.50
2473 Fixes for obtaining the IP address from reject messages.
2475 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2476 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2477 whitesspace changes from 4.64-PH/27
2479 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2482 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2483 are found, allow negative numbers in numeric criteria)
2485 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2487 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2488 to match changes made in 4.64-PH/43
2490 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2492 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2493 "domain" is actually an IP address, and force a failure. This locks out
2494 those revolvers/nameservers that support "A-for-A" lookups, in
2495 contravention of the specifications.
2497 PH/48 When a host name was looked up from an IP address, and the subsequent
2498 forward lookup of the name timed out, the host name was left in
2499 $sender_host_name, contrary to the specification.
2501 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2502 restricted to single-key lookups, Exim was not diagnosing an error if
2503 * or *@ was used with a query-style lookup.
2505 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2507 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2508 long ago, but noone interested enough thought of it. Let's just say that
2509 the "1.1" means that there are some new functions that weren't there at
2510 some point in the past.
2512 PH/51 Error processing for expansion failure of helo_data from an smtp
2513 transport during callout processing was broken.
2515 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2516 tested/used via the -bh/-bhc/-bs options.
2518 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2519 bug, fixed in subsequent PCRE releases).
2521 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2522 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2524 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2530 SC/01 Use a glob alias rather than an array ref in eximstats generated
2531 parser. This improves both readability and performance.
2533 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2534 Don't display local sender or destination tables in eximstats unless
2535 there is data to show.
2536 Added average volumes into the eximstats top table text output.
2538 SC/03 Collect data on the number of addresses (recipients) as well
2539 as the number of messages in eximstats.
2541 TF/01 Correct an error in the documentation for the redirect router. Exim
2542 does (usually) call initgroups() when daemonizing.
2544 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2545 with consistent privilege compared to when running as a daemon.
2547 TF/03 Note in the spec that $authenticated_id is not set for local
2548 submissions from trusted users.
2550 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2551 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2553 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2554 by adding some example configuration directives to the default
2555 configuration file. A little bit of work is required to uncomment the
2556 directives and define how usernames and passwords are checked, but
2557 there is now a framework to start from.
2559 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2560 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2561 without this. I don't know how relevant this is to other LDAP libraries.
2563 PH/02 Add the verb name to the "unknown ACL verb" error.
2565 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2567 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2569 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2570 directory not expanded when it should be if an expanded home directory
2571 was set for the address (which is overridden by the transport).
2573 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2576 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2577 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2578 because it is too late at that time, and has no effect.
2580 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2581 security issue with \' (bugzilla #107). I could not use the
2582 PQescapeStringConn() function, because it needs a PGconn value as one of
2585 PH/08 When testing addresses using -bt, indicate those final addresses that
2586 are duplicates that would not cause an additional delivery. At least one
2587 person was confused, thinking that -bt output corresponded to deliveries.
2588 (Suppressing duplicates isn't a good idea as you lose the information
2589 about possibly different redirections that led to the duplicates.)
2591 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2592 systems where poll() doesn't work, in particular OS X.
2594 PH/10 Added more information to debugging output for retry time not reached.
2596 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2597 operations in malware.c.
2599 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2602 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2603 syslog_duplication set false, log lines that would normally be written
2604 both the the main log and to the reject log were not written to syslog at
2607 PH/14 In the default configuration, change the use of "message" in ACL warn
2608 statements to "add_header".
2610 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2611 not followed by a command (e.g. "seen endif").
2613 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2614 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2617 PH/17 Added extra conditions to the default value of delay_warning_condition
2621 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2622 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2623 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2626 The Auto-Submitted: and various List- headers are standardised, whereas I
2627 don't think Precedence: ever was.
2629 PH/18 Refactored debugging code in route_finduser() to show more information,
2630 in particular, the error code if getpwnam() issues one.
2632 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2633 This is apparently needed in addition to the PH/07 change above to avoid
2634 any possible encoding problems.
2636 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2637 but not after initializing Perl.
2639 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2640 output them only if debugging. By default they are written stderr,
2641 apparently, which is not desirable.
2643 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2646 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2649 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2651 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2652 authenticated or an ident call has been made. Suppress the default
2653 values for $authenticated_id and $authenticated_sender (but permit -oMai
2654 and -oMas) when testing with -bh.
2656 PH/24 Re-jigged the order of the tests in the default configuration so that the
2657 tests for valid domains and recipients precede the DNS black list and CSA
2658 tests, on the grounds that those ones are more expensive.
2660 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2661 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2662 This bug exists in every version of Exim that I still have, right back to
2665 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2666 However, an attempt to turn on case-sensitivity in a regex key by
2667 including (?-i) didn't work because the subject string was already
2668 lowercased, and the effects were non-intuitive. It turns out that a
2669 one-line patch can be used to allow (?-i) to work as expected.
2675 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2676 other effects) broke the use of negated acl sub-conditions.
2678 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2681 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2682 "Deny" causes Exim to reject the incoming connection with a 554 error.
2683 Unfortunately, if there is a major crisis, such as a disk failure,
2684 tcp-wrappers gives "deny", whereas what one would like would be some
2685 kind of temporary error. A kludge has been added to help with this.
2686 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2687 554 error is used if errno is still zero or contains ENOENT (which occurs
2688 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2691 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2693 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2694 errors. Otherwise a message that provokes a temporary error (when other
2695 messages do not) can cause a whole host to time out.
2697 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2698 addresses were routed directly to files or pipes from a redirect router.
2699 File deliveries just didn't batch; pipe deliveries might have suffered
2702 PH/06 A failure to get a lock for a hints database would erroneously always say
2703 "Failed to get write lock", even when it was really a read lock.
2705 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2706 of 0600. This has been changed to use the value of the lockfile_mode
2707 option (which defaults to 0600).
2709 PH/08 Applied small patch from the Sieve maintainer.
2711 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2712 folder from quota calculations, a direct delivery into this folder messed
2713 up the contents of the maildirsize file. This was because the regex was
2714 used only to exclude .Trash (or whatever) when the size of the mailbox
2715 was calculated. There was no check that a delivery was happening into an
2716 excluded directory. This bug has been fixed by ignoring all quota
2717 processing for deliveries into excluded directories.
2719 PH/10 Added the maildirfolder_create_regex option to appendfile.
2725 PH/01 The code for finding all the local interface addresses on a FreeBSD
2726 system running IPv6 was broken. This may well have applied to all BSD
2727 systems, as well as to others that have similar system calls. The broken
2728 code found IPv4 interfaces correctly, but gave incorrect values for the
2729 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2730 that it would not match correctly against @[] and not recognize the IPv6
2733 PH/02 The ipliteral router was not recognizing addresses of the form user@
2734 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2736 PH/03 Added disable_ipv6.
2738 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2739 decoded form, because it is most often used to construct To: headers
2740 lines in autoreplies, and the decoded form may well be syntactically
2741 invalid. However, $reply_address has leading white space removed, and all
2742 newlines turned into spaces so that the autoreply transport does not
2745 PH/05 If group was specified without a user on a router, and no group or user
2746 was specified on a transport, the group from the router was ignored.
2748 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2749 for visible compile-time settings that can be used to change these
2750 numbers, for those that want even more. Backwards compatibility with old
2751 spool files has been maintained. However, going back to a previous Exim
2752 release will lost any variables that are in spool files.
2754 PH/07 Two small changes when running in the test harness: increase delay when
2755 passing a TCP/IP connection to a new process, in case the original
2756 process has to generate a bounce, and remove special handling of
2757 127.0.0.2 (sic), which is no longer necessary.
2759 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2760 be the same on different OS.
2762 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2765 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2766 whether --show-vars was specified or not
2768 JJ/02 exipick: Added support for new ACL variable spool format introduced
2771 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2772 syntactically invalid From: or Reply-to: line, and a filter used this to
2773 generate an autoreply, and therefore failed to obtain an address for the
2774 autoreply, Exim could try to deliver to a non-existent relative file
2775 name, causing unrelated and misleading errors. What now happens is that
2776 it logs this as a hard delivery error, but does not attempt to create a
2779 PH/11 The exinext utility has a -C option for testing purposes, but although
2780 the given file was scanned by exinext itself; it wasn't being passed on
2781 when Exim was called.
2783 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2784 an end-of-file indication when reading a command response.
2786 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2787 compiled. In many other places in Exim, IPv6 addresses are always
2788 recognized, so I have changed this. It also means that IPv4 domain
2789 literals of the form [IPV4:n.n.n.n] are now always recognized.
2791 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2792 used if the router is not running as root, for example, when verifying at
2793 ACL time, or when using -bh. The debugging output from this situation was
2794 non-existent - all you got was a failure to exec. I have made two
2797 (a) Failures to set uid/gid, the current directory, or a process leader
2798 in a subprocess such as that created by queryprogram now generate
2799 suitable debugging ouput when -d is set.
2801 (b) The queryprogram router detects when it is not running as root,
2802 outputs suitable debugging information if -d is set, and then runs
2803 the subprocess without attempting to change uid/gid.
2805 PH/15 Minor change to Makefile for building test_host (undocumented testing
2808 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2809 additional section of a DNS packet that returns MX or SRV records.
2810 Instead, it always explicitly searches for A/AAAA records. This avoids
2811 major problems that occur when a DNS server includes only records of one
2812 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2813 fixed another bug: if SRV records were looked up and the corresponding
2814 address records were *not* found in the additional section, the port
2815 values from the SRV records were lost.
2817 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2818 using the correct key (the original address) when searching the retry
2819 rules in order to find which one to use for generating the retry hint.
2821 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2822 adding the default one. Similarly, if it contains a Reply-To: header, the
2823 errors_reply_to option, if set, is not used.
2825 PH/19 When calculating a retry time, Exim used to measure the "time since
2826 failure" by looking at the "first failed" field in the retry record. Now
2827 it does not use this if it is later than than the arrival time of the
2828 message. Instead it uses the arrival time. This makes for better
2829 behaviour in cases where some deliveries succeed, thus re-setting the
2830 "first failed" field. An example is a quota failure for a huge message
2831 when small messages continue to be delivered. Without this change, the
2832 "time since failure" will always be short, possible causing more frequent
2833 delivery attempts for the huge message than are intended.
2834 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2836 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2837 $1, $2, $3) because the numerical variables can be reset during some
2838 expansion items (e.g. "match"), thereby losing the authentication data.
2840 PH/21 Make -bV show the size of off_t variables so that the test suite can
2841 decide whether to run tests for quotas > 2G.
2843 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2844 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2845 filecount value is greater than 2G or if a quota value is greater than 2G
2846 on a system where the size of off_t is not greater than 4, a panic error
2849 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2850 never match. The debug and -bh output now contains an explicit error
2851 message indicating a malformed IPv4 address or mask.
2853 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2854 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2855 PH/23 above applies.
2857 PH/25 Do not write to syslog when running in the test harness. The only
2858 occasion when this arises is a failure to open the main or panic logs
2859 (for which there is an explicit test).
2861 PH/26 Added the /no_tell option to "control=freeze".
2863 PH/27 If a host name lookup failed very early in a connection, for example, if
2864 the IP address matched host_lookup and the reverse lookup yielded a name
2865 that did not have a forward lookup, an error message of the form "no IP
2866 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2867 could be logged. Now it outputs the IP address instead of "NULL".
2869 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2870 allows the sender and the authenticated sender to be set when
2871 submitting a message from within Exim. Since child_open_exim() is
2872 documented for local_scan(), the new function should be too.
2874 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2875 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2876 results in an empty string is now treated as unset.
2878 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2880 PH/31 Added sender_verify_fail logging option.
2882 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2883 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2884 tidied the source and removed it altogether.
2886 PH/33 When a queue run was abandoned because the load average was too high, a
2887 log line was always written; now it is written only if the queue_run log
2888 selector is set. In addition, the log line for abandonment now contains
2889 information about the queue run such as the pid. This is always present
2890 in "start" and "stop" lines but was omitted from the "abandon" line.
2892 PH/34 Omit spaces between a header name and the colon in the error message that
2893 is given when verify = headers_syntax fails (if there are lots of them,
2894 the message gets confusing).
2896 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2897 names, as there are now some PTR records that contain slashes. This check
2898 is only to protect against broken name servers that fall over on strange
2899 characters, so the fact that it applies to all lookups doesn't matter.
2901 PH/36 Now that the new test suite is complete, we can remove some of the
2902 special code in Exim that was needed for the old test suite. For example,
2903 sorting DNS records because real resolvers return them in an arbitrary
2904 order. The new test suite's fake resolver always returns records in the
2907 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2908 bounces) except when queue_only is set, to avoid logging races between
2909 the different processes.
2911 PH/38 Panic-die if .include specifies a non-absolute path.
2913 PH/39 A tweak to the "H" retry rule from its user.
2915 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2916 a label. They prevented compilation on older perls.
2918 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2919 a warning to be raised on newish perls.
2921 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2922 on queue. Changes to match documented behaviour of showing count of
2923 messages matching specified criteria.
2925 PH/40 Changed the default ident timeout from 30s to 5s.
2927 PH/41 Added support for the use of login_cap features, on those BSD systems
2928 that have them, for controlling the resources used by pipe deliveries.
2930 PH/42 The content-scanning code uses fopen() to create files in which to put
2931 message data. Previously it was not paying any attention to the mode of
2932 the files. Exim runs with umask(0) because the rest of the code creates
2933 files with open(), and sets the required mode explicitly. Thus, these
2934 files were ending up world-writeable. This was not a big issue, because,
2935 being within the spool directory, they were not world-accessible. I have
2936 created a function called modefopen, which takes an additional mode
2937 argument. It sets umask(777), creates the file, chmods it to the required
2938 mode, then resets the umask. All the relevant calls to fopen() in the
2939 content scanning code have been changed to use this function.
2941 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2942 to 24 hours. This avoids potential overflow problems when processing G
2943 and H retry rules. I suspect nobody ever tinkers with this value.
2945 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2947 PH/45 When the plaintext authenticator is running as a client, the server's
2948 challenges are checked to ensure they are valid base64 strings. By
2949 default, the authentication attempt is cancelled if an invalid string is
2950 received. Setting client_ignore_invalid_base64 true ignores these errors.
2951 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2952 they are received. Thus, the responses can be made to depend on the
2953 challenges. If an invalid string is ignored, an empty string is placed in
2956 PH/46 Messages that are created by the autoreply transport now contains a
2957 References: header, in accordance with RFCs 2822 and 3834.
2959 PH/47 Added authenticated_sender_force to the smtp transport.
2961 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2963 PH/49 Installed latest patch from the Sieve maintainer.
2965 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2966 was also set, the appendfile transport was unnecessarily scanning a
2967 directory of message files (e.g. for maildir delivery) to find the count
2968 of files (along with the size), even though it did not need this
2969 information. It now does the scan only if it needs to find either the
2970 size of the count of files.
2972 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2974 PH/52 Two bugs concerned with error handling when the smtp transport is
2977 (i) Exim was not creating retry information for temporary errors given
2978 for individual recipients after the DATA command when the smtp transport
2979 was used in LMTP mode. This meant that they could be retried too
2980 frequently, and not timed out correctly.
2982 (ii) Exim was setting the flag that allows error details to be returned
2983 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2984 recipients that were returned after the DATA command.
2986 PH/53 This is related to PH/52, but is more general: for any failing address,
2987 when detailed error information was permitted to be returned to the
2988 sender, but the error was temporary, then after the final timeout, only
2989 "retry timeout exceeded" was returned. Now it returns the full error as
2990 well as "retry timeout exceeded".
2992 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2993 do this, and (what is worse) MTAs that accept it.
2995 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2996 will now be deprecated.
2998 PH/56 New os.c-cygwin from the Cygwin maintainer.
3000 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3001 formats (previously only available in exim formats via -bpr, -bpru,
3002 and -bpra. Now also available in native and exiqgrep formats)
3004 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3005 with very large, slow to parse queues
3007 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3009 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3011 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3012 responses to authentication challenges, though it was showing the
3013 challenges; (ii) I've removed the CR characters from the debug output for
3016 PH/58 Allow for the insertion of a newline as well as a space when a string
3017 is turned into more than one encoded-word during RFC 2047 encoding. The
3018 Sieve code now uses this.
3020 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3021 data_4xx, lost_connection, tls_required.
3023 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3024 message was being sent as an SMTP response.
3026 PH/61 Add -l and -k options to exicyclog.
3028 PH/62 When verifying, if an address was redirected to one new address, so that
3029 verification continued, and the new address failed or deferred after
3030 having set something in $address_data, the value of $address_data was not
3031 passed back to the ACL. This was different to the case when no
3032 redirection occurred. The value is now passed back in both cases.
3034 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3035 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3036 use login_cap.h, so on its own it isn't the distinguishing feature. The
3037 new name refers directly to the setclassresources() function.
3039 PH/65 Added configuration files for NetBSD3.
3041 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3043 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3044 is preferred over IPv4.
3046 PH/68 The bounce_return_message and bounce_return_body options were not being
3047 honoured for bounces generated during the reception of non-SMTP messages.
3048 In particular, this applied to messages rejected by the ACL. This bug has
3049 been fixed. However, if bounce_return_message is true and bounce_return_
3050 body is false, the headers that are returned for a non-SMTP message
3051 include only those that have been read before the error was detected.
3052 (In the case of an ACL rejection, they have all been read.)
3054 PH/69 The HTML version of the specification is now built in a directory called
3055 spec_html instead of spec.html, because the latter looks like a path with
3056 a MIME-type, and this confuses some software.
3058 PH/70 Catch two compiler warnings in sieve.c.
3060 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3061 function verify_get_ident() calls ip_connect() to connect a socket, but
3062 if the "connect()" function timed out, ip_connect() used to close the
3063 socket. However, verify_get_ident() also closes the socket later, and in
3064 between Exim writes to the log, which may get opened at this point. When
3065 the socket was closed in ip_connect(), the log could get the same file
3066 descriptor number as the socket. This naturally causes chaos. The fix is
3067 not to close the socket in ip_connect(); the socket should be closed by
3068 the function that creates it. There was only one place in the code where
3069 this was missing, in the iplookup router, which I don't think anybody now
3070 uses, but I've fixed it anyway.
3072 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3073 well as to direct DNS lookups. Otherwise the handling of names in host
3074 lists is inconsistent and therefore confusing.
3080 PH/01 Two changes to the default runtime configuration:
3082 (1) Move the checks for relay_from_hosts and authenticated clients from
3083 after to before the (commented out) DNS black list checks.
3085 (2) Add control=submission to the relay_from_hosts and authenticated
3086 clients checks, on the grounds that messages accepted by these
3087 statements are most likely to be submissions.
3089 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3091 (1) Generate an error if the third argument for the ${prvs expansion is
3094 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3097 (3) Reset the variables that are obtained from the first argument of
3098 ${prvscheck and used in the second argument before leaving the code,
3099 because their memory is reclaimed, so using them afterwards may do
3102 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3103 one (it's much easier than Tom thought :-).
3105 (5) Because of (4), we can now allow for the use of $prvscheck_result
3106 inside the third argument.
3108 PH/03 For some reason, the default setting of PATH when running a command from
3109 a pipe transport was just "/usr/bin". I have changed it to
3112 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3113 anything to be listed in the output from -bV.
3115 PH/05 When a filter generated an autoreply, the entire To: header line was
3116 quoted in the delivery log line, like this:
3118 => >A.N.Other <ano@some.domain> <original@ddress> ...
3120 This has been changed so that it extracts the operative address. There
3121 may be more than one such address. If so, they are comma-separated, like
3124 => >ano@some.domain,ona@other.domain <original@ddress> ...
3126 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3127 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3128 not being looked up in the rDNS to get a host name, Exim was showing the
3129 IP address twice in Received: lines, even though the IP addresses were
3130 identical. For example:
3132 Received: from [1.2.3.4] (helo=[1.2.3.4])
3134 However, if the real host name was known, it was omitting the HELO data
3135 if it matched the actual IP address. This has been tidied up so that it
3136 doesn't show the same IP address twice.
3138 PH/07 When both +timestamp and +memory debugging was on, the value given by
3139 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3140 called by the debug printing, thereby overwriting the timestamp buffer.
3141 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3143 PH/08 When the original message was included in an autoreply transport, it
3144 always said "this is a copy of the message, including all the headers",
3145 even if body_only or headers_only was set. It now gives an appropriate
3148 PH/09 Applied a patch from the Sieve maintainer which:
3150 o fixes some comments
3151 o adds the (disabled) notify extension core
3152 o adds some debug output for the result of if/elsif tests
3153 o points to the current vacation draft in the documentation
3154 and documents the missing references header update
3158 o fixes a bug in processing the envelope test (when testing
3159 multiple envelope elements, the last element determinted the
3162 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3163 Electronic Mail") by including:
3165 Auto-submitted: auto-generated
3167 in the messages that it generates (bounce messages and others, such as
3168 warnings). In the case of bounce messages for non-SMTP mesages, there was
3169 also a typo: it was using "Auto_submitted" (underscore instead of
3170 hyphen). Since every message generated by Exim is necessarily in response
3171 to another message, thes have all been changed to:
3173 Auto-Submitted: auto-replied
3175 in accordance with these statements in the RFC:
3177 The auto-replied keyword:
3179 - SHOULD be used on messages sent in direct response to another
3180 message by an automatic process,
3182 - MUST NOT be used on manually-generated messages,
3184 - MAY be used on Delivery Status Notifications (DSNs) and Message
3185 Disposition Notifications (MDNs),
3187 - MUST NOT be used on messages generated by automatic or periodic
3188 processes, except for messages which are automatic responses to
3191 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3192 to the default Received: header definition.
3194 PH/12 Added log selector acl_warn_skipped (default on).
3196 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3197 variables because (a) they are in the wrong storage pool and (b) even if
3198 they were copied, it wouldn't work properly because of the caching.
3200 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3201 checking when decoding. Apparently there are clients that generate
3202 overlong encoded strings. Why am I not surprised?
3204 PH/15 If the first argument of "${if match_address" was not empty, but did not
3205 contain an "@" character, Exim crashed. Now it writes a panic log message
3206 and treats the condition as false.
3208 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3210 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3211 extension "envelope-auth". The code is finished and in agreement with
3212 other implementations, but there is no documentation so far and in fact,
3213 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3214 not changing the active code.
3216 Print executed "if" and "elsif" statements when debugging is used. This
3217 helps a great deal to understand what a filter does.
3219 Document more things not specified clearly in RFC3028. I had all this
3220 sorted out, when out of a sudden new issues came to my mind. Oops."
3222 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3225 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3226 canonical form (as documented). However, after a host name lookup from
3227 the IP address, check_host() was doing a simple string comparison with
3228 addresses acquired from the DNS when checking that the found name did
3229 have the original IP as one of its addresses. Since any found IPv6
3230 addresses are likely to be in abbreviated form, the comparison could
3231 fail. Luckily, there already exists a function for doing the comparison
3232 by converting both addresses to binary, so now that is used instead of
3233 the text comparison.
3235 PH/20 There was another similar case to PH/19, when a complete host name was
3236 given in a host list; looking up its IP address could give an abbreviated
3237 form, whereas the current host's name might or might not be abbreviated.
3238 The same fix has been applied.
3244 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3245 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3248 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3249 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3251 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3253 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3254 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3255 Also, for "Auto-Submitted", treat anything other than "no" as
3256 non-personal, in accordance with RFC 3834. (Previously it treated
3257 anything starting "auto-" as non-personal.)
3259 TF/01 The control=submission/name=... option had a problem with syntax
3260 errors if the name included a slash character. The /name= option
3261 now slurps the rest of the string, so it can include any characters
3262 but it must come last in the list of options (after /sender_retain
3265 PH/05 Some modifications to the interface to the fake nameserver for the new
3273 TK/01 Added the "success_on_redirect" address verification option. See
3274 NewStuff for rationale and an example.
3276 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3278 PH/02 Patch to exigrep to allow it to work on syslog lines.
3280 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3281 fread() to skip over the body file's header line, because in Cygwin the
3282 header line is locked and is inaccessible.
3284 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3285 co-exist for some time) to make it clear that it is the Exim ID that is
3286 referenced, not the Message-ID: header line.
3288 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3289 string_format() function, because snprintf() does not exist on all
3292 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3293 ${stat: expansion item.
3295 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3296 protocol synchronization error", to keep the pedants happy.
3298 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3299 well as for IRIX systems, when gcc is being used. See the host.c source
3302 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3304 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3307 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3308 transport and to the smtp transport in LMTP mode.
3310 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3312 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3313 run a filter in a subprocess. This could lead to confusion in subsequent
3314 lookups in the parent process. There should also be a search_tidyup() at
3315 the end of the subprocess.
3317 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3318 only if the host matched helo_try_verify_hosts, which caused the
3319 verification to occur when the EHLO/HELO command was issued. The ACL just
3320 tested the remembered result. Now, if a previous verification attempt has
3321 not happened, "verify = helo" does it there and then.
3323 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3325 TK/03 Fix log output including CR from clamd.
3327 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3328 exist provoked a memory error which could cause a segfault.
3330 PH/15 Installed PCRE 6.2
3332 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3334 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3335 of the problem. Specifically, suggested +O2 rather than +O1 for the
3338 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3340 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3341 record, the port was not correctly logged when the outgoing_port log
3342 selector was set (it logged the transort's default port).
3344 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3345 fallback_hosts, and "hosts" in the smtp transport.
3347 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3348 host errors such as "Connection refused".
3350 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3351 authentication with radiusclient 0.4.9:
3353 - Error returned from rc_read_config was caught wrongly
3354 - Username/password not passed on to radius server due to wrong length.
3356 The presumption is that some radiusclient API changes for 4.51/PH/17
3357 were not taken care of correctly. The code is still untested by me (my
3358 Linux distribution still has 0.3.2 of radiusclient), but it was
3359 contributed by a Radius user.
3361 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3362 expanding the "port" option of the smtp transport.
3364 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3365 while reading a MIME header. Thanks to Tom Hughes for a patch.
3367 PH/24 Include config.h inside local_scan.h so that configuration settings are
3370 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3371 that in an ACL for RCPT (for example), you can examine exactly what was
3374 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3375 commands, but it was not correctly comparing the address with the actual
3376 client host address. Thus, it would show the EHLO address in Received:
3377 header lines when this was not necessary.
3379 PH/27 Added the % operator to ${eval:}.
3381 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3382 it should be ignoring failures (because with -C, for example, it has lost
3383 privilege). It wasn't ignoring creation failures other than "already
3386 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3389 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3390 because an input line was too long, either on its own, or by virtue of
3391 too many continuations, the temporary file was not being removed, and the
3392 return code was incorrect.
3394 PH/31 Missing "BOOL" in function definition in filtertest.c.
3396 PH/32 Applied Sieve patches from the maintainer.
3398 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3400 PH/33 Added "verify = not_blind".
3402 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3403 Local/Makefile (with some defaults set). These are used in built scripts
3404 such as exicyclog, but they have never been used in the exim_install
3405 script (though there are many overriding facilities there). I have
3406 arranged that the exim_install script now takes note of these two
3409 PH/35 Installed configuration files for Dragonfly.
3411 PH/36 When a locally submitted message by a trusted user did not contain a
3412 From: header, and the sender address was obtained from -f or from an SMTP
3413 MAIL command, and the trusted user did not use -F to supply a sender
3414 name, $originator_name was incorrectly used when constructing a From:
3415 header. Furthermore, $originator_name was used for submission mode
3416 messages from external hosts without From: headers in a similar way,
3417 which is clearly wrong.
3419 PH/37 Added control=suppress_local_fixups.
3421 PH/38 When log_selector = +received_sender was set, and the addition of the
3422 sender made the log line's construction buffer exactly full, or one byte
3423 less than full, an overflow happened when the terminating "\n" was
3426 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3427 when the result of a list match is failure because a DNS lookup failed.
3429 PH/40 RM_COMMAND is now used in the building process.
3431 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3432 the "build-* directories that it finds.
3434 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3435 address in a domain literal was a prefix of an interface address.
3437 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3438 when verifying a sender address, unless rewrite_headers is false.
3440 PH/44 Wrote a long comment about why errors_to addresses are verified as
3441 recipients, not senders.
3443 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3444 the ratelimit ACL was added.
3446 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3448 PH/46 Added extra information about PostgreSQL errors to the error string.
3450 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3451 avoiding the need to install special zones in a real server. This is
3452 backwards compatible; if it can't find the fake resolver, it drops back.
3453 Thus, both old and new test suites can be run.
3455 TF/02 Added util/ratelimit.pl
3457 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3458 clock is set back in time.
3460 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3461 Candler <B.Candler@pobox.com>.
3463 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3464 OK for addresses that are the result of redirections.
3466 PH/48 A number of further additions for the benefit of the new test suite,
3467 including a fake gethostbyname() that interfaces to the fake DNS resolver
3470 TF/06 The fix for widen_domains has also been applied to qualify_single and
3471 search_parents which are the other dnslookup options that can cause
3474 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3477 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3479 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3480 service" if it encountered an MX record with an empty target hostname.
3481 The message is now "an MX or SRV record indicated no SMTP service".
3483 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3484 if the DNS of the sending site is misconfigured. This is quite a
3485 common situation. This change restores the behaviour of treating a
3486 helo verification defer as a failure.
3488 PH/51 If self=fail was set on a router, the bounce message did not include the
3489 actual error message.
3495 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3497 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3498 command itself ended in error, the underlying message about the transport
3499 filter timeout was being overwritten with the pipe command error. Now the
3500 underlying error message should be appended to the second error message.
3502 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3504 PH/02 Reduce the amount of output that "make" produces by default. Full output
3505 can still be requested.
3507 PH/03 The warning log line about a condition test deferring for a "warn" verb
3508 was being output only once per connection, rather than after each
3509 occurrence (because it was using the same function as for successful
3510 "warn" verbs). This seems wrong, so I have changed it.
3512 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3513 it should not have, which might have caused a crash in the right
3514 circumstances, but probably never did.
3516 PH/04 Installed a modified version of Tony Finch's patch to make submission
3517 mode fix the return path as well as the Sender: header line, and to
3518 add a /name= option so that you can make the user's friendly name appear
3521 TF/03 Added the control = fakedefer ACL modifier.
3523 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3524 Mark Lowes for thorough testing.
3526 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3528 TK/03 Merged latest SRS patch from Miles Wilton.
3530 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3531 with the definition in sysexits.h (which is #included earlier).
3532 Fortunately, Exim does not actually use EX_OK. The code used to try to
3533 preserve the sysexits.h value, by assumimg that macro definitions were
3534 scanned for macro replacements. I have been disabused of this notion,
3535 so now the code just undefines EX_OK before #including unistd.h.
3537 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3538 in the smtp transport. When a block could not be written in a single
3539 write() function, the timeout was being re-applied to each part-write.
3540 This seems wrong - if the receiver was accepting one byte at a time it
3541 would take for ever. The timeout is now adjusted when this happens. It
3542 doesn't have to be particularly precise.
3544 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3545 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3547 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3548 without a domain if the check to <postmaster@domain> fails.
3550 SC/01 Eximstats: added -xls and the ability to specify output files
3551 (patch written by Frank Heydlauf).
3553 SC/02 Eximstats: use FileHandles for outputing results.
3555 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3557 SC/04 Eximstats: fixed display of large numbers with -nvr option
3559 SC/05 Eximstats: fixed merging of reports with empty tables.
3561 SC/06 Eximstats: added the -include_original_destination flag
3563 SC/07 Eximstats: removed tabs and trailing whitespace.
3565 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3567 TK/06 MBOX spool code: Add real "From " MBOX separator line
3568 so the .eml file is really in mbox format (even though
3569 most programs do not really care). Patch from Alex Miller.
3571 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3572 The latter is generated from $received_to and is only set if the
3573 message has one envelope recipient. SA can use these headers,
3574 obviously out-of-the-box. Patch from Alex Miller.
3576 PH/08 The ${def test on a variable was returning false if the variable's
3577 value was "0", contrary to what the specification has always said!
3578 The result should be true unless the variable is empty.
3580 PH/09 The syntax error of a character other than { following "${if
3581 def:variable_name" (after optional whitespace) was not being diagnosed.
3582 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3583 accidental colon was present, for example, could give incorrect results.
3585 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3586 result is used (not including appendfile, where other changes are about
3589 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3590 This involved changing a lot of size variables from int to off_t. It
3591 should work with maildirs and everything.
3593 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3594 spamd dying while we are connected to it.
3596 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3599 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3600 function for BDB 4.3.
3602 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3604 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3605 the sources for PCRE have been re-arranged and more files are now
3608 PH/15 The code I had for printing potentially long long variables in PH/11
3609 above was not the best (it lost precision). The length of off_t variables
3610 is now inspected at build time, and an appropriate printing format (%ld
3611 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3612 to be "long long int" or "long int". This is needed for the internal
3613 formatting function string_vformat().
3615 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3616 the configuration file to be ":syslog", then the script "guesses" where
3617 the logs files are, rather than using the compiled in default. In our
3618 case the guess is not the same as the compiled default, so the script
3619 suddenly stopped working when I started to use syslog. The patch checks
3620 to see if log_file_path is "". If so, it attempts to read it from exim
3621 with no configuration file to get the compiled in version, before it
3622 falls back to the previous guessing code."
3624 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3625 implementing BATV in an Exim configuration. See NewStuff for the gory
3628 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3629 Makefile that are specific to HP-UX.
3631 PH/18 If the "use_postmaster" option was set for a recipient callout together
3632 with the "random" option, the postmaster address was used as the MAIL
3633 FROM address for the random test, but not for the subsequent recipient
3634 test. It is now used for both.
3636 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3637 patch removes a few documentation additions to RFC 3028, because the
3638 latest draft now contains them. It adds the new en;ascii-case comparator
3639 and a new error check for 8bit text in MIME parts. Comparator and
3640 require names are now matched exactly. I enabled the subaddress
3641 extension, but it is not well tested yet (read: it works for me)."
3643 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3644 rework some of the code of TK/09 above to avoid the hardwired use of
3645 "%lld" and "long long". Replaced the call to snprintf() with a call to
3648 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3649 records point to non-existent hosts", "retry timeout exceeded", and
3650 "retry time not reached for any host after a long failure period".
3652 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3653 experimental DomainKeys support:
3655 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3656 (2) On an error such as an illegally used "control", the wrong name for
3657 the control was given.
3659 These problems did NOT occur unless DomainKeys support was compiled.
3661 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3663 PH/24 Added ${if match_ip condition.
3665 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3666 databases so that it will be absolutely obvious if a crash occurs in the
3667 DB library. This is a regular occurrence (often caused by mis-matched
3670 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3671 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3672 picked up on a user's system that detects such things. There doesn't seem
3673 to be a gcc warning option for this - only an attribute that has to be
3674 put on the function's prototype. It seems that in Fedora Core 4 they have
3675 set this on a number of new functions. No doubt there will be more in due
3678 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3679 specify a transport. However, if an address that was verified by such a
3680 router was the subject of a callout, Exim crashed because it tried to
3681 read the rcpt_include_affixes from the non-existent transport. Now it
3682 just assumes that the setting of that option is false. This bug was
3683 introduced by 4.51/PH/31.
3685 PH/28 Changed -d+all to exclude +memory, because that information is very
3686 rarely of interest, but it makes the output a lot bigger. People tend to
3687 do -d+all out of habit.
3689 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3690 code in os-type was giving problems when libc.so lives in lib64, like on
3693 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3694 aren't the modern standard, and it seems that some systems' include files
3695 don't always have them. Exim was already checking for some of the newer
3696 ones like T_AAAA, and defining it itself. I've added checks for all the
3697 record types that Exim uses.
3699 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3700 not automatically generating a new one, as it is supposed to. This
3701 prevented TLS from working. If the file did exist, but contained invalid
3702 data, a new version was generated, as expected. It was only the case of a
3703 non-existent file that was broken.
3705 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3706 with a change in libdomainkeys > 0.64.
3708 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3709 from DNS. If the selector record carries the flag, it now has
3710 precedence over the domain-wide flag.
3712 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3714 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3715 the use of a port name that isn't defined in /etc/services) occurred, the
3716 message was deferred as in a normal delivery, and thus remained on the
3717 spool, instead of being failed because of the mua_wrapper setting. This
3718 is now fixed, and I tidied up some of the mua_wrapper messages at the
3721 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3722 the messages in a hash of arrays rather than using individual hashes.
3723 This is a bit cleaner and results in dramatic memory savings, albeit
3724 at a slight CPU cost.
3726 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3727 as requested by Marc Sherman.
3729 SC/10 Eximstats: added histograms for user specified patterns as requested
3732 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3734 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3735 fopen() in the content-scanning modules that did not already have it.
3741 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3742 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3744 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3746 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3748 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3749 to test Sieve filters that use "vacation".
3751 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3752 that changes the way the GnuTLS parameters are stored in the cache file.
3753 The new format can be generated externally. For backward compatibility,
3754 if the data in the cache doesn't make sense, Exim assumes it has read an
3755 old-format file, and it generates new data and writes a new file. This
3756 means that you can't go back to an older release without removing the
3759 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3760 work if there are any delivery delays because "one_time" forces the
3761 parent to be marked "delivered", so its unseen clone is never tried
3762 again. For this reason, Exim now forbids the simultaneous setting of
3765 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3766 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3767 entry if you want to know the details. The fix, however, neglected to
3768 consider the case when local delivery batching is involved. The test for
3769 "previously delivered" was not happening when checking to see if an
3770 address could be batched with a previous (undelivered) one; under
3771 certain circumstances this could lead to multiple deliveries to the same
3774 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3775 in its include files, and this causes problems building Exim.
3777 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3778 header_syntax) but Exim was just ignoring anything given after a slash.
3779 In particular, this caused confusion with an attempt to use "verify =
3780 reverse_host_lookup/defer_ok". An error is now given when options are
3781 supplied for verify items that do not have them. (Maybe reverse_host_
3782 lookup should have a defer_ok option, but that's a different point.)
3784 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3785 defined by RFC 821) to 2048, because there were problems with some AUTH
3786 commands, and RFC 1869 says the size should be increased for extended
3787 SMTP commands that take arguments.
3789 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3792 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3793 "unknown" error; now it says that the functionality isn't in the binary.
3795 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3796 an address' error message when a string expansion fails (syntax or
3797 whatever). Otherwise the password may appear in the log. Following change
3798 PH/42 below, there is no longer a chance of it appearing in a bounce
3801 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3803 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3804 of its addresses was ever tried. (Bugzilla bug #2.)
3806 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3807 the result incorrectly in the debug output. (It correctly added a newline
3808 to what was transported.)
3810 TF/01 Added $received_time.
3812 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3813 commented out examples of how to interface to a virus scanner and to
3814 SpamAssassin. Also added commented examples of av_scanner and
3815 spamd_address settings.
3817 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3818 and controls are allowed in which ACLs. There were a couple of minor
3819 errors. Some of the entries in the conditions table (which is a table of
3820 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3821 negation of where the condition IS allowed.
3823 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3825 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3826 header file does not have a version number, so I've had to invent a new
3827 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3828 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3829 radiusclient), but it was contributed by a Radius user.
3831 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3832 files or directories, for OpenSSL.
3834 PH/19 When an Exim process that is running as root has to create an Exim log
3835 file, it does so in a subprocess that runs as exim:exim so as to get the
3836 ownership right at creation (otherwise, other Exim processes might see
3837 the file with the wrong ownership). There was no test for failure of this
3838 fork() call, which would lead to the process getting stuck as it waited
3839 for a non-existent subprocess. Forks do occasionally fail when resources
3840 run out. I reviewed all the other calls to fork(); they all seem to check
3843 PH/20 When checking for unexpected SMTP input at connect time (before writing
3844 the banner), Exim was not dealing correctly with a non-positive return
3845 from the read() function. If the client had disconnected by this time,
3846 the result was a log entry for a synchronization error with an empty
3847 string after "input=" when read() returned zero. If read() returned -1
3848 (an event I could not check), uninitialized data bytes were printed.
3849 There were reports of junk text (parts of files, etc) appearing after
3852 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3854 PH/22 Added support for macro redefinition, and (re)definition in between
3855 driver and ACL definitions.
3857 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3858 forgetting to use the resulting value; it was using the unexpanded value.
3860 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3861 hadn't been configured. The fix is from Juergen Kreileder, who
3862 understands it better than I do:
3864 "Here's what I see happening with three configured cyrus_sasl
3865 authenticators configured (plain, login, cram-md5):
3867 On startup auth_cyrus_sasl_init() gets called for each of these.
3868 This means three calls to sasl_listmech() without a specified mech_list.
3869 => SASL tests which mechs of all available mechs actually work
3870 => three warnings about OTP not working
3871 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3873 With the patch, sasl_listmech() also gets called three times. But now
3874 SASL's mech_list option is set to the server_mech specified in the the
3875 authenticator. Or in other words, the answer from sasl_listmech()
3876 gets limited to just the mech you're testing for (which is different
3878 => the return list contains just 'plain' or 'login', 'cram-md5' or
3879 nothing depending on the value of ob->server_mech.
3881 I've just tested the patch: Authentication still works fine,
3882 unavailable mechs specified in the exim configuration are still
3883 caught, and the auth.log warnings about OTP are gone."
3885 PH/25 When debugging is enabled, the contents of the command line are added
3886 to the debugging output, even when log_selector=+arguments is not
3889 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3890 answer is "GNU", and only if the return is "GNU/something" is the answer
3893 PH/27 $acl_verify_message is now set immediately after the failure of a
3894 verification in an ACL, and so is available in subsequent modifiers. In
3895 particular, the message can be preserved by coding like this:
3897 warn !verify = sender
3898 set acl_m0 = $acl_verify_message
3900 Previously, $acl_verify_message was set only while expanding "message"
3901 and "log_message" when a very denied access.
3903 PH/28 Modified OS/os.c-Linux with
3905 -#ifndef OS_LOAD_AVERAGE
3906 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3908 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3909 nomenclature these days.)
3911 PH/29 Installed patch from the Sieve maintainer that adds the options
3912 sieve_useraddress and sieve_subaddress to the redirect router.
3914 PH/30 In these circumstances:
3915 . Two addresses routed to the same list of hosts;
3916 . First host does not offer TLS;
3917 . First host accepts first address;
3918 . First host gives temporary error to second address;
3919 . Second host offers TLS and a TLS session is established;
3920 . Second host accepts second address.
3921 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3922 and peerdn, if requested) that were in fact used only for the second
3925 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3926 attention to any local part prefix or suffix that was matched by the
3927 router that accepted the address. It now behaves in the same way as it
3928 does for delivery: the affixes are removed from the local part unless
3929 rcpt_include_affixes is set on the transport.
3931 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3932 timeout during the DATA phase of an incoming message.
3934 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3935 applied a patch sanctioned by the Sieve maintainer.
3937 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3938 the uid or gid is negative. A case of a negative gid caused this to be
3939 noticed. The fix allows for either to be negative.
3941 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3942 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3945 PH/36 The change PH/12 above was broken. Fixed it.
3947 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3948 the grounds that routing the same address twice would always produce the
3949 same answer. This might have been true once, but it is certainly no
3950 longer true now. Routing a child address may depend on the previous
3951 routing that produced that child. Some complicated redirection strategies
3952 went wrong when messages had multiple recipients, and made Exim's
3953 behaviour dependent on the order in which the addresses were given.
3955 I have moved the duplicate checking until after the routing is complete.
3956 Exim scans the addresses that are assigned to local and remote
3957 transports, and removes any duplicates. This means that more work will be
3958 done, as duplicates will always all be routed, but duplicates are
3959 presumably rare, so I don't expect this is of any significance.
3961 For deliveries to pipes, files, and autoreplies, the duplicate checking
3962 still happens during the routing process, since they are not going to be
3965 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3966 It corrects a timeout issue with spamd. This is Ian's comment: "The
3967 background is that sometimes spamd either never reads data from a
3968 connection it has accepted, or it never writes response data. The exiscan
3969 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3970 blindly assumes that writes won't block so it may never time out."
3972 PH/39 Allow G after quota size as well as K and M.
3974 PH/40 The value set for $authenticated_id in an authenticator may not contain
3975 binary zeroes or newlines because the value is written to log lines and
3976 to spool files. There was no check on this. Now the value is run through
3977 the string_printing() function so that such characters are converted to
3978 printable escape sequences.
3980 PH/41 $message_linecount is a new variable that contains the total number of
3981 lines in the message. Compare $body_linecount, which is the count for the
3984 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3985 bounce and delay warning messages, except in certain special cases, which
3988 (a) An SMTP error message from a remote host;
3989 (b) A message specified in a :fail: redirection;
3990 (c) A message specified in a "fail" command in a system filter;
3991 (d) A message specified in a FAIL return from the queryprogram router;
3992 (e) A message specified by the cannot_route_message router option.
3994 In these cases only, Exim does include the error details in bounce and
3995 warning messages. There are also a few cases where bland messages such
3996 as "unrouteable address" or "local delivery error" are given.
3998 PH/43 $value is now also set for the "else" part of a ${run expansion.
4000 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4001 being worked on, but at least Exim now implements the latest version to
4004 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4005 process to complete was treated as a delivery failure, a timeout while
4006 writing the message to the pipe was logged, but erroneously treated as a
4007 successful delivery. Such timeouts include transport filter timeouts. For
4008 consistency with the overall process timeout, these timeouts are now
4009 treated as errors, giving rise to delivery failures by default. However,
4010 there is now a new Boolean option for the pipe transport called
4011 timeout_defer, which, if set TRUE, converts the failures into defers for
4012 both kinds of timeout. A transport filter timeout is now identified in
4015 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4016 systems where "make" and "gmake" are different, calling "gmake" at top
4017 level broke things. I've arranged for the value of $(MAKE) to be passed
4018 from the Makefile to this script so that it can call the same version of
4022 A note about Exim versions 4.44 and 4.50
4023 ----------------------------------------
4025 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4026 changes of various kinds. As a consequence, a big documentation update was
4027 needed. This delayed the release for rather longer than seemed good, especially
4028 in the light of a couple of (minor) security issues. Therefore, the changes
4029 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4030 release. So 4.44 and 4.50 are in effect two different branches that both start
4033 I have left the 4.50 change log unchanged; it contains all the changes since
4034 4.43. The change log for 4.44 is below; many of its items are identical to
4035 those for 4.50. This seems to be the most sensible way to preserve the
4036 historical information.
4042 1. Minor wording change to the doc/README.SIEVE file.
4044 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4045 computation of the current number of files was incorrect.
4047 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4048 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4051 4. Give more explanation in the error message when the command for a transport
4052 filter fails to execute.
4054 5. There are several places where Exim runs a non-Exim command in a
4055 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4056 was being done only for the command run by the queryprogram router. It is
4057 now done for all such subprocesses. The other cases are: ${run, transport
4058 filters, and the commands run by the lmtp and pipe transports.
4060 6. Added CONFIGURE_GROUP build-time option.
4062 7. Some older OS have a limit of 256 on the maximum number of file
4063 descriptors. Exim was using setrlimit() to set 1000 as a large value
4064 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4065 systems. I've change it so that if it can't get 1000, it tries for 256.
4067 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4068 was an oversight, and furthermore, ever since the addition of extra
4069 controls (e.g. 4.43/32), the checks on when to allow different forms of
4070 "control" were broken. There should now be diagnostics for all cases when a
4071 control that does not make sense is encountered.
4073 9. Added the /retain_sender option to "control=submission".
4075 10. $recipients is now available in the predata ACL (oversight).
4077 11. Tidy the search cache before the fork to do a delivery from a message
4078 received from the command line. Otherwise the child will trigger a lookup
4079 failure and thereby defer the delivery if it tries to use (for example) a
4080 cached ldap connection that the parent has called unbind on.
4082 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4083 of $address_data from the recipient verification was clobbered by the
4084 sender verification.
4086 13. The value of address_data from a sender verification is now available in
4087 $sender_address_data in subsequent conditions in the ACL statement.
4089 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4091 15. Added a new option "connect=<time>" to callout options, to set a different
4094 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4095 was its contents. (It was OK if the option was not defined at all.)
4097 17. A "Completed" log line is now written for messages that are removed from
4098 the spool by the -Mrm option.
4100 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4101 information about exactly what failed.
4103 19. Added -dd to debug only the daemon process.
4105 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4106 handles timeouts, both on the server side and network timeouts. Renamed the
4107 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4109 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4110 It is now set to "smtps".
4112 22. $host_address is now set to the target address during the checking of
4113 ignore_target_hosts.
4115 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4116 being passed; this would have caused $sender_host_name to have been used if
4117 matching the list had actually called for a host name (not very likely,
4118 since this list is usually IP addresses). A host name is now passed as
4121 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4122 code that specifies a non-restarting handler (typically sigaction() in
4123 modern systems) in an attempt to fix a rare and obscure crash bug.
4125 25. Narrowed the window for a race in the daemon that could cause it to ignore
4126 SIGCHLD signals. This is not a major problem, because they are used only to
4127 wake it up if nothing else does.
4129 26. A malformed maildirsize file could cause Exim to calculate negative values
4130 for the mailbox size or file count. Odd effects could occur as a result.
4131 The maildirsize information is now recalculated if the size or filecount
4134 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4135 support for a long time. Removed HAVE_SYS_VFS_H.
4137 28. Installed the latest version of exipick from John Jetmore.
4139 29. In an address list, if the pattern was not a regular expression, an empty
4140 subject address (from a bounce message) matched only if the pattern was an
4141 empty string. Non-empty patterns were not even tested. This was the wrong
4142 because it is perfectly reasonable to use an empty address as part of a
4143 database query. An empty address is now tested by patterns that are
4144 lookups. However, all the other forms of pattern expect the subject to
4145 contain a local part and a domain, and therefore, for them, an empty
4146 address still always fails if the pattern is not itself empty.
4148 30. Exim went into a mad DNS loop when attempting to do a callout where the
4149 host was specified on an smtp transport, and looking it up yielded more
4150 than one IP address.
4152 31. Re-factored the code for checking spool and log partition space into a
4153 function that finds that data and another that does the check. The former
4154 is then used to implement four new variables: $spool_space, $log_space,
4155 $spool_inodes, and $log_inodes.
4157 32. The RFC2047 encoding function was originally intended for short strings
4158 such as real names; it was not keeping to the 75-character limit for
4159 encoded words that the RFC imposes. It now respects the limit, and
4160 generates multiple encoded words if necessary. To be on the safe side, I
4161 have increased the buffer size for the ${rfc2047: expansion operator from
4164 33. It is now permitted to omit both strings after an "if" condition; if the
4165 condition is true, the result is "true". As before, when the second string
4166 is omitted, a false condition yields an empty string. This makes it less
4167 cumbersome to write custom ACL and router conditions.
4169 34. Failure to deliver a bounce message always caused it to be frozen, even if
4170 there was an errors_to setting on the router. The errors_to setting is now
4173 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4174 canonical form (fully expanded) before being placed in
4175 $sender_host_address.
4177 36. The table in the code that translates DNS record types into text (T_A to
4178 "A" for instance) was missing entries for NS and CNAME. It is just possible
4179 that this could have caused confusion if both these types were looked up
4180 for the same domain, because the text type is used as part of Exim's
4181 per-process caching. But the chance of anyone hitting this buglet seems
4184 37. The dnsdb lookup has been extended in a number of ways.
4186 (1) There is a new type, "zns", which walks up the domain tree until it
4187 finds some nameserver records. It should be used with care.
4189 (2) There is a new type, "mxh", which is like "mx" except that it returns
4190 just the host names, not the priorities.
4192 (3) It is now possible to give a list of domains (or IP addresses) to be
4193 looked up. The behaviour when one of the lookups defers can be
4194 controlled by a keyword.
4196 (4) It is now possible to specify the separator character for use when
4197 multiple records are returned.
4199 38. The dnslists ACL condition has been extended: it is now possible to supply
4200 a list of IP addresses and/or domains to be looked up in a particular DNS
4203 39. Added log_selector=+queue_time_overall.
4205 40. When running the queue in the test harness, wait just a tad after forking a
4206 delivery process, to get repeatability of debugging output.
4208 41. Include certificate and key file names in error message when GnuTLS fails
4209 to set them up, because the GnuTLS error message doesn't include the name
4210 of the failing file when there is a problem reading it.
4212 42. Allow both -bf and -bF in the same test run.
4214 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4216 44. The "Exiscan patch" is now merged into the mainline Exim source.
4218 45. Sometimes the final signoff response after QUIT could fail to get
4219 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4220 < 0 before doing a fflush(). This bug looks as though it goes back to the
4221 introduction of TLS in release 3.20, but "sometimes" must have been rare
4222 because the tests only now provoked it.
4224 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4225 (this can affect the format of dates).
4227 47. exim_tidydb, when checking for the continued existence of a message for
4228 which it has found a message-specific retry record, was not finding
4229 messages that were in split spool directories. Consequently, it was
4230 deleting retry records that should have stayed in existence.
4232 48. Steve fixed some bugs in eximstats.
4234 49. The SPA authentication driver was not abandoning authentication and moving
4235 on to the next authenticator when an expansion was forced to fail,
4236 contradicting the general specification for all authenticators. Instead it
4237 was generating a temporary error. It now behaves as specified.
4239 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4240 (the order specifies the preference for clients). The order is now AES256,
4241 AES128, 3DES, ARCFOUR128.
4243 51. Small patch to Sieve code - explicitly set From: when generating an
4246 52. Exim crashed if a remote delivery caused a very long error message to be
4247 recorded - for instance if somebody sent an entire SpamAssassin report back
4248 as a large number of 550 error lines. This bug was coincidentally fixed by
4249 increasing the size of one of Exim's internal buffers (big_buffer) that
4250 happened as part of the Exiscan merge. However, to be on the safe side, I
4251 have made the code more robust (and fixed the comments that describe what
4254 53. Now that there can be additional text after "Completed" in log lines (if
4255 the queue_time_overall log selector is set), a one-byte patch to exigrep
4256 was needed to allow it to recognize "Completed" as not the last thing in
4259 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4260 patch that reportedly fixes this has been added. I am not expert enough to
4261 create a test for it. This is what the patch creator wrote:
4263 "I found a little strange behaviour of ldap code when working with
4264 Windows 2003 AD Domain, where users was placed in more than one
4265 Organization Units. When I tried to give exim partial DN, the exit code
4266 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4267 But simultaneously result of request was absolutely normal ldap result,
4268 so I produce this patch..."
4270 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4271 REFERENCE, so I have modified the code to exclude the patch when that macro
4274 55. Some experimental protocols are using DNS PTR records for new purposes. The
4275 keys for these records are domain names, not reversed IP addresses. The
4276 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4277 leaves it alone. Component reversal etc. now happens only for IP addresses.
4280 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4282 57. Double the size of the debug message buffer (to 2048) so that more of very
4283 long debug lines gets shown.
4285 58. The exicyclog utility now does better if the number of log files to keep
4286 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4288 59. Two changes related to the smtp_active_hostname option:
4290 (1) $smtp_active_hostname is now available as a variable.
4291 (2) The default for smtp_banner uses $smtp_active_hostname instead
4292 of $primary_hostname.
4294 60. The host_aton() function is supposed to be passed a string that is known
4295 to be a valid IP address. However, in the case of IPv6 addresses, it was
4296 not checking this. This is a hostage to fortune. Exim now panics and dies
4297 if the condition is not met. A case was found where this could be provoked
4298 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4299 components; fortuitously, this particular loophole had already been fixed
4300 by change 4.50/55 above.
4302 If there are any other similar loopholes, the new check in host_aton()
4303 itself should stop them being exploited. The report I received stated that
4304 data on the command line could provoke the exploit when Exim was running as
4305 exim, but did not say which command line option was involved. All I could
4306 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4307 running as the user.
4310 61. There was a buffer overflow vulnerability in the SPA authentication code
4311 (which came originally from the Samba project). I have added a test to the
4312 spa_base64_to_bits() function which I hope fixes it.
4315 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4316 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4318 63. The daemon start-up calls getloadavg() while still root for those OS that
4319 need the first call to be done as root, but it missed one case: when
4320 deliver_queue_load_max is set with deliver_drop_privilege. This is
4321 necessary for the benefit of the queue runner, because there is no re-exec
4322 when deliver_drop_privilege is set.
4324 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4325 This has been fixed.
4327 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4328 was in use, was not putting the data itself into the right store pool;
4329 consequently, it could be overwritten for a subsequent message in the same
4330 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4333 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4335 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4336 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4337 the function were treating the return as a boolean value, which happened to
4338 work because 0=false and not-0=true, but is not correct code.
4340 68. The host_aton() function was not handling scoped IPv6 addresses (those
4341 with, for example, "%eth0" on the end) correctly.
4343 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4344 negated items (that is, ~something) in unsigned ints. Some compilers
4345 apparently mutter when there is no cast.
4347 70. If an address verification called from an ACL failed, and did not produce a
4348 user-specific message (i.e. there was only a "system" message), nothing was
4349 put in $acl_verify_message. In this situation, it now puts the system
4352 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4353 session; change 4.31/43 added the unwanted input to the log line - except
4354 that it did not do this in the start of session case. It now does.
4356 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4357 This is wrong and can cause the other end to generate a synchronization
4358 error if it is another Exim or anything else that does the synchronization
4359 check. A QUIT command is no longer sent after a timeout.
4361 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4362 during host lookups.
4364 74. The defer_ok option of callout verification was not working if it was used
4365 when verifying addresses in header lines, that is, for this case:
4367 verify = header_sender/callout=defer_ok
4369 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4370 those file descriptors could be used for SMTP connections. If anything
4371 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4372 could be sent to the SMTP client, causing chaos. The daemon now opens
4373 stdin, stdout, and stderr to /dev/null when it puts itself into the
4376 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4377 log by default. The user can override this with suitable Perl magic.
4379 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4380 the log message when discard triggers, was not working for the DATA ACL or
4381 for the non-SMTP ACL.
4383 78. Error message wording change in sieve.c.
4385 79. If smtp_accept_max_per_host was set, the number of connections could be
4386 restricted to fewer than expected, because the daemon was trying to set up
4387 a new connection before checking whether the processes handling previous
4388 connections had finished. The check for completed processes is now done
4389 earlier. On busy systems, this bug wouldn't be noticed because something
4390 else would have woken the daemon, and it would have reaped the completed
4393 80. If a message was submitted locally by a user whose login name contained one
4394 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4395 It caused a spool format error. I have fixed the spool reading code. A
4396 related problem was that the "from" clause in the Received: line became
4397 illegal because of the space(s). It is now covered by ${quote_local_part.
4399 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4402 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4404 83. Patch from Sieve maintainer to fix unterminated string problem in
4405 "vacation" handling.
4407 84. Some minor changes to the Linux configuration files to help with other
4408 OS variants using glibc.
4410 85. One more patch for Sieve to update vacation handling to latest spec.
4413 ----------------------------------------------------
4414 See the note above about the 4.44 and 4.50 releases.
4415 ----------------------------------------------------
4421 1. Change 4.43/35 introduced a bug that caused file counts to be
4422 incorrectly computed when quota_filecount was set in an appendfile
4425 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4426 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4429 3. Give more explanation in the error message when the command for a transport
4430 filter fails to execute.
4432 4. There are several places where Exim runs a non-Exim command in a
4433 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4434 was being done only for the command run by the queryprogram router. It is
4435 now done for all such subprocesses. The other cases are: ${run, transport
4436 filters, and the commands run by the lmtp and pipe transports.
4438 5. Some older OS have a limit of 256 on the maximum number of file
4439 descriptors. Exim was using setrlimit() to set 1000 as a large value
4440 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4441 systems. I've change it so that if it can't get 1000, it tries for 256.
4443 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4444 was an oversight, and furthermore, ever since the addition of extra
4445 controls (e.g. 4.43/32), the checks on when to allow different forms of
4446 "control" were broken. There should now be diagnostics for all cases when a
4447 control that does not make sense is encountered.
4449 7. $recipients is now available in the predata ACL (oversight).
4451 8. Tidy the search cache before the fork to do a delivery from a message
4452 received from the command line. Otherwise the child will trigger a lookup
4453 failure and thereby defer the delivery if it tries to use (for example) a
4454 cached ldap connection that the parent has called unbind on.
4456 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4457 of $address_data from the recipient verification was clobbered by the
4458 sender verification.
4460 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4461 was its contents. (It was OK if the option was not defined at all.)
4463 11. A "Completed" log line is now written for messages that are removed from
4464 the spool by the -Mrm option.
4466 12. $host_address is now set to the target address during the checking of
4467 ignore_target_hosts.
4469 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4470 being passed; this would have caused $sender_host_name to have been used if
4471 matching the list had actually called for a host name (not very likely,
4472 since this list is usually IP addresses). A host name is now passed as
4475 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4476 code that specifies a non-restarting handler (typically sigaction() in
4477 modern systems) in an attempt to fix a rare and obscure crash bug.
4479 15. Narrowed the window for a race in the daemon that could cause it to ignore
4480 SIGCHLD signals. This is not a major problem, because they are used only to
4481 wake it up if nothing else does.
4483 16. A malformed maildirsize file could cause Exim to calculate negative values
4484 for the mailbox size or file count. Odd effects could occur as a result.
4485 The maildirsize information is now recalculated if the size or filecount
4488 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4489 support for a long time. Removed HAVE_SYS_VFS_H.
4491 18. Updated exipick to current release from John Jetmore.
4493 19. Allow an empty sender to be matched against a lookup in an address list.
4494 Previously the only cases considered were a regular expression, or an
4497 20. Exim went into a mad DNS lookup loop when doing a callout where the
4498 host was specified on the transport, if the DNS lookup yielded more than
4501 21. The RFC2047 encoding function was originally intended for short strings
4502 such as real names; it was not keeping to the 75-character limit for
4503 encoded words that the RFC imposes. It now respects the limit, and
4504 generates multiple encoded words if necessary. To be on the safe side, I
4505 have increased the buffer size for the ${rfc2047: expansion operator from
4508 22. Failure to deliver a bounce message always caused it to be frozen, even if
4509 there was an errors_to setting on the router. The errors_to setting is now
4512 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4513 canonical form (fully expanded) before being placed in
4514 $sender_host_address.
4516 24. Updated eximstats to version 1.33
4518 25. Include certificate and key file names in error message when GnuTLS fails
4519 to set them up, because the GnuTLS error message doesn't include the name
4520 of the failing file when there is a problem reading it.
4522 26. Expand error message when OpenSSL has problems setting up cert/key files.
4525 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4526 (this can affect the format of dates).
4528 28. exim_tidydb, when checking for the continued existence of a message for
4529 which it has found a message-specific retry record, was not finding
4530 messages that were in split spool directories. Consequently, it was
4531 deleting retry records that should have stayed in existence.
4533 29. eximstats updated to version 1.35
4534 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4535 1.35 - bugfix such that pie charts by volume are generated correctly
4537 30. The SPA authentication driver was not abandoning authentication and moving
4538 on to the next authenticator when an expansion was forced to fail,
4539 contradicting the general specification for all authenticators. Instead it
4540 was generating a temporary error. It now behaves as specified.
4542 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4543 (the order specifies the preference for clients). The order is now AES256,
4544 AES128, 3DES, ARCFOUR128.
4546 31. Small patch to Sieve code - explicitly set From: when generating an
4549 32. Exim crashed if a remote delivery caused a very long error message to be
4550 recorded - for instance if somebody sent an entire SpamAssassin report back
4551 as a large number of 550 error lines. This bug was coincidentally fixed by
4552 increasing the size of one of Exim's internal buffers (big_buffer) that
4553 happened as part of the Exiscan merge. However, to be on the safe side, I
4554 have made the code more robust (and fixed the comments that describe what
4557 33. Some experimental protocols are using DNS PTR records for new purposes. The
4558 keys for these records are domain names, not reversed IP addresses. The
4559 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4560 leaves it alone. Component reversal etc. now happens only for IP addresses.
4563 34. The host_aton() function is supposed to be passed a string that is known
4564 to be a valid IP address. However, in the case of IPv6 addresses, it was
4565 not checking this. This is a hostage to fortune. Exim now panics and dies
4566 if the condition is not met. A case was found where this could be provoked
4567 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4568 components; fortuitously, this particular loophole had already been fixed
4569 by change 4.50/55 or 4.44/33 above.
4571 If there are any other similar loopholes, the new check in host_aton()
4572 itself should stop them being exploited. The report I received stated that
4573 data on the command line could provoke the exploit when Exim was running as
4574 exim, but did not say which command line option was involved. All I could
4575 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4576 running as the user.
4579 35. There was a buffer overflow vulnerability in the SPA authentication code
4580 (which came originally from the Samba project). I have added a test to the
4581 spa_base64_to_bits() function which I hope fixes it.
4584 36. The daemon start-up calls getloadavg() while still root for those OS that
4585 need the first call to be done as root, but it missed one case: when
4586 deliver_queue_load_max is set with deliver_drop_privilege. This is
4587 necessary for the benefit of the queue runner, because there is no re-exec
4588 when deliver_drop_privilege is set.
4590 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4591 was in use, was not putting the data itself into the right store pool;
4592 consequently, it could be overwritten for a subsequent message in the same
4593 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4596 38. Sometimes the final signoff response after QUIT could fail to get
4597 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4598 < 0 before doing a fflush(). This bug looks as though it goes back to the
4599 introduction of TLS in release 3.20, but "sometimes" must have been rare
4600 because the tests only now provoked it.
4606 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4607 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4608 Now it returns an integer. A number of calls were still expecting a T/F
4609 return. Fortuitously, in all cases, the tests worked in OK situations,
4610 which is the norm. However, things would have gone wrong on any write
4611 failures on the smtp file descriptor. This function is used when sending
4612 messages over SMTP and also when doing verify callouts.
4614 2. When Exim is called to do synchronous delivery of a locally submitted
4615 message (the -odf or -odi options), it no longer closes stderr before doing
4618 3. Implemented the mua_wrapper option.
4620 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4622 5. Implemented the functions header_remove(), header_testname(),
4623 header_add_at_position(), and receive_remove_recipient(), and exported them
4626 6. If an ACL "warn" statement specified the addition of headers, Exim already
4627 inserted X-ACL-Warn: at the start if there was no header name. However, it
4628 was not making this test for the second and subsequent header lines if
4629 there were newlines in the string. This meant that an invalid header could
4630 be inserted if Exim was badly configured.
4632 7. Allow an ACL "warn" statement to add header lines at the start or after all
4633 the Received: headers, as well as at the end.
4635 8. Added the rcpt_4xx retry error code.
4637 9. Added postmaster_mailfrom=xxx to callout verification option.
4639 10. Added mailfrom=xxxx to the callout verification option, for verify=
4642 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4643 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4645 12. Inserted some casts to stop certain compilers warning when using pointer
4646 differences as field lengths or precisions in printf-type calls (mostly
4647 affecting debugging statements).
4649 13. Added optional readline() support for -be (dynamically loaded).
4651 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4652 same clock tick as a message's arrival, so that its received time was the
4653 same as the "first fail" time on the retry record, and that message
4654 remained on the queue past the ultimate address timeout, every queue runner
4655 would try a delivery (because it was past the ultimate address timeout) but
4656 after another failure, the ultimate address timeout, which should have then
4657 bounced the address, did not kick in. This was a "< instead of <=" error;
4658 in most cases the first failure would have been in the next clock tick
4659 after the received time, and all would be well.
4661 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4662 being recognized when the domain list was tested by the match_domain
4663 condition in an expansion string.
4665 16. Added the ${str2b64: operator.
4667 17. Exim was always calling setrlimit() to set a large limit for the number of
4668 processes, without checking whether the existing limit was already
4669 adequate. (It did check for the limit on file descriptors.) Furthermore,
4670 errors from getrlimit() and setrlimit() were being ignored. Now they are
4671 logged to the main and panic logs, but Exim does carry on, to try to do its
4672 job under whatever limits there are.
4674 18. Imported PCRE 5.0.
4676 19. Trivial typo in log message " temporarily refused connection" (the leading
4679 20. If the log selector return_path_on_delivery was set and an address was
4680 redirected to /dev/null, the delivery process crashed because it assumed
4681 that a return path would always be set for a "successful" delivery. In this
4682 case, the whole delivery is bypassed as an optimization, and therefore no
4685 21. Internal re-arrangement: the function for sending a challenge and reading
4686 a response while authentication was assuming a zero-terminated challenge
4687 string. It's now changed to take a pointer and a length, to allow for
4688 binary data in such strings.
4690 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4692 23. Exim was not respecting finduser_retries when seeking the login of the
4693 uid under which it was called; it was always trying 10 times. (The default
4694 setting of finduser_retries is zero.) Also, it was sleeping after the final
4695 failure, which is pointless.
4697 24. Implemented tls_on_connect_ports.
4699 25. Implemented acl_smtp_predata.
4701 26. If the domain in control=submission is set empty, Exim assumes that the
4702 authenticated id is a complete email address when it generates From: or
4703 Sender: header lines.
4705 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4706 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4707 chown and chgrp in /bin and hostname in /usr/bin.
4709 28. Exim was keeping the "process log" file open after each use, just as it
4710 does for the main log. This opens the possibility of it remaining open for
4711 long periods when the USR1 signal hits a daemon. Occasional processlog
4712 errors were reported, that could have been caused by this. Anyway, it seems
4713 much more sensible not to leave this file open at all, so that is what now
4716 29. The long-running daemon process does not normally write to the log once it
4717 has entered its main loop, and it closes the log before doing so. This is
4718 so that log files can straightforwardly be renamed and moved. However,
4719 there are a couple of unusual error situations where the daemon does write
4720 log entries, and I had neglected to close the log afterwards.
4722 30. The text of an SMTP error response that was received during a remote
4723 delivery was being truncated at 512 bytes. This is too short for some of
4724 the long messages that one sometimes sees. I've increased the limit to
4727 31. It is now possible to make retry rules that apply only when a message has a
4728 specific sender, in particular, an empty sender.
4730 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4731 it possible to be selective about when SMTP synchronization is enforced.
4733 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4735 32. Implemented hosts_connection_nolog.
4737 33. Added an ACL for QUIT.
4739 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4742 35. Added mailbox_size and mailbox_filecount to appendfile.
4744 36. Added control = no_multiline_responses to ACLs.
4746 37. There was a bug in the logic of the code that waits for the clock to tick
4747 in the case where the clock went backwards by a substantial amount such
4748 that the microsecond fraction of "now" was more than the microsecond
4749 fraction of "then" (but the whole seconds number was less).
4751 38. Added support for the libradius Radius client library this is found on
4752 FreeBSD (previously only the radiusclient library was supported).
4758 1. When certain lookups returned multiple values in the form name=value, the
4759 quoting of the values was not always being done properly. Specifically:
4760 (a) If the value started with a double quote, but contained no whitespace,
4762 (b) If the value contained whitespace other than a space character (i.e.
4763 tabs or newlines or carriage returns) it was not quoted.
4764 This fix has been applied to the mysql and pgsql lookups by writing a
4765 separate quoting function and calling it from the lookup code. The fix
4766 should probably also be applied to nisplus, ibase and oracle lookups, but
4767 since I cannot test any of those, I have not disturbed their existing code.
4769 2. A hit in the callout cache for a specific address caused a log line with no
4770 reason for rejecting RCPT. Now it says "Previous (cached) callout
4771 verification failure".
4773 3. There was an off-by-one bug in the queryprogram router. An over-long
4774 return line was truncated at 256 instead of 255 characters, thereby
4775 overflowing its buffer with the terminating zero. As well as fixing this, I
4776 have increased the buffer size to 1024 (and made a note to document this).
4778 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4779 when Exim is waiting for an SMTP response from a remote server, Exim
4780 restarts its select() call on the socket, thereby resetting its timeout.
4781 This is not a problem when such interrupts are rare. Somebody set up a cron
4782 job to run exiwhat every 2 minutes, which is less than the normal select()
4783 timeout (5 or 10 minutes). This meant that the select() timeout never
4784 kicked in because it was always reset. I have fixed this by comparing the
4785 time when an interrupt arrives with the time at the start of the first call
4786 to select(). If more time than the timeout has elapsed, the interrupt is
4787 treated as a timeout.
4789 5. Some internal re-factoring in preparation for the addition of Sieve
4790 extensions (by MH). In particular, the "personal" test is moved to a
4791 separate function, and given an option for scanning Cc: and Bcc: (which is
4792 not set for Exim filters).
4794 6. When Exim created an email address using the login of the caller as the
4795 local part (e.g. when creating a From: or Sender: header line), it was not
4796 quoting the local part when it contained special characters such as @.
4798 7. Installed new OpenBSD configuration files.
4800 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4801 try to make them clearer.
4803 9. Callout options, other than the timeout value, were being ignored when
4804 verifying sender addresses in header lines. For example, when using
4806 verify = header_sender/callout=no_cache
4808 the cache was (incorrectly) being used.
4810 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4811 only the Cygwin environment.
4813 11. When return_path_on_delivery was set as a log selector, if different remote
4814 addresses in the same message used different return paths and parallel
4815 remote delivery occurred, the wrong values would sometimes be logged.
4816 (Whenever a remote delivery process finished, the return path value from
4817 the most recently started remote delivery process was logged.)
4819 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4820 lines when AUTH and/or TLS are in use. This is the "received protocol"
4821 field. Exim used to use "asmtp" for authenticated SMTP, without any
4822 indication (in the protocol name) for TLS use. Now it follows the RFC and
4823 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4824 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4825 names appear in log lines as well as in Received: header lines.
4827 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4828 extensions, and comparison tests, and to fix some bugs.
4830 14. Changes to the "personal" filter test:
4832 (1) The test was buggy in that it was just doing the equivalent of
4833 "contains" tests on header lines. For example, if a user's address was
4834 anne@some.where, the "personal" test would incorrectly be true for
4836 To: susanne@some.where
4838 This test is now done by extracting each address from the header in turn,
4839 and checking the entire address. Other tests that are part of "personal"
4840 are now done using regular expressions (for example, to check local parts
4841 of addresses in From: header lines).
4843 (2) The list of non-personal local parts in From: addresses has been
4844 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4845 taken from the Sieve specification recommendations.
4847 (3) If the message contains any header line starting with "List-" it is
4848 treated as non-personal.
4850 (4) The test for "circular" in the Subject: header line has been removed
4851 because it now seems ill-conceived.
4853 15. Minor typos in src/EDITME comments corrected.
4855 16. Installed latest exipick from John Jetmore.
4857 17. If headers_add on a router specified a text string that was too long for
4858 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4859 of string_sprintf() is now avoided.
4861 18. $message_body_size was not set (it was always zero) when running the DATA
4862 ACL and the local_scan() function.
4864 19. For the "mail" command in an Exim filter, no default was being set for
4865 the once_repeat time, causing a random time value to be used if "once" was
4866 specified. (If the value happened to be <= 0, no repeat happened.) The
4867 default is now 0s, meaning "never repeat". The "vacation" command was OK
4868 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4869 (I found it when inspecting the code).
4871 20. There is now an overall timeout for performing a callout verification. It
4872 defaults to 4 times the callout timeout, which applies to individual SMTP
4873 commands during the callout. The overall timeout applies when there is more
4874 than one host that can be tried. The timeout is checked before trying the
4875 next host. This prevents very long delays if there are a large number of
4876 hosts and all are timing out (e.g. when the network connections are timing
4877 out). The value of the overall timeout can be changed by specifying an
4878 additional sub-option for "callout", called "maxwait". For example:
4880 verify = sender/callout=5s,maxwait=20s
4882 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4883 to the end before writing, but this should make it even safer).
4885 22. Exim was forgetting that it had advertised PIPELINING for the second and
4886 subsequent messages on an SMTP connection. It was also not resetting its
4887 memory on STARTTLS and an internal HELO.
4889 23. When Exim logs an SMTP synchronization error within a session, it now
4890 records whether PIPELINING has been advertised or not.
4892 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4893 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4894 rather than long int.
4896 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4898 26. Added the never_mail option to autoreply.
4904 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4905 crash if the getsockname() call failed; this can happen if a connection is
4906 closed very soon after it is established. The problem was simply in the
4907 order in which certain operations were done, causing Exim to try to write
4908 to the SMTP stream before it had set up the file descriptor. The bug has
4909 been fixed by making things happen in the correct order.
4915 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4916 before the connection was closed, thus losing the rejection response.
4918 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4919 some early Solaris releases, but causes trouble in current releases where
4920 socklen_t is defined.
4922 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4925 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4928 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4929 that releasing the top of it at the end releases what was used for sub-
4930 expansions (unless the block got too big). However, discard this block if
4931 the first thing is a variable or header, so that we can use its block when
4932 it is dynamic (useful for very large $message_headers, for example).
4934 6. Lookups now cache *every* query, not just the most recent. A new, separate
4935 store pool is used for this. It can be recovered when all lookup caches are
4936 flushed. Lookups now release memory at the end of their result strings.
4937 This has involved some general refactoring of the lookup sources.
4939 7. Some code has been added to the store_xxx() functions to reduce the amount
4940 of flapping under certain conditions.
4942 8. log_incoming_interface used to affect only the <= reception log lines. Now
4943 it causes the local interface and port to be added to several more SMTP log
4944 lines, for example "SMTP connection from", and rejection lines.
4946 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4948 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4950 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4951 could be overwritten at the end of the current message (or the start of a
4952 new message if it was set in a HELO ACL). The value is now preserved for
4953 the duration of the SMTP connection.
4955 12. If a transport had a headers_rewrite setting, and a matching header line
4956 contained an unqualified address, that address was qualified, even if it
4957 did not match any rewriting rules. The underlying bug was that the values
4958 of the flags that permit the existence of unqualified sender and recipient
4959 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4960 non-local messages, and by -bnq for local messages) were not being
4961 preserved with the message after it was received.
4963 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4964 "next input=" as part of the text comprising the host identity instead of
4965 the correct text. The code was using the same buffer for two different
4966 strings. However, depending on which order the printing function evaluated
4967 its arguments, the bug did not always show up. Under Linux, for example, my
4968 test suite worked just fine.
4970 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4971 doesn't work with some older versions of Perl. It has been changed to "my",
4972 which in any case is probably the better facility to use.
4974 15. A really picky compiler found some instances of statements for creating
4975 error messages that either had too many or two few arguments for the format
4978 16. The size of the buffer for calls to the DNS resolver has been increased
4979 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4980 for addresses that have a lot of PTR records. This alleviates a problem; it
4981 does not fully solve it.
4983 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4984 buffer now truncates the list and logs the incident, which is the same
4985 action as happens when Exim is looking up a host name and its aliases.
4986 Previously in this situation something unpredictable would happen;
4987 sometimes it was "internal error: store_reset failed".
4989 18. If a server dropped the connection unexpectedly when an Exim client was
4990 using GnuTLS and trying to read a response, the client delivery process
4991 crashed while trying to generate an error log message.
4993 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4994 string, for example:
4996 warn message = H1: something\nH2: something
4998 the text was added as a single header line from Exim's point of view
4999 though it ended up OK in the delivered message. However, searching for the
5000 second and subsequent header lines using $h_h2: did not work. This has been
5001 fixed. Similarly, if a system filter added multiple headers in this way,
5002 the routers could not see them.
5004 20. Expanded the error message when iplsearch is called with an invalid key to
5005 suggest using net-iplsearch in a host list.
5007 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5008 ACLs are no longer actually imposed (and a message to that effect is
5011 22. If a "gecos" field in a passwd entry contained escaped characters, in
5012 particular, if it contained a \" sequence, Exim got it wrong when building
5013 a From: or a Sender: header from that name. A second bug also caused
5014 incorrect handling when an unquoted " was present following a character
5015 that needed quoting.
5017 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5018 was not being matched caselessly.
5020 24. Arranged for all hyphens in the exim.8 source to be escaped with
5023 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5024 properly. Recipient callout cache records were still being keyed to include
5025 the sender, even when use_sender was set false. This led to far more
5026 callouts that were necessary. The sender is no longer included in the key
5027 when use_sender is false.
5029 26. Added "control = submission" modifier to ACLs.
5031 27. Added the ${base62d: operator to decode base 62 numbers.
5033 28. dnsdb lookups can now access SRV records.
5035 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5036 the configuration file.
5038 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5039 (-v) mode. This makes the output for a verbose queue run more intelligible.
5041 31. Added a use_postmaster feature to recipient callouts.
5043 32. Added the $body_zerocount variable, containing the number of binary zero
5044 bytes in the message body.
5046 33. The time of last modification of the "new" subdirectory is now used as the
5047 "mailbox time last read" when there is a quota error for a maildir
5050 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5052 35. Added +ignore_unknown as a special item in host lists.
5054 36. Code for decoding IPv6 addresses in host lists is now included, even if
5055 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5056 address was recognized as an IP address, but was then not correctly decoded
5057 into binary, causing unexpected and incorrect effects when compared with
5064 1. Very minor rewording of debugging text in manualroute to say "list of
5065 hosts" instead of "hostlist".
5067 2. If verify=header_syntax was set, and a header line with an unqualified
5068 address (no domain) and a large number of spaces between the end of the
5069 name and the colon was received, the reception process suffered a buffer
5070 overflow, and (when I tested it) crashed. This was caused by some obsolete
5071 code that should have been removed. The fix is to remove it!
5073 3. When running in the test harness, delay a bit after writing a bounce
5074 message to get a bit more predictability in the log output.
5076 4. Added a call to search_tidyup() just before forking a reception process. In
5077 theory, someone could use a lookup in the expansion of smtp_accept_max_
5078 per_host which, without the tidyup, could leave open a database connection.
5080 5. Added the variables $recipient_data and $sender_data which get set from a
5081 lookup success in an ACL "recipients" or "senders" condition, or a router
5082 "senders" option, similar to $domain_data and $local_part_data.
5084 6. Moved the writing of debug_print from before to after the "senders" test
5087 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5088 problems for message scanning, either using a data ACL, or using
5089 local_scan() because the Received: header was not generated till after they
5090 were called (in order to set the time as the time of reception completion).
5091 I have revised the way this works. The header is now generated after the
5092 body is received, but before the ACL or local_scan() are called. After they
5093 are run, the timestamp in the header is updated.
5099 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5100 before starting a queue runner without re-exec. This happened only when
5101 deliver_drop_privilege was set or when the Exim user was set to root. The
5102 effect of the bug was that timeouts during subsequent deliveries caused
5103 crashes instead of being properly handled. The handler is now left at its
5104 default (and expected) setting.
5106 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5107 message, again when deliver_drop_privilege is set or Exim is run as root.
5108 The bug described in (1) was not present in this case, but the tidying up
5109 of the other signals was missing. I have made the two cases consistent.
5111 3. The ignore_target_hosts setting on a manualroute router was being ignored
5112 for hosts that were looked up using the /MX notation.
5114 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5117 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5118 operated on the sender address. After changing the $sender_address to <>
5119 for the sender address verify, Exim was re-instated it as the original
5120 (before rewriting) address, but remembering that it had rewritten it, so it
5121 wasn't rewriting it again. This bug also had the effect of breaking the
5122 sender address verification caching when the sender address was rewritten.
5124 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5125 This has been changed so that if the ip literal address matches
5126 ignore_target_hosts, the router declines.
5128 7. Added expansion conditions match_domain, match_address, and match_local_
5129 part (NOT match_host).
5131 8. The placeholder for the Received: header didn't have a length field set.
5133 9. Added code to Exim itself and to exim_lock to test for a specific race
5134 condition that could lead to file corruption when using MBX delivery. The
5135 issue is with the lockfile that is created in /tmp. If this file is removed
5136 after a process has opened it but before that process has acquired a lock,
5137 there is the potential for a second process to recreate the file and also
5138 acquire a lock. This could lead to two Exim processes writing to the file
5139 at the same time. The added code performs the same test as UW imapd; it
5140 checks after acquiring the lock that its file descriptor still refers to
5141 the same named file.
5143 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5144 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5147 11. Added the smtp_active_hostname option. If used, this will typically be made
5148 to depend on the incoming interface address. Because $interface_address is
5149 not set up until the daemon has forked a reception process, error responses
5150 that can happen earlier (such as "too many connections") no longer contain
5153 12. If an expansion in a condition on a "warn" statement fails because a lookup
5154 defers, the "warn" statement is abandoned, and the next ACL statement is
5155 processed. Previously this caused the whole ACL to be aborted.
5157 13. Added the iplsearch lookup type.
5159 14. Added ident_timeout as a log selector.
5161 15. Added tls_certificate_verified as a log selector.
5163 16. Added a global option tls_require_ciphers (compare the smtp transport
5164 option of the same name). This controls incoming TLS connections.
5166 17. I finally figured out how to make tls_require_ciphers do a similar thing
5167 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5168 before starting the TLS session.
5170 18. Tabs are now shown as \t in -bP output.
5172 19. If the log selector return_path_on_delivery was set, Exim crashed when
5173 bouncing a message because it had too many Received: header lines.
5175 20. If two routers both had headers_remove settings, and the first one included
5176 a superfluous trailing colon, the final name in the first list and the
5177 first name in the second list were incorrectly joined into one item (with a
5178 colon in the middle).
5184 1. Added -C and -D options to the exinext utility, mainly to make it easier
5185 to include in the automated testing, but these could be helpful when
5186 multiple configurations are in use.
5188 2. The exinext utility was not formatting the output nicely when there was
5189 an alternate port involved in the retry record key, nor when there was a
5190 message id as well (for retries that were specific to a specific message
5191 and a specific host). It was also confused by IPv6 addresses, because of
5192 the additional colons they contain. I have fixed the IPv4 problem, and
5193 patched it up to do a reasonable job for IPv6.
5195 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5196 delivery, the log line now contains "pipelined" if PIPELINING was used.
5198 4. An SMTP transport process used to panic and die if the bind() call to set
5199 an explicit outgoing interface failed. This has been changed; it is now
5200 treated in the same way as a connect() failure.
5202 5. A reference to $sender_host_name in the part of a conditional expansion
5203 that was being skipped was still causing a DNS lookup. This no longer
5206 6. The def: expansion condition was not recognizing references to header lines
5207 that used bh_ and bheader_.
5209 7. Added the _cache feature to named lists.
5211 8. The code for checking quota_filecount in the appendfile transport was
5212 allowing one more file than it should have been.
5214 9. For compatibility with Sendmail, the command line option
5222 and sets the incoming protocol and host name (for trusted callers). The
5223 host name and its colon can be omitted when only the protocol is to be set.
5224 Note the Exim already has two private options, -pd and -ps, that refer to
5225 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5226 "s", but I don't think that's a major issue.
5228 10. A number of refactoring changes to the code, none of which should affect
5231 (a) The number of logging options was getting close to filling up the
5232 32-bit word that was used as a bit map. I have split them into two classes:
5233 those that are passed in the argument to log_write(), and those that are
5234 only ever tested independently outside of that function. These are now in
5235 separate 32-bit words, so there is plenty of room for expansion again.
5236 There is no change in the user interface or the logging behaviour.
5238 (b) When building, for example, log lines, the code previously used a
5239 macro that called string_cat() twice, in order to add two strings. This is
5240 not really sufficiently general. Furthermore, there was one instance where
5241 it was actually wrong because one of the argument was used twice, and in
5242 one call a function was used. (As it happened, calling the function twice
5243 did not affect the overall behaviour.) The macro has been replaced by a
5244 function that can join an arbitrary number of extra strings onto a growing
5247 (c) The code for expansion conditions now uses a table and a binary chop
5248 instead of a serial search (which was left over from when there were very
5249 few conditions). Also, it now recognizes conditions like "pam" even when
5250 the relevant support is not compiled in: a suitably worded error message is
5251 given if an attempt is made to use such a condition.
5253 11. Added ${time_interval:xxxxx}.
5255 12. A bug was causing one of the ddress fields not to be passed back correctly
5256 from remote delivery subprocesses. The field in question was not being
5257 subsequently used, so this caused to problems in practice.
5259 13. Added new log selectors queue_time and deliver_time.
5261 14. Might have fixed a bug in maildirsizefile handling that threw up
5262 "unexpected character" debug warnings, and recalculated the data
5263 unnecessarily. In any case, I expanded the warning message to give more
5266 15. Added the message "Restricted characters in address" to the statements in
5267 the default ACL that block characters like @ and % in local parts.
5269 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5270 Three changes have been made:
5272 (a) There was a serious bug; a negative response to MAIL caused the whole
5273 recipient domain to be cached as invalid, thereby blocking all messages
5274 to all local parts at the same domain, from all senders. This bug has
5275 been fixed. The domain is no longer cached after a negative response to
5276 MAIL if the sender used is not empty.
5278 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5281 (c) A new callout option, "use_sender" has been added for people who want
5282 the modified behaviour.
5288 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5291 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5292 indeed breaks things for older releases.
5294 3. Added additional logging to the case where there is a problem reading data
5295 from a filter that is running in a subprocess using a pipe, in order to
5296 try to track down a specific problem.
5298 4. Testing facility fudge: when running in the test harness and attempting
5299 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5300 getting "No route to host". Convert this to a timeout.
5302 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5305 6. Some OS don't have socklen_t but use size_t instead. This affects the
5306 fifth argument of getsockopt() amongst other things. This is now
5307 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5308 can be set for individual OS. I have set it for SunOS5, OSF1, and
5309 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5310 some earlier ones do not.
5312 7. Change 4.30/15 was not doing the test caselessly.
5314 8. The standard form for an IPv6 address literal was being rejected by address
5315 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5316 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5317 this, as well as the form without the "IPv6" on the front (but only when
5318 address literals are enabled, of course).
5320 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5322 10. Exim crashed if a message with an empty sender address specified by -f
5323 encountered a router with an errors_to setting. This could be provoked only
5324 by a command such as
5328 where an empty string was supplied; "<>" did not hit this bug.
5330 11. Installed PCRE release 4.5.
5332 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5333 remained set. It is now erased.
5335 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5336 times from message ids (which are base 36 rather than the normal 62).
5338 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5339 were being counted as actual protocol errors, and logged if the log
5340 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5341 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5342 and DATA following a set of rejected RCPTs do not count as protocol errors.
5343 In other words, Exim assumes they were pipelined, though this may not
5344 actually be the case. Of course, in all cases the client gets an
5345 appropriate error code.
5347 15. If a lookup fails in an ACL condition, a message about the failure may
5348 be available; it is used if testing the ACL cannot continue, because most
5349 such messages specify what the cause of the deferral is. However, some
5350 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5351 that caused an old message to be retained and used if a later statement
5352 caused a defer, replacing the real cause of the deferral.
5354 16. If an IP address had so many PTR records that the DNS lookup buffer
5355 was not large enough to hold them, Exim could crash while trying to process
5356 the truncated data. It now detects and logs this case.
5358 17. Further to 4.21/58, another change has been made: if (and only if) the
5359 first line of a message (the first header line) ends with CRLF, a bare LF
5360 in a subsequent header line has a space inserted after it, so as not to
5361 terminate the header.
5363 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5364 unnecessarily, used atoi() instead of strtol(), and didn't check the
5365 termination when getting file sizes from file names by regex.
5367 19. Completely re-implemented the support for maildirsize files, in the light
5368 of a number of problems with the previous contributed implementation
5369 (4.30/29). In particular:
5371 . If the quota is zero, the maildirsize file is maintained, but no quota is
5374 . If the maildir directory does not exist, it is created before any attempt
5375 to write a maildirsize file.
5377 . The quota value in the file is just a cache; if the quota is changed in
5378 the transport, the new value overrides.
5380 . A regular expression is available for excluding directories from the
5383 20. The autoreply transport checks the characters in options that define the
5384 message's headers; it allows continued headers, but it was checking with
5385 isspace() after an embedded newline instead of explicitly looking for a
5388 21. If all the "regular" hosts to which an address was routed had passed their
5389 expiry times, and had not reached their retry times, the address was
5390 bounced, even if fallback hosts were defined. Now Exim should go on to try
5393 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5394 equivalent code in the SMTP transport. Some hosts send humungous responses
5395 to HELO/EHLO, more than 1024 it seems.
5397 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5398 gives compiler warnings in some environments. I've now done it "properly",
5401 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5402 (because of problems with the built-in one) was declared to return uschar *
5403 instead of char *, causing compiler failure.
5405 25. Fixed a file descriptor leak when processing alias/forward files.
5407 26. Fixed a minor format string issue in dbfn.c.
5409 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5411 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5412 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5415 29. When a sender address is verified, it is cached, to save repeating the test
5416 when there is more than one recipient in a message. However, when the
5417 verification involves a callout, it is possible for different callout
5418 options to be set for different recipients. It is too complicated to keep
5419 track of this in the cache, so now Exim always runs a verification when a
5420 callout is required, relying on the callout cache for the optimization.
5421 The overhead is duplication of the address routing, but this should not be
5424 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5425 to be verified with callout=postmaster, and the main callout worked but the
5426 postmaster check failed, the verification correctly failed. However, if a
5427 subsequent RCPT command asked for sender verification *without* the
5428 postmaster check, incorrect caching caused this verification also to fail,
5431 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5432 it was not caching the DNS options (qualify_single, search_parents) that
5433 were used when the lookup failed. A subsequent lookup with different
5434 options therefore always gave the same answer, though there were cases
5435 where it should not have. (Example: a "domains = !$mx_any" option on a
5436 dnslookup router: the "domains" option is always processed without any
5437 widening, but the router might have qualify_single set.) Now Exim uses the
5438 cached value only when the same options are set.
5440 32. Added John Jetmore's "exipick" utility to the distribution.
5442 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5443 than a timeout (e.g. a certificate is required, and is not provided), an
5444 Exim server now closes the connection immediately. Previously it waited for
5445 the client to close - but if the client is SSL, it seems that they each
5446 wait for each other, leading to a delay before one of them times out.
5448 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5449 maintained 0.8.x compatibility because I don't think many are using it, and
5450 it is clearly obsolete.
5452 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5455 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5456 client certificate was expired. A simple patch fixes this, though I don't
5457 understand the full logic of why the verify callback is called multiple
5460 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5461 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5462 which causes problems with some clients (such as the Certicom SSL Plus
5463 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5464 disables the coutermeasure allowing Eudora to connect."
5466 38. Exim was not checking that a write() to a log file succeeded. This could
5467 lead to Bad Things if a log got too big, in particular if it hit a file
5468 size limit. Exim now panics and dies if it cannot write to a log file, just
5469 as it does if it cannot open a log file.
5471 39. Modified OS/Makefile-Linux so that it now contains
5473 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5475 The two -D definitions ensure that Exim is compiled with large file
5476 support, which makes it possible to handle log files that are bigger than
5479 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5480 instance) a domain was checked against a named list that involved a lookup,
5481 causing $domain_data to be set, then another domain was checked against the
5482 same list, then the first domain was re-checked, the value of $domain_data
5483 after the final check could be wrong. In particular, if the second check
5484 failed, it could be set empty. This bug probably also applied to
5487 41. The strip_trailing_dot option was not being applied to the address given
5488 with the -f command-line option.
5490 42. The code for reading a message's header from the spool was incrementing
5491 $received_count, but never initializing it. This meant that the value was
5492 incorrect (doubled) while delivering a message in the same process in which
5493 it was received. In the most common configuration of Exim, this never
5494 happens - a fresh exec is done - but it can happen when
5495 deliver_drop_privilege is set.
5497 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5498 it now includes up to 150 characters of the unexpected data in the log
5501 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5502 and building data strings. The size of both of these buffers was 10 000
5503 bytes - far larger than anybody would *ever* want, thought I. Needless to
5504 say, somebody hit the limit. I have increased the maximum line length to
5505 20 000 and the maximum data length of concatenated lines to 100 000. I have
5506 also fixed two bugs, because there was no checking on these buffers. Tsk,
5507 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5508 buffer is too small.
5510 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5511 lsearch lookups. Now it does.
5513 46. When parsing a route_list item in a manualroute router, a fixed-length
5514 buffer was used for the list of hosts. I made this 1024 bytes long,
5515 thinking that nobody would ever have a list of hosts that long. Wrong.
5516 Somebody had a whole pile of complicated expansion conditions, and the
5517 string was silently truncated, leading to an expansion error. It turns out
5518 that it is easier to change to an unlimited length (owing to other changes
5519 that have happened since this code was originally written) than to build
5520 structure for giving a limitation error. The length of the item that
5521 expands into the list of hosts is now unlimited.
5523 47. The lsearch lookup could not handle data where the length of text line was
5524 more than 4095 characters. Such lines were truncated, leading to shortened
5525 data being returned. It should now handle lines of any length.
5527 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5528 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5531 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5532 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5533 no longer happens in comment lines. A list of replacements is now placed
5534 at the head of all of the source files, except those whose only change is
5535 to replace PERL_COMMAND in the very first #! line.
5537 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5538 messages on the queue, with a bottom-up merge sort, using code contributed
5539 by Michael Haardt. This should make operations like -bp somewhat faster on
5540 large queues. It won't affect queue runners, except when queue_run_in_order
5543 51. Installed eximstats 1.31 in the distribution.
5545 52. Added support for SRV lookups to the dnslookup router.
5547 53. If an ACL referred to $message_body or $message_body_end, the value was not
5548 reset for any messages that followed in the same SMTP session.
5550 54. The store-handling optimization for building very long strings was not
5551 differentiating between the different store pools. I don't think this
5552 actually made any difference in practice, but I've tidied it.
5554 55. While running the routers to verify a sender address, $sender_address
5555 was still set to the sender address. This is wrong, because when routing to
5556 send a bounce to the sender, it would be empty. Therefore, I have changed
5557 it so that, while verifying a sender address, $sender_address is set to <>.
5558 (There is no change to what happens when verifying a recipient address.)
5560 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5561 target A or AAAA records (if not already returned) without resetting the
5562 qualify_single or search_parents options of the DNS resolver. These are
5563 inappropriate in this case because the targets of MX and SRV records must
5564 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5565 target that, when qualified, matched something in the local domain. These
5566 two options are now turned off when doing these lookups.
5568 57. It seems that at least some releases of Reiserfs (which does not have the
5569 concept of a fixed number of inodes) returns zero and not -1 for the
5570 number of available inodes. This interacted badly with check_spool_inodes,
5571 which assumed that -1 was the "no such thing" setting. What I have done is
5572 to check that the total number of inodes is greater than zero before doing
5573 the test of how many are available.
5575 58. When a "warn" ACL statement has a log_message modifier, the message is
5576 remembered, and not repeated. This is to avoid a lot of repetition when a
5577 message has many recipients that cause the same warning to be written.
5578 Howewer, Exim was preserving the list of already written lines for an
5579 entire SMTP session, which doesn't seem right. The memory is now reset if a
5580 new message is started.
5582 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5583 debugging output unless log_rewrite was also set.
5585 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5586 of (int)(handle) when we know that handle contains (void *)(-1).
5588 61. The Exim daemon panic-logs an error return when it closes the incoming
5589 connection. However "connection reset by peer" seems to be common, and
5590 isn't really an error worthy of noting specially, so that particular error
5593 62. When Exim is trying to find all the local interfaces, it used to panic and
5594 die if the ioctl to get the interface flags failed. However, it seems that
5595 on at least one OS (Solaris 9) it is possible to have an interface that is
5596 included in the list of interfaces, but for which you get a failure error
5597 for this call. This happens when the interface is not "plumbed" into a
5598 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5599 failure of the "get flags" call assumes that the interface is down.
5601 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5602 makes life easier for people who are doing arithmetic on fields extracted
5603 from dates, where you often get leading zeros that should not be
5604 interpreted as octal.
5606 64. Added qualify_domain to the redirect router, to override the global
5609 65. If a pathologically long header line contained very many addresses (the
5610 report of this problem mentioned 10 000) and each of them was rewritten,
5611 Exim could use up a very large amount of memory. (It kept on making new
5612 copies of the header line as it rewrote, and never released the old ones.)
5613 At the expense of a bit more processing, the header rewriting function has
5614 been changed so that it no longer eats memory in this way.
5616 66. The generation of the Received: header has been moved from the time that a
5617 message starts to be received, to the time that it finishes. The timestamp
5618 in the Received: header should now be very close to that of the <= log
5619 line. There are two side-effects of this change:
5621 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5622 logged header lines no longer include the local Received: line, because
5623 it has not yet been created. The same applies to a copy of the message
5624 that is returned to a non-SMTP sender when a message is rejected.
5626 (b) When a filter file is tested using -bf, no additional Received: header
5627 is added to the test message. After some thought, I decided that this
5630 This change does not affect the value of $received_for. It is still set
5631 after address rewriting, but before local_scan() is called.
5633 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5635 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5636 gave an unhelpful panic error message, and a defer error. I have managed to
5637 change this behaviour so that it now rejects any supplied certificate,
5638 which seems right, as the list of acceptable certificates is empty.
5640 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5641 gave an unhelpful defer error. I have not managed to make this reject any
5642 supplied certificates, but the error message it gives is "no certificate
5643 supplied", which is not helpful.
5645 70. exigrep's output now also includes lines that are not associated with any
5646 message, but which match the given pattern. Implemented by a patch from
5647 Martin Sluka, which also tidied up the Perl a bit.
5649 71. Recipient callout verification, like sender verification, was using <> in
5650 the MAIL FROM command. This isn't really the right thing, since the actual
5651 sender may affect whether the remote host accepts the recipient or not. I
5652 have changed it to use the actual sender in the callout; this means that
5653 the cache record is now keyed on a recipient/sender pair, not just the
5654 recipient address. There doesn't seem to be a real danger of callout loops,
5655 since a callout by the remote host to check the sender would use <>.
5656 [SEE ABOVE: changed after hitting problems.]
5658 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5659 temporary errors. However, in the case of such a code being given after
5660 the end of a data transmission (i.e. after ".") Exim was failing to write
5661 a retry record for the message. (Yes, there was some broken host that was
5662 actually sending 8xx at this point.)
5664 73. An unknown lookup type in a host list could cause Exim to panic-die when
5665 the list was checked. (An example that provoked this was putting <; in the
5666 middle of a list instead of at the start.) If this happened during a DATA
5667 ACL check, a -D file could be left lying around. This kind of configuration
5668 error no longer causes Exim to die; instead it causes a defer errror. The
5669 incident is still logged to the main and panic logs.
5671 74. Buglet left over from Exim 3 conversion. The message "too many messages
5672 in one connection" was written to the rejectlog but not the mainlog, except
5673 when address rewriting (yes!) was being logged.
5675 75. Added write_rejectlog option.
5677 76. When a system filter was run not as root (that is, when system_filter_user
5678 was set), the values of the $n variables were not being returned to the
5679 main process; thus, they were not subsequently available in the $sn
5682 77. Added +return_path_on_delivery log selector.
5684 78. A connection timeout was being treated differently from recipients deferred
5685 when testing hosts_max_try with a message that was older than the host's
5686 retry timeout. (The host should not be counted, thus allowing all hosts to
5687 be tried at least once before bouncing.) This may have been the cause of an
5688 occasionally reported bug whereby a message would remain on the queue
5689 longer than the retry timeout, but would be bounced if a delivery was
5690 forced. I say "may" because I never totally pinned down the problem;
5691 setting up timeout/retry tests is difficult. See also the next item.
5693 79. The ultimate address timeout was not being applied to errors that involved
5694 a combination of host plus message (for example, a timeout on a MAIL
5695 command). When an address resolved to a number of possible hosts, and they
5696 were not all tried for each delivery (e.g. because of hosts_max_try), a
5697 message could remain on the queue longer than the retry timeout.
5699 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5702 81. Fixed an obscure SMTP outgoing bug which required at least the following
5703 conditions: (a) there was another message waiting for the same server;
5704 (b) the server returned 5xx to all RCPT commands in the first message so
5705 that the message was not completed; (c) the server dropped the connection
5706 or gave a negative response to the RSET that Exim sends to abort the
5707 transaction. The observed case was a dropped connection after DATA that had
5708 been sent in pipelining mode. That is, the server had advertised PIPELINING
5709 but was not implementing it correctly. The effect of the bug was incorrect
5710 behaviour, such as trying another host, and this could lead to a crash.
5716 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5717 and daemon.c were passed as pointers to ints; they should have been
5718 pointers to socklen_t variables (which are typically unsigned ints).
5720 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5723 3. Fixed a really odd bug that affected only the testing scheme; patching a
5724 certain fixed string in the binary changed the value of another string that
5725 happened to be identical to the end of the original first string.
5727 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5728 name", it returns that address as the IP address. On some operating
5729 systems (e.g. Solaris), it also passes back the IP address string as the
5730 "host name". However, on others (e.g. Linux), it passes back an empty
5731 string. Exim wasn't checking for this, and was changing the host name to an
5732 empty string, assuming it had been canonicized.
5734 5. Although rare, it is permitted to have more than one PTR record for a given
5735 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5736 all the names associated with an address, because they do in Solaris.
5737 However, it seems that they do not in Linux for data that comes from the
5738 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5739 I found this out when I moved to a new Linux workstation and tried to run
5740 the Exim test suite.
5742 To get round this problem I have changed the code so that it now does its
5743 own call to the DNS to look up PTR records when searching for a host name.
5744 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5745 addresses that are only in /etc/hosts are still found.
5747 This behaviour is, however, controlled by an option called host_lookup_
5748 order, which defaults to "bydns:byaddr". If people want to use the other
5749 order, or indeed, just use one or the other means of lookup, they can
5750 specify it in this variable.
5752 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5753 some operating systems, this comes back from gethostbyaddr() as an empty
5754 string, and this is what Exim used to test for. However, it seems that in
5755 other systems, "." is yielded. Exim now tests for this case too.
5757 7. The values of check_spool_space and check_log_space are now held internally
5758 as a number of kilobytes instead of an absolute number of bytes. If a
5759 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5760 kilobyte. This means that much larger values can be stored.
5762 8. Exim monitor: an attempt to get the action menu when not actually pointing
5763 at a message produces an empty menu entitled "No message selected". This
5764 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5765 no entries in it ("Shell widget menu has zero width and/or height"). So I
5766 have added a single, blank menu entry in this case.
5768 9. Added ${quote_local_part.
5770 10. MIME decoding is now applied to the contents of Subject: header lines when
5773 11. Now that a reference to $sender_host_address automatically causes a reverse
5774 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5775 host lookup before query-style lookups in lists that might use this
5776 variable. This has therefore been abolished, and the "net-" prefix is no
5777 longer necessary for query-style lookups.
5779 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5780 have been a typo for LFLAGS, so it has been changed.
5782 13. The install script calls Exim with "-C /dev/null" in order to find the
5783 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5784 to be output. Howeve, since Exim outputs its version number before the
5785 error, it didn't break the script. It just looked ugly. I fixed this by
5786 always allowing "-C /dev/null" if the caller is root.
5788 14. Ignore overlarge ACL variable number when reading spool file - insurance
5789 against a later release with more variables having written the file.
5791 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5792 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5793 this, as well as the form without the "IPv6" on the front.
5795 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5796 OS/Makefile-Darwin file.
5798 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5799 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5801 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5802 variable. However, some OS use ints for inodes. I've added cast to long int
5803 to get rid of the compiler warning.
5805 19. I had forgotten to lock out "/../" in configuration file names when
5806 ALT_CONFIG_PREFIX was set.
5808 20. Routers used for verification do not need to specify transports. However,
5809 if such a router generated a host list, and callout was configured, Exim
5810 crashed, because it could not find a port number from the (non-existent)
5811 transport. It now assumes port 25 in this circumstance.
5813 21. Added the -t option to exigrep.
5815 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5816 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5817 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5818 them is set, LOOKUP_LSEARCH is forced.
5820 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5822 24. Added sender and host information to the "rejected by local_scan()" log
5823 line; previously there was no indication of these.
5825 25. Added .include_if_exists.
5827 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5828 Linux. It turns out that not all file systems support this. Apparently some
5829 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5830 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5831 not supported on the file descriptor, is now ignored when Exim is trying to
5832 sync a directory. This applies only to Linux.
5834 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5836 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5837 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5840 29. Added support for maildirsize files from supplied patch (modified a bit).
5842 30. The use of :fail: followed by an empty string could lead Exim to respond to
5843 sender verification failures with (e.g.):
5845 550 Verification failed for <xxx>
5846 550 Sender verify failed
5848 where the first response line was missing the '-' that indicates it is not
5849 the final line of the response.
5851 31. The loop for finding the name of the user that called Exim had a hardwired
5852 limit of 10; it now uses the value of finduser_retries, which is used for
5853 all other user lookups.
5855 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5858 33. Exim was neglecting to zero errno before one call of strtol() when
5859 expanding a string and expecting an integer value. On some systems this
5860 resulted in spurious "integer overflow" errors. Also, it was casting the
5861 result into an int without checking.
5863 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5864 did not work. The code looks as if it has *never* worked, though it appears
5865 to have been documented since at least releast 1.62. I have made it work.
5867 35. The "timeout_DNS" error in retry rules, also documented since at least
5868 1.62, also never worked. As it isn't clear exactly what this means, and
5869 clearly it isn't a major issue, I have abolished the feature by treating it
5870 as "timeout", and writing a warning to the main and panic logs.
5872 36. The display of retry rules for -brt wasn't always showing the error code
5875 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5876 timeout_connect_A, timeout_connect_MX.
5878 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5879 to the empty sender.
5881 39. The daemon was not analysing the content of -oX till after it had closed
5882 stderr and disconnected from the controlling terminal. This meant that any
5883 syntax errors were only noted on the panic log, and the return code from
5884 the command was 0. By re-arranging the code a little, I've made the
5885 decoding happen first, so such errors now appear on stderr, and the return
5886 code is 1. However, the actual setting up of the sockets still happens in
5887 the disconnected process, so errors there are still only recorded on the
5890 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5891 connections (as happens on some IP stacks) was logged at start up time as
5892 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5893 it from "IPv6 and IPv4", which means that two separate sockets are being
5896 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5897 says whether AF_INET or AF_INET6 was passed as an argument.
5899 42. Exiwhat output was messed up when time zones were included in log
5902 43. Exiwhat now gives more information about the daemon's listening ports,
5903 and whether -tls-on-connect was used.
5905 44. The "port" option of the smtp transport is now expanded.
5907 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5908 silently ignored. Now an error message is written to the main and panic
5911 46. There's a new ACL modifier called "logwrite" which writes to a log file
5912 as soon as it is encountered.
5914 47. Added $local_user_uid and $local_user_gid at routing time.
5916 48. Exim crashed when trying to verify a sender address that was being
5919 49. Exim was recognizing only a space character after ".include". It now also
5920 recognizes a tab character.
5922 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5923 extracting the relevant information from the specification. The man page no
5924 longer contains scrambled data for the -d option, and I've added a section
5925 at the front about calling Exim under different names.
5927 51. Added "extra_headers" argument to the "mail" command in filter files.
5929 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5932 53. Installed eximstats 1.29.
5934 54. Added transport_filter_timeout as a generic transport option.
5936 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5937 Cc: header lines. This was required by RFC 822, but it not required by RFC
5940 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5941 incoming messages that did not have them. Now it does so only if the
5942 message originates locally, that is, if there is no associated remote host
5943 address. When Resent- header lines are present, this applies to the Resent-
5944 lines rather than the non-Resent- lines.
5946 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5947 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5949 58. Messages for configuration errors now include the name of the main
5950 configuration file - useful now that there may be more than one file in a
5951 list (.included file names were always shown).
5953 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5954 for those rare installations that do not start the daemon as root or run it
5955 setuid root. I've cut out the call to initgroups() if the daemon is not
5958 60. The Exim user and group can now be bound into the binary as text strings
5959 that are looked up at the start of Exim's processing.
5961 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5963 62. Added $mailstore_basename variable.
5965 63. Installed patch to sieve.c from Michael Haardt.
5967 64. When Exim failed to open the panic log after failing to open the main log,
5968 the original message it was trying to log was written to stderr and debug
5969 output, but if they were not available (the usual case in production), it
5970 was lost. Now it is written to syslog before the two lines that record the
5971 failures to open the logs.
5973 65. Users' Exim filters run in subprocesses under the user's uid. It is
5974 possible for a "deliver" command or an alias in a "personal" command to
5975 provoke an address rewrite. If logging of address rewriting is configured,
5976 this fails because the process is not running as root or exim. There may be
5977 a better way of dealing with this, but for the moment (because 4.30 needs
5978 to be released), I have disabled address rewrite logging when running a
5979 filter in a non-root, non-exim process.
5985 1. The buildconfig auxiliary program wasn't quoting the value set for
5986 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5987 not defined. This bug was masked in 4.22 by the effect that was fixed in
5990 2. Some messages that were rejected after a message id was allocated were
5991 shown as "incomplete" by exigrep. It no longer does this for messages that
5992 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5994 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5995 have allow_domain_literals set, the ID did not get logged in the <= line.
5996 Domain literals are now always recognized in Message-ID: header lines.
5998 4. The first argument for a ${extract expansion item is the key name or field
5999 number. Leading and trailing spaces in this item were not being ignored,
6000 causing some misleading effects.
6002 5. When deliver_drop_privilege was set, single queue runner processes started
6003 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6004 same command in the process it spins off) were not dropping privilege.
6006 6. When the daemon running as "exim" started a queue runner, it always
6007 re-executed Exim in the spun-off process. This is a waste of effort when
6008 deliver_drop_privilege is set. The new process now just calls the
6009 queue-runner function directly.
6015 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6018 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6019 ignored. Though the use of -oP was forcing the writing of a pid file, it
6020 was always written to the default place.
6022 3. If the message "no IP address found for host xxxx" is generated during
6023 incoming verification, it is now followed by identification of the incoming
6024 connection (so you can more easily find what provoked it).
6026 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6028 5. Added some features to "harden" Exim a bit more against certain attacks:
6030 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6031 be put in Local/Makefile. This is like the never_users runtime option,
6032 but it cannot be overridden. The default setting is "root".
6034 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6035 prefix string with which any file named in a -C command line option
6038 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6039 is retained for -C and -D only if the caller of Exim is root. Without
6040 it, the exim user may also use -C and -D and retain privilege.
6042 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6043 command line option is disabled.
6045 6. Macro names set by the -D option must start with an upper case letter, just
6046 like macro names defined in the configuration file.
6048 7. Added "dereference=" facility to LDAP.
6050 8. Two instances of the typo "uknown" in the source files are fixed.
6052 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6053 the Configure-Makefile script screwed up while processing it.
6055 10. Incorporated PCRE 4.4.
6057 11. The SMTP synchronization check was not operating right at the start of an
6058 SMTP session. For example, it could not catch a HELO sent before the client
6059 waited for the greeting. There is now a check for outstanding input at the
6060 point when the greeting is written. Because of the duplex, asynchronous
6061 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6062 way, but not yet received, when the check is performed.
6064 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6065 on TCP/IP sockets, because this apparently causes some broken clients to
6068 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6069 unchanged) from the Cygwin maintainer.
6071 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6072 was supported instead of testing for mbx. Effectively a typo.
6074 15. The spa authenticator server code was not checking that the input it
6075 received was valid base64.
6077 16. The debug output line for the "set" modifier in ACLs was not showing the
6078 name of the variable that was being set.
6080 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6082 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6083 lookup on its own. Something else was needed to trigger the lookup. For
6084 example, a match in host_lookup or the need for a host name in a host list.
6085 Now, if $sender_host_name is referenced and the host name has not yet been
6086 looked up, a lookup is performed. If the lookup fails, the variable remains
6087 empty, and $host_lookup_failed is set to "1".
6089 19. Added "eqi" as a case-independent comparison operator.
6091 20. The saslauthd authentication condition could segfault if neither service
6092 nor realm was specified.
6094 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6095 error message that was logged was misleading, and incoming SMTP
6096 connections were dropped. The message is now more accurate, and temporary
6097 errors are given to SMTP connections.
6099 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6100 (except RSET) with a 421 error, until QUIT is received. However, it was
6101 failing to send a response to QUIT.
6103 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6104 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6105 ACL is now run after the helo_try_verify_hosts code.
6107 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6108 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6109 case-independent, so other case variants are also recognized). Apparently
6110 some systems use these upper case variants.
6112 25. If more than two messages were waiting for the same host, and a transport
6113 filter was specified for the transport, Exim sent two messages over the
6114 same TCP/IP connection, and then failed with "socket operation on non-
6115 socket" when it tried to send the third.
6117 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6119 27. The extern definition of crypt16() in expand.c was not being excluded when
6120 the OS had its own crypt16() function.
6122 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6123 as a preferred synonym for return_size_limit, both as an option and as an
6126 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6128 30. Changed the default configuration ACL to relax the local part checking rule
6129 for addresses that are not in any local domains. For these addresses,
6130 slashes and pipe symbols are allowed within local parts, but the sequence
6131 /../ is explicitly forbidden.
6133 31. SPA server authentication was not clearing the challenge buffer before
6136 32. log_message in a "warn" ACL statement was writing to the reject log as
6137 well as to the main log, which contradicts the documentation and doesn't
6138 seem right (because no rejection is happening). So I have stopped it.
6140 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6141 However, I am unable to do any testing of this.
6143 34. Fixed an infelicity in the appendfile transport. When checking directories
6144 for a mailbox, to see if any needed to be created, it was accidentally
6145 using path names with one or more superfluous leading slashes; tracing
6146 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6148 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6149 contents are added to the log line that is written for every discarded
6150 recipient. (Previously a log_message setting was ignored.)
6152 36. The ${quote: operator now quotes the string if it is empty.
6154 37. The install script runs exim in order to find its version number. If for
6155 some reason other than non-existence or emptiness, which it checks, it
6156 could not run './exim', it was installing it with an empty version number,
6157 i.e. as "exim-". This error state is now caught, and the installation is
6160 38. An argument was missing from the function that creates an error message
6161 when Exim fails to connect to the socket for saslauthd authentication.
6162 This could cause Exim to crash, or give a corrupted message.
6164 39. Added isip, isip4, and isip6 to ${if conditions.
6166 40. The ACL variables $acl_xx are now saved with the message, and can be
6167 accessed later in routers, transports, and filters.
6169 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6170 strings in the file are not string-expanded.
6172 42. If a MAIL command specified a SIZE value that was too large to fit into an
6173 int variable, the check against message_size_limit failed. Such values are
6174 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6175 day this will have to be increased, but I don't think I want to be around
6176 when emails are that large.
6183 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6184 iconv() is not standard in FreeBSD.
6186 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6187 IPv6 enabled. The observed symptom was a segmentation fault on return from
6188 the function os_common_find_running_interfaces() in src/os.c.
6190 3. In the check_special_case() function in daemon.c I had used "errno" as an
6191 argument name, which causes warnings on some systems. This was basically a
6192 typo, since it was named "eno" in the comments!
6194 4. The code that waits for the clock to tick (at a resolution of some fraction
6195 of a second) so as to ensure message-id uniqueness was always waiting for
6196 at least one whole tick, when it could have waited for less. [This is
6197 almost certainly not relevant at current processor speeds, where it is
6198 unlikely to ever wait at all. But we try to future-proof.]
6200 5. The function that sleeps for a time interval that includes fractions of a
6201 second contained a race. It did not block SIGALRM between setting the
6202 timer, and suspending (a couple of lines later). If the interval was short
6203 and the sigsuspend() was delayed until after it had expired, the suspension
6204 never ended. On busy systems this could lead to processes getting stuck for
6207 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6208 process, before it forks any delivery processes. The open lookup caching
6209 mechanism meant that the open file or database connection was passed into
6210 the delivery process. The problem was that delivery processes always tidy
6211 up cached lookup data. This could cause a problem for the next delivery
6212 process started by the queue runner, because the external queue runner
6213 process does not know about the closure. So the next delivery process
6214 still has data in the lookup cache. In the case of a file lookup, there was
6215 no problem because closing a file descriptor in a subprocess doesn't affect
6216 the parent. However, if the lookup was caching a connection to a database,
6217 the connection was closed, and the second delivery process was likely to
6218 see errors such as "PGSQL: query failed: server closed the connection
6219 unexpectedly". The problem has been fixed by closing all cached lookups
6220 in a queue runner before running a delivery process.
6222 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6223 seem to have the "const" qualifier which it has on other OS. I've
6226 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6227 *of the same type* (client or server) with the same public name that an
6228 error should be diagnosed.
6230 9. When Exim looked up a host name for an IP address, but failed to find the
6231 original IP address when looking up the host name (a safety check), it
6232 output the message "<ip address> does not match any IP for NULL", which was
6233 confusing, to say the least. The bug was that the host name should have
6234 appeared instead of "NULL".
6236 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6237 user that is built into the binary, and the -C or -D options is used, root
6238 privilege is dropped before the configuration file is read. In addition,
6239 logging is switched to stderr instead of the normal log files. If the
6240 configuration then re-defines the Exim user, the unprivileged environment
6241 is probably not what is expected, so Exim logs a panic warning message (but
6244 However, if deliver_drop_privilege is set, the unprivileged state may well
6245 be exactly what is intended, so the warning has been cut out in that case,
6246 and Exim is allowed to try to write to its normal log files.
6252 1. smtp_return_error_details was not giving details for temporary sender
6253 or receiver verification errors.
6255 2. Diagnose a configuration error if two authenticators have the same public
6258 3. Exim used not to create the message log file for a message until the first
6259 delivery attempt. This could be confusing when incoming messages were held
6260 for policy or load reasons. The message log file is now created at the time
6261 the message is received, and an initial "Received" line is written to it.
6263 4. The automatically generated man page for command line options had a minor
6264 bug that caused no ill effects; however, a more serious problem was that
6265 the procedure for building the man page automatically didn't always
6266 operate. Consequently, release 4.20 contains an out-of-date version. This
6267 shouldn't happen again.
6269 5. When building Exim with embedded Perl support, the script that builds the
6270 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6271 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6273 6. The freeze_tell option was not being used for messages that were frozen on
6274 arrival, either by an ACL or by local_scan().
6276 7. Added the smtp_incomplete_transaction log selector.
6278 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6279 was accepting AUTH without a new EHLO.
6281 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6282 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6285 10. Logging of TCP/IP connections (when configured) now happens in the main
6286 daemon process instead of the child process, so that the TCP/IP connection
6287 count is more accurate (but it can never be perfect).
6289 11. The use of "drop" in a nested ACL was not being handled correctly in the
6290 outer ACL. Now, if condition failure induced by the nested "drop" causes
6291 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6292 or "require"), the connection is dropped.
6294 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6295 that yield "discard" can now be used with an "accept" or a "discard" verb,
6296 but an error is generated for any others (because I can't see a useful way
6297 to define what should happen).
6299 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6300 are now processed in the same way as lines in the Exim configuration file.
6301 In particular, continuation lines are supported.
6303 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6305 15. Added -ti meaning -t -i.
6307 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6308 domains, and warn by logging if others are found.
6310 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6311 returned from GIFCONF when Exim is trying to find the list of interfaces on
6312 a host. The code in os.c has been modified to copy each ifreq to an aligned
6313 structure in all cases.
6315 Also, in some cases, the returned ifreq's were being copied to a 'struct
6316 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6317 means the last couple of bytes of an IPv6 address could be chopped if the
6318 ifreq contained only a normal sockaddr (14 bytes storage).
6320 18. Named domain lists were not supported in the hosts_treat_as_local option.
6321 An entry such as +xxxx was not recognized, and was treated as a literal
6324 19. Ensure that header lines added by a DATA ACL are included in the reject log
6325 if the ACL subsequently rejects the message.
6327 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6328 MD5 (which is deprecated).
6330 21. When testing a filter file using -bf, Exim was writing a message when it
6331 took the sender from a "From " line in the message, but it was not doing so
6332 when it took $return_path from a Return-Path: header line. It now does.
6334 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6335 with a valid header line field name (a series of printing characters
6336 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6338 23. Changed "disc" in the source to "disk" to conform to the documentation and
6339 the book and for uniformity.
6341 24. Ignore Sendmail's -Ooption=value command line item.
6343 25. When execve() failed while trying to run a command in a pipe transport,
6344 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6345 could be confused with a return value of 69 from the command itself. This
6346 has been changed to 127, the value the shell returns if it is asked to run
6347 a non-existent command. The wording for the related log line suggests a
6348 non-existent command as the problem.
6350 26. If received_header_text expands to an empty string, do not add a Received:
6351 header line to the message. (Well, it adds a token one on the spool, but
6352 marks it "old" so that it doesn't get used or transmitted.)
6354 27. Installed eximstats 1.28 (addition of -nt option).
6356 28. There was no check for failure on the call to getsockname() in the daemon
6357 code. This can fail if there is a shortage of resources on the system, with
6358 ENOMEM, for example. A temporary error is now given on failure.
6360 29. Contrary to the C standard, it seems that in some environments, the
6361 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6362 program. Exim now does this explicitly; it affects the formatting of
6363 timestamps using strftime().
6365 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6366 complaints. I've now initialized all the variables, to avoid this.
6368 32. Header lines added by a system filter were not being "seen" during
6369 transport-time rewrites.
6371 33. The info_callback() function passed to OpenSSL is set up with type void
6372 (*)(SSL *, int, int), as described somewhere. However, when calling the
6373 function (actually a macro) that sets it up, the type void(*)() is
6374 expected. I've put in a cast to prevent warnings from picky compilers.
6376 34. If a DNS black list lookup found a CNAME record, but there were no A
6377 records associated with the domain it pointed at, Exim crashed.
6379 35. If a DNS black list lookup returned more than one A record, Exim ignored
6380 all but the first. It now scans all returned addresses if a particular IP
6381 value is being sought. In this situation, the contents of the
6382 $dnslist_value variable are a list of all the addresses, separated by a
6385 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6386 to accept a host name and all its aliases if the forward lookup for any of
6387 them yielded the IP address of the incoming connection. Now it accepts only
6388 those names whose forward lookup yields the correct IP address. Any other
6389 names are discarded. This closes a loophole whereby a rogue DNS
6390 administrator could create reverse DNS records to break through a
6391 wildcarded host restriction in an ACL.
6393 37. If a user filter or a system filter that ran in a subprocess used any of
6394 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6395 the wrong values were passed to the pipe command ($thisaddress had the
6396 value of $0, $0 had the value of $1, etc). This bug was introduced by
6397 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6399 38. Improved the line breaking for long SMTP error messages from ACLs.
6400 Previously, if there was no break point between 40 and 75 characters, Exim
6401 left the rest of the message alone. Two changes have been made: (a) I've
6402 reduced the minimum length to 35 characters; (b) if it can't find a break
6403 point between 35 and 75 characters, it looks ahead and uses the first one
6404 that it finds. This may give the occasional overlong line, but at least the
6405 remaining text gets split now.
6407 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6408 file descriptors might be low, and that setting 1000 would always raise it.
6409 It turns out that in some environments, the limit is already over 1000 and
6410 that lowering it causes trouble. So now Exim takes care not to decrease it.
6412 40. When delivering a message, the value of $return_path is set to $sender_
6413 address at the start of routing (routers may change the value). By an
6414 oversight, this default was not being set up when an address was tested by
6415 -bt or -bv, which affected the outcome if any router or filter referred to
6418 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6419 log only when "message" or "log_message" are set. However, if one of the
6420 conditions was an address verification, or a call to a nested ACL, the
6421 messages generated by the underlying test were being passed through. This
6422 no longer happens. The underlying message is available in $acl_verify_
6423 message for both "message" and "log_message" expansions, so it can be
6424 passed through if needed.
6426 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6427 new expansion $bh_ to give the encoded byte string without charset
6428 translation. Translation happens only if iconv() is available; HAVE_ICONV
6429 indicates this at build time. HEADERS_CHARSET gives the charset to
6430 translate to; headers_charset can change it in the configuration, and
6431 "headers charset" can change it in an individual filter file.
6433 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6434 that creates RFC 2047 encoded "words" labels them as that charset instead
6435 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6436 expansion operator; (ii) when Exim creates a From: line for a local
6437 message; (iii) when a header line is rewritten to include a "phrase" part.
6439 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6440 buggy, causing it to skip the first lines of messages whose message ID
6441 ended in 'D'. This would not have bitten before Exim release 4.14, because
6442 message IDs were unlikely to end in 'D' before then. The effect was to have
6443 incorrect size information for certain domains.
6445 45. #include "config.h" was missing at the start of the crypt16.c module. This
6446 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6449 46. If there was a timeout during a "random" callout check, Exim treated it as
6450 a failure of the random address, and carried on sending RSET and the real
6451 address. If the delay was just some slowness somewhere, the response to the
6452 original RCPT would be taken as a response to RSET and so on, causing
6453 mayhem of various kinds.
6455 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6456 when I implemented it. It didn't allow for the fact that some option values
6457 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6458 the right test for positive values.
6460 48. Domain names in DNS records are case-independent. Exim always looks them up
6461 in lower case. Some resolvers return domain names in exactly the case they
6462 appear in the zone file, that is, they may contain uppercase letters. Not
6463 all resolvers do this - some return always lower case. Exim was treating a
6464 change of case by a resolver as a change of domain, similar to a widening
6465 of a domain abbreviation. This triggered its re-routing code and so it was
6466 trying to route what was effectively the same domain again. This normally
6467 caused routing to fail (because the router wouldn't handle the domain
6468 twice). Now Exim checks for this case specially, and just changes the
6469 casing of the domain that it ultimately uses when it transmits the message
6472 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6475 50. If a filter generated a file delivery with a non-absolute name (possible if
6476 no home directory exists for the router), the forbid_file option was not
6479 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6480 the existing equality matching.
6482 52. Exim was using ints instead of ino_t variables in some places where it was
6483 dealing with inode numbers.
6485 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6486 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6487 called TMPDIR, and if it finds it is different, it changes its value.
6489 54. The smtp_printf() function is now made available to local_scan() so
6490 additional output lines can be written before returning. There is also an
6491 smtp_fflush() function to enable the detection of a dropped connection.
6492 The variables smtp_input and smtp_batched_input are exported to
6495 55. Changed the default runtime configuration: the message "Unknown user"
6496 has been removed from the ACL, and instead placed on the localuser router,
6497 using the cannot_route_message feature. This means that any verification
6498 failures that generate their own messages won't get overridden. Similarly,
6499 the "Unrouteable address" message that was in the ACL for unverifiable
6500 relay addresses has also been removed.
6502 56. Added hosts_avoid_esmtp to the smtp transport.
6504 57. The exicyclog script was not checking for the esoteric option
6505 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6506 will work only if exicyclog is run under the appropriate euid.
6508 58. Following a discussion on the list, the rules by which Exim recognises line
6509 endings on incoming messages have been changed. The -dropcr and drop_cr
6510 options are now no-ops, retained only for backwards compatibility. The
6511 following line terminators are recognized: LF CRLF CR. However, special
6512 processing applies to CR:
6514 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6515 nor a local message in the state where . is a terminator.
6517 (ii) If a bare CR is encountered in a header line, an extra space is added
6518 after the line terminator so as not to end the header. The reasoning
6519 behind this is that bare CRs in header lines are most likely either
6520 to be mistakes, or people trying to play silly games.
6522 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6523 was being incorrectly given as 18 bytes larger than it should have been.
6524 This is a VOB (very old bug).
6526 60. This may never have affected anything current, but just in case it has:
6527 When the local host is found other than at the start of a list of hosts,
6528 the local host, those with the same MX, and any that follow, are discarded.
6529 When the list in question was part of a longer list of hosts, the following
6530 hosts (not currently being processed) were also being discarded. This no
6531 longer happens. I'm not sure if this situation could ever has previously
6534 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6537 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6538 header. This is something that is recommended in a new Internet Draft, and
6539 is something that is documented as being done by Sendmail. There are two
6540 possible values. For messages generated by the autoreply transport, Exim
6543 Auto-Submitted: auto-replied
6545 whereas for all other generated messages (e.g. bounces) it adds
6547 Auto-Submitted: auto-generated
6549 63. The "personal" condition in filters now includes a test for the
6550 Auto-Submitted: header. If it contains the string "auto-" the message it
6551 not considered personal.
6553 64. Added rcpt_include_affixes as a generic transport option.
6555 65. Added queue_only_override (default true).
6557 66. Added the syslog_duplication option.
6559 67. If what should have been the first header line of a message consisted of
6560 a space followed by a colon, Exim was mis-interpreting it as a header line.
6561 It isn't of course - it is syntactically invalid and should therefore be
6562 treated as the start of the message body. The misbehaviour could have
6563 caused a number of strange effects, including loss of data in subsequent
6564 header lines, and spool format errors.
6566 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6567 client host had authenticated. This control can now be exercised by an ACL
6568 for more flexibility.
6570 69. By default, callouts do not happen when testing with -bh. There is now a
6571 variant, -bhc, which does actually run the callout code, including
6572 consulting and updating the callout cache.
6574 70. Added support for saslauthd authentication, courtesy of Alexander
6577 71. If statvfs() failed on the spool or log directories while checking their
6578 size for availability, Exim confusingly gave the error "space shortage".
6579 Furthermore, in debugging mode it crashed with a floating point exception.
6580 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6581 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6582 serious problem, Exim now writes to the main and panic logs when this
6583 happens, with details of the failure. It then refuses to accept the
6584 incoming message, giving the message "spool directory problem" or "log
6585 directory problem" with a 421 code for SMTP messages.
6587 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6588 0, 1, and 2 exist, because some OS complain for execs without them (see
6589 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6590 descriptors. However, the code omitted to check that the open succeeded,
6591 causing mysterious errors if for some reason the permissions on /dev/null
6592 got screwed. Now Exim writes a message to the main and panic logs, and
6593 bombs out if it can't open /dev/null.
6595 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6596 interact so that it is all more flexible. It is supposed to remain
6597 backwards compatible. Also added extra_local_interfaces.
6599 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6600 to bomb out with an assertion failure - to the client this appears as a
6601 connection drop. This problem occurs in the part of the code that was taken
6602 from the Samba project. Fortunately, the assertion is in a very simple
6603 function, so I have fixed this by reproducing the function inline in the
6604 one place where it is called, and arranging for authentication to fail
6605 instead of killing the process with assert().
6607 75. The SPA client code was not working when the server requested OEM rather
6608 than Unicode encoding.
6610 76. Added code to make require_files with a specific uid setting more usable in
6611 the case where statting the file as root fails - usually a non-root-mounted
6612 NFS file system. When this happens and the failure is EACCES, Exim now
6613 forks a subprocess and does the per-uid checking as the relevant uid.
6615 77. Added process_log_path.
6617 78. If log_file_path was not explicitly set, a setting of check_log_space or
6618 check_log_inodes was ignored.
6620 79. If a space check for the spool or log partitions fails, the incident is now
6621 logged. Of course, in the latter case the data may get lost...
6623 80. Added the %p formatting code to string_format() so that it can be used to
6624 print addresses in debug_print(). Adjusted all the address printing in the
6625 debugging in store.c to use %p rather than %d.
6627 81. There was a concern that a line of code in smtp_in.c could overflow a
6628 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6629 initially expressed, the concern was not well-founded, because trailing
6630 spaces are removed early. However, if the trailing spaces were followed by
6631 a NULL, they did not get removed, so the overflow was possible. Two fixes
6634 (a) I re-wrote the offending code in a cleaner fashion.
6635 (b) If an incoming SMTP command contains a NULL character, it is rejected
6638 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6639 runs initgroups(), so that if the Exim user is in any additional groups,
6640 they will be used during message reception.
6646 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob