1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 selfsigned certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the sourcecode
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
105 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
106 and 3.4.4 - once the server is enabled to respond to an OCSP request
107 it does even when not requested, resulting in a stapling non-aware
108 client dropping the TLS connection.
110 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
111 support variable-length bit vectors. No functional change.
113 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
114 The I= interface field on outgoing lines is now after the H= remote
115 host field, same as incoming lines. There is a separate
116 outgoing_interface log selector which allows you to disable the
119 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
120 If not running log_selector +smtp_connection the mainlog would be held
121 open indefinitely after a "too many connections" event, including to a
122 deleted file after a log rotate. Leave the per net connection logging
123 leaving it open for efficiency as that will be quickly detected by the
124 check on the next write.
126 HS/01 Bug 1671: Fix post transport crash.
127 Processing the wait-<transport> messages could crash the delivery
128 process if the message IDs didn't exist for some reason. When
129 using 'split_spool_directory=yes' the construction of the spool
130 file name failed already, exposing the same netto behaviour.
132 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
133 mime_regex ACL conditions.
135 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
136 to DSN fail messages (bounces): remote IP, remote greeting, remote response
137 to HELO, local diagnostic string.
139 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
140 log line to debug. Even when configured with a tls authenticator many
141 client connections are expected to not authenticate in this way, so
142 an authenticate fail is not an error.
144 HS/02 Add the Exim version string to the process info. This way exiwhat
145 gives some more detail about the running daemon.
147 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
148 matter for fast-change records such as DNSBLs.
150 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
151 retry record, even if constant. There may be multiple transports with
152 different interface settings and the retry behaviour needs to be kept
155 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
157 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
159 JH/10 Bug 840: fix log_defer_output option of pipe transport
161 JH/11 Bug 830: use same host for all RCPTS of a message, even under
162 hosts_randomize. This matters a lot when combined with mua_wrapper.
164 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
165 ${quote_pgsql:<string>} operator.
167 JH/13 Bug 1708: avoid misaligned access in cached lookup.
169 JH/14 Change header file name for freeradius-client. Relevant if compiling
170 with Radius support; from the Gentoo tree and checked under Fedora.
172 JH/15 Bug 1712: Introduce $prdr_requested flag variable
174 JH/16 Bug 1714: Permit an empty string as expansion result for transport
175 option transport_filter, meaning no filtering.
177 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
179 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
180 defaults to "*" (all hosts). The variable is now available when not built
181 with TLS, default unset, mainly to enable keeping the testuite sane.
182 If a server certificate is not supplied (via tls_certificate) an error is
183 logged, and clients will find TLS connections fail on startup. Presumably
184 they will retry in-clear.
185 Packagers of Exim are strongly encouraged to create a server certificate
186 at installation time.
188 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
189 with the $config_file variable.
191 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
192 in transport context, after the attempt, and per-recipient. The latter type
193 is per host attempted. The event data is the error message, and the errno
194 information encodes the lookup type (A vs. MX) used for the (first) host,
195 and the trailing two digits of the smtp 4xx reponse.
197 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
198 to write to mainlog (or rejectlog, paniclog) in the window between file
199 creation and permissions/ownership being changed. Particularly affects
200 installations where exicyclog is run as root, rather than exim user;
201 result is that the running daemon panics and dies.
203 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
205 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
206 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
207 "pri" and "weight". Note that the previous implicit priority given by the
208 list order is no longer honoured.
210 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
213 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
214 by defining SUPPORT_SOCKS.
216 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
217 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
218 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
219 variables are renamed to proxy_{local,external}_{address,port}.
221 JH/27 Move Internationalisation support from Experimental to mainline, enabled
222 for a build by defining SUPPORT_I18N
224 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
225 of the query string, and make ${quote_redis:} do that quoting.
227 JH/29 Move Events support from Experimental to mainline, enabled by default
228 and removable for a build by defining DISABLE_EVENT.
230 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
232 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
233 cached by the daemon.
235 JH/32 Move Redis support from Experimental to mainline, enabled for a build
236 by defining LOOKUP_REDIS. The libhiredis library is required.
238 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
239 keys are given for lookup.
241 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
242 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
243 only supported when built with TLS support. The PolarSSL SHA routines
244 are still used when the TLS library is too old for convenient support.
246 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
247 openssl_options), for security. OpenSSL forces this from version 1.1.0
248 server-side so match that on older versions.
250 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
251 allocation for $value could be released as the expansion processing
252 concluded, but leaving the global pointer active for it.
254 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
255 and to use the domains and local_parts ACL conditions.
257 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
258 incorrectly not doubled on cutthrough transmission, hence seen as a
259 body-termination at the receiving system - resulting in truncated mails.
260 Commonly the sender saw a TCP-level error, and retransmitted the message
261 via the normal store-and-forward channel. This could result in duplicates
262 received - but deduplicating mailstores were liable to retain only the
263 initial truncated version.
265 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
267 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
269 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
270 we're in there, support oversigning also; bug 1309.
272 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
274 HS/04 Add support for keep_environment and add_environment options.
276 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
277 either intentional arithmetic overflow during PRNG, or testing config-
280 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
281 delivery resulted in actual delivery. Cancel cutthrough before DATA
284 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
285 rejects a recipient: pass the reject to the originator.
287 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
288 Many were false-positives and ignorable, but it's worth fixing the
291 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
292 for the new environment-manipulation done at startup. Move the routines
293 from being local to tls.c to being global via the os.c file.
295 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
296 an extract embedded as result-arg for a map, the first arg for extract
297 is unavailable so we cannot tell if this is a numbered or keyed
298 extraction. Accept either.
304 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
307 JH/02 The smtp transport option "multi_domain" is now expanded.
309 JH/03 The smtp transport now requests PRDR by default, if the server offers
312 JH/04 Certificate name checking on server certificates, when exim is a client,
313 is now done by default. The transport option tls_verify_cert_hostnames
314 can be used to disable this per-host. The build option
315 EXPERIMENTAL_CERTNAMES is withdrawn.
317 JH/05 The value of the tls_verify_certificates smtp transport and main options
318 default to the word "system" to access the system default CA bundle.
319 For GnuTLS, only version 3.0.20 or later.
321 JH/06 Verification of the server certificate for a TLS connection is now tried
322 (but not required) by default. The verification status is now logged by
323 default, for both outbound TLS and client-certificate supplying inbound
326 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
329 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
330 Status Notification (bounce) messages are now MIME format per RFC 3464.
331 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
332 under the control of the dsn_advertise_hosts option, and routers may
333 have a dsn_lasthop option.
335 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
336 default, modifiable by a malware= option. The list separator for
337 the options can now be changed in the usual way. Bug 68.
339 JH/10 The smtp_receive_timeout main option is now expanded before use.
341 JH/11 The incoming_interface log option now also enables logging of the
342 local interface on delivery outgoing connections.
344 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
345 if the interface and destination host and port all match.
347 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
350 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
351 Patch from Andrew Lewis.
353 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
354 now supports optional time-restrictions, weighting, and priority
355 modifiers per server. Patch originally by <rommer@active.by>.
357 JH/16 The spamd_address main option now supports a mixed list of local
358 and remote servers. Remote servers can be IPv6 addresses, and
359 specify a port-range.
361 JH/17 Bug 68: The spamd_address main option now supports an optional
362 timeout value per server.
364 JH/18 Bug 1581: Router and transport options headers_add/remove can
365 now have the list separator specified.
367 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
370 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
373 JH/21 Support for the A6 type of dns record is withdrawn.
375 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
376 rather than the verbs used.
378 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
379 from 255 to 1024 chars.
381 JH/24 Verification callouts now attempt to use TLS by default.
383 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
384 are generic router options now. The defaults didn't change.
386 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
387 Original patch from Alexander Shikoff, worked over by JH.
389 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
390 files if ZCAT_COMMAND is not executable.
392 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
394 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
396 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
397 Normally benign, it bites when the pair was led to by a CNAME;
398 modern usage is to not canoicalize the domain to a CNAME target
399 (and we were inconsistent anyway for A-only vs AAAA+A).
401 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
403 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
404 when evaluating $sender_host_dnssec.
406 JH/31 Check the HELO verification lookup for DNSSEC, adding new
407 $sender_helo_dnssec variable.
409 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
411 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
413 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
415 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
416 documented as working, but never had. Support all but $spam_report.
418 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
419 added for tls authenticator.
421 HS/03 Add perl_taintmode main config option
426 TL/01 When running the test suite, the README says that variables such as
427 no_msglog_check are global and can be placed anywhere in a specific
428 test's script, however it was observed that placement needed to be near
429 the beginning for it to behave that way. Changed the runtest perl
430 script to read through the entire script once to detect and set these
431 variables, reset to the beginning of the script, and then run through
432 the script parsing/test process like normal.
434 TL/02 The BSD's have an arc4random API. One of the functions to induce
435 adding randomness was arc4random_stir(), but it has been removed in
436 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
437 function when detected.
439 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
440 cause callback expansion.
442 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
443 syntax errors in an expansion can be treated as a string instead of
444 logging or causing an error, due to the internal use of bool_lax
445 instead of bool when processing it.
447 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
448 server certificates when making smtp deliveries.
450 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
452 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
454 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
456 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
457 Merged patch from Sebastian Wiedenroth.
459 JH/05 Fix results-pipe from transport process. Several recipients, combined
460 with certificate use, exposed issues where response data items split
461 over buffer boundaries were not parsed properly. This eventually
462 resulted in duplicates being sent. This issue only became common enough
463 to notice due to the introduction of conection certificate information,
464 the item size being so much larger. Found and fixed by Wolfgang Breyha.
466 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
467 size buffer was used, resulting in syntax errors when an expansion
470 JH/07 Add support for directories of certificates when compiled with a GnuTLS
471 version 3.3.6 or later.
473 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
474 is EXPERIMENTAL_EVENT, the main-configuration and transport options
475 both become "event_action", the variables become $event_name, $event_data
476 and $event_defer_errno. There is a new variable $verify_mode, usable in
477 routers, transports and related events. The tls:cert event is now also
478 raised for inbound connections, if the main configuration event_action
481 TL/06 In test suite, disable OCSP for old versions of openssl which contained
482 early OCSP support, but no stapling (appears to be less than 1.0.0).
484 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
485 server certificate names available under the smtp transport option
486 "tls_verify_cert_hostname" now do not permit multi-component wildcard
489 JH/10 Time-related extraction expansions from certificates now use the main
490 option "timezone" setting for output formatting, and are consistent
491 between OpenSSL and GnuTLS compilations. Bug 1541.
493 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
494 encoded parameter in the incoming message. Bug 1558.
496 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
497 include certificate info, eximon was claiming there were spoolfile
500 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
502 JH/14 Log delivery-related information more consistently, using the sequence
503 "H=<name> [<ip>]" wherever possible.
505 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
506 are problematic for Debian distribution, omit them from the release
509 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
511 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
514 JH/17 Fix a null-indirection in certextract expansions when a nondefault
515 output list separator was used.
520 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
521 checkers that were complaining about end of non-void function with no
524 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
525 This was a regression intruduced in 4.83 by another bugfix.
527 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
529 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
530 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
536 TF/01 Correctly close the server side of TLS when forking for delivery.
538 When a message was received over SMTP with TLS, Exim failed to clear up
539 the incoming connection properly after forking off the child process to
540 deliver the message. In some situations the subsequent outgoing
541 delivery connection happened to have the same fd number as the incoming
542 connection previously had. Exim would try to use TLS and fail, logging
543 a "Bad file descriptor" error.
545 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
546 utilities have not been installed.
548 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
549 temporary space as the ACL may create new global variables.
551 TL/01 LDAP support uses per connection or global context settings, depending
552 upon the detected version of the libraries at build time.
554 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
555 to extract and use the src ip:port in logging and expansions as if it
556 were a direct connection from the outside internet. PPv2 support was
557 updated based on HAProxy spec change in May 2014.
559 JH/02 Add ${listextract {number}{list}{success}{fail}}.
561 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
562 Properly escape header and check for NULL return.
564 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
567 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
569 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
570 characters in header names, implemented as a verify condition.
571 Contributed by Michael Fischer v. Mollard.
573 TL/05 Rename SPF condition results err_perm and err_temp to standardized
574 results permerror and temperror. Previous values are deprecated but
575 still accepted. In a future release, err_perm and err_temp will be
576 completely removed, which will be a backward incompatibility if the
577 ACL tests for either of these two old results. Patch contributed by
578 user bes-internal on the mailing list.
580 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
582 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
583 selectors, in both main and reject logs.
585 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
588 JH/07 Add malware type "sock" for talking to simple daemon.
590 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
592 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
593 routers/transports under cutthrough routing.
595 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
596 numbers. Touch up "bool" conditional to keep the same definition.
598 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
600 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
602 JH/12 Expand items in router/transport headers_add or headers_remove lists
603 individually rather than the list as a whole. Bug 1452.
605 Required for reasonable handling of multiple headers_ options when
606 they may be empty; requires that headers_remove items with embedded
607 colons must have them doubled (or the list-separator changed).
609 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
610 view the policy declared in the DMARC record. Currently, $dmarc_status
611 is a combined value of both the record presence and the result of the
614 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
616 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
617 dnslookup router and the smtp transport (applying to the forward
620 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
621 of ldap servers used for a specific lookup. Patch provided by Heiko
624 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
625 New variable $lookup_dnssec_authenticated for observability.
627 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
628 Patch submitted by Lars Timman.
630 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
632 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
633 Requires trusted mode and valid format message id, aborts otherwise.
634 Patch contributed by Heiko Schlichting.
636 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
637 certextract with support for various fields. Bug 1358.
639 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
640 is requested by default, modifiable by smtp transport option
643 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
644 operate on certificate variables to give certificate fingerprints
645 Also new ${sha256:cert_variable}.
647 JH/23 The PRDR feature is moved from being Experimental into the mainline.
649 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
650 Christian Aistleitner.
652 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
654 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
655 file. Patch from Wolfgang Breyha.
657 JH/25 Expand the coverage of the delivery $host and $host_address to
658 client authenticators run in verify callout. Bug 1476.
660 JH/26 Port service names are now accepted for tls_on_connect_ports, to
661 align with daemon_smtp_ports. Bug 72.
663 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
664 support and error reporting did not work properly.
666 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
667 and is readable. Patch from Andrew Colin Kissa.
669 TL/14 Enhance documentation of ${run expansion and how it parses the
670 commandline after expansion, particularly in the case when an
671 unquoted variable expansion results in an empty value.
673 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
675 PP/02 Fix internal collision of T_APL on systems which support RFC3123
676 by renaming away from it. Addresses GH issue 15, reported by
679 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
681 TL/15 SECURITY: prevent double expansion in math comparison functions
682 (can expand unsanitized data). Not remotely exploitable.
689 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
691 PP/02 Make -n do something, by making it not do something.
692 When combined with -bP, the name of an option is not output.
694 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
697 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
698 $sender_host_name and config options to manage this, and basic check
701 PP/05 DSCP support for outbound connections and control modifier for inbound.
703 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
704 (Only plugin which currently uses this is kerberos4, which nobody should
705 be using, but we should make it available and other future plugins might
706 conceivably use it, even though it would break NAT; stuff *should* be
707 using channel bindings instead).
709 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
710 name; added for Sendmail compatibility; requires admin caller.
711 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
712 just ignore it); requires trusted caller.
713 Also parse but ignore: -Ac -Am -X<logfile>
716 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
718 TL/02 Add +smtp_confirmation as a default logging option.
720 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
721 Patch by Magnus Holmgren from 2007-02-20.
723 TL/04 Bugzilla 1281 - Spec typo.
724 Bugzilla 1283 - Spec typo.
725 Bugzilla 1290 - Spec grammar fixes.
727 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
729 TL/06 Add Experimental DMARC support using libopendmarc libraries.
731 TL/07 Fix an out of order global option causing a segfault. Reported to dev
732 mailing list by by Dmitry Isaikin.
734 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
736 JH/02 Support "G" suffix to numbers in ${if comparisons.
738 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
740 NM/01 Bugzilla 1197 - Spec typo
741 Bugzilla 1196 - Spec examples corrections
743 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
745 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
746 gnutls_enable_pkcs11, but renamed to more accurately indicate its
749 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
750 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
752 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
753 "acl {{name}{arg}...}", and optional args on acl condition
756 JH/05 Permit multiple router/transport headers_add/remove lines.
758 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
760 JH/07 Avoid using a waiting database for a single-message-only transport.
761 Performance patch from Paul Fisher. Bugzilla 1262.
763 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
766 JH/09 Add $headers_added variable, with content from use of ACL modifier
767 add_header (but not yet added to the message). Bugzilla 199.
769 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
770 Pulled from Bugzilla 817 by Wolfgang Breyha.
772 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
774 (nb: this is the same fix as in Exim 4.80.1)
776 JH/11 Add A= logging on delivery lines, and a client_set_id option on
779 JH/12 Add optional authenticated_sender logging to A= and a log_selector
782 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
784 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
785 advertise SMTP AUTH mechanism to us, instead of a generic
786 protocol violation error. Also, make Exim more robust to bad
787 data from the Dovecot auth socket.
789 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
791 When a queue runner is handling a message, Exim first routes the
792 recipient addresses, during which it prunes them based on the retry
793 hints database. After that it attempts to deliver the message to
794 any remaining recipients. It then updates the hints database using
797 So if a recipient address works intermittently, it can get repeatedly
798 deferred at routing time. The retry hints record remains fresh so the
799 address never reaches the final cutoff time.
801 This is a fairly common occurrence when a user is bumping up against
802 their storage quota. Exim had some logic in its local delivery code
803 to deal with this. However it did not apply to per-recipient defers
804 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
806 This change adds a proper retry rule check during routing so that the
807 final cutoff time is checked against the message's age. We only do
808 this check if there is an address retry record and there is not a
809 domain retry record; this implies that previous attempts to handle
810 the address had the retry_use_local_parts option turned on. We use
811 this as an approximation for the destination being like a local
812 delivery, as in LMTP.
814 I suspect this new check makes the old local delivery cutoff check
815 redundant, but I have not verified this so I left the code in place.
817 TF/02 Correct gecos expansion when From: is a prefix of the username.
819 Test 0254 submits a message to Exim with the header
823 When I ran the test suite under the user fanf2, Exim expanded
824 the header to contain my full name, whereas it should have added
825 a Resent-Sender: header. It erroneously treats any prefix of the
826 username as equal to the username.
828 This change corrects that bug.
830 GF/01 DCC debug and logging tidyup
831 Error conditions log to paniclog rather than rejectlog.
832 Debug lines prefixed by "DCC: " to remove any ambiguity.
834 TF/03 Avoid unnecessary rebuilds of lookup-related code.
836 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
837 Bug spotted by Jeremy Harris; was flawed since initial commit.
838 Would have resulted in OCSP responses post-SNI triggering an Exim
839 NULL dereference and crash.
841 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
843 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
844 Bug detection, analysis and fix by Samuel Thibault.
845 Bugzilla 1331, Debian bug #698092.
847 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
849 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
850 Server implementation by Todd Lyons, client by JH.
851 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
852 config variable "prdr_enable" controls whether the server
853 advertises the facility. If the client requests PRDR a new
854 acl_data_smtp_prdr ACL is called once for each recipient, after
855 the body content is received and before the acl_smtp_data ACL.
856 The client is controlled by bolth of: a hosts_try_prdr option
857 on the smtp transport, and the server advertisement.
858 Default client logging of deliveries and rejections involving
859 PRDR are flagged with the string "PRDR".
861 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
862 fclose(). Diagnosis by Todd Lyons.
864 PP/17 Update configure.default to handle IPv6 localhost better.
865 Patch by Alain Williams (plus minor tweaks).
868 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
869 This is now consistent with GnuTLS, and is now documented: the
870 previous undocumented portable approach to treating the option as
871 unset was to force an expansion failure. That still works, and
872 an empty string is now equivalent.
874 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
875 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
876 not performing validation itself.
878 PP/20 Added force_command boolean option to pipe transport.
879 Patch from Nick Koston, of cPanel Inc.
881 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
884 TF/04 Added udpsend ACL modifer and hexquote expansion operator
886 PP/21 Fix eximon continuous updating with timestamped log-files.
887 Broken in a format-string cleanup in 4.80, missed when I repaired the
888 other false fix of the same issue.
889 Report and fix from Heiko Schlichting.
892 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
893 Report from Prashanth Katuri.
895 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
896 It's SecureTransport, so affects any MacOS clients which use the
897 system-integrated TLS libraries, including email clients.
899 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
900 using a MIME ACL for non-SMTP local injection.
901 Report and assistance in diagnosis by Warren Baker.
903 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
905 JH/16 Fix comparisons for 64b. Bugzilla 1385.
907 TL/09 Add expansion variable $authenticated_fail_id to keep track of
908 last id that failed so it may be referenced in subsequent ACL's.
910 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
913 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
914 ldap library initialization, allowing self-signed CA's to be
915 used. Also properly sets require_cert option later in code by
916 using NULL (global ldap config) instead of ldap handle (per
917 session). Bug diagnosis and testing by alxgomz.
919 TL/12 Enhanced documentation in the ratelimit.pl script provided in
920 the src/util/ subdirectory.
922 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
923 renamed to Transport Post Delivery Action by Jeremy Harris, as
926 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
927 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
928 redis_servers = needs to be configured which will be used by the redis
929 lookup. Patch from Warren Baker, of The Packet Hub.
931 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
933 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
934 hostname or reverse DNS when processing a host list. Used suggestions
935 from multiple comments on this bug.
937 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
939 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
940 Missed a few lines, added it to make the runtest require no keyboard
943 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
944 contains upper case chars. Make router use caseful_local_part.
946 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
947 support when GnuTLS has been built with p11-kit.
953 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
955 This, or similar/improved, will also be change PP/11 of 4.82.
961 PP/01 Handle short writes when writing local log-files.
962 In practice, only affects FreeBSD (8 onwards).
963 Bugzilla 1053, with thanks to Dmitry Isaikin.
965 NM/01 Bugzilla 949 - Documentation tweak
967 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
970 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
972 PP/02 Implemented gsasl authenticator.
974 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
976 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
977 `pkg-config foo` for cflags/libs.
979 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
980 with rest of GSASL and with heimdal_gssapi.
982 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
983 `pkg-config foo` for cflags/libs for the TLS implementation.
985 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
986 properties get this fed in as external SSF. A number of robustness
987 and debugging improvements to the cyrus_sasl authenticator.
989 PP/08 cyrus_sasl server now expands the server_realm option.
991 PP/09 Bugzilla 1214 - Log authentication information in reject log.
992 Patch by Jeremy Harris.
994 PP/10 Added dbmjz lookup type.
996 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
998 PP/12 MAIL args handles TAB as well as SP, for better interop with
999 non-compliant senders.
1000 Analysis and variant patch by Todd Lyons.
1002 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1003 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1004 Patch from Dirk Mueller <dmueller@suse.com>
1006 PP/13 tls_peerdn now print-escaped for spool files.
1007 Observed some $tls_peerdn in wild which contained \n, which resulted
1008 in spool file corruption.
1010 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1011 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1012 or write after TLS renegotiation, which otherwise led to messages
1015 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1016 as a tracking header (ie: a signed header comes before the signature).
1017 Patch from Wolfgang Breyha.
1019 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1020 comma-sep list; embedded commas doubled.
1022 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1024 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1026 Report and patch from Dmitry Banschikov.
1028 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1029 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1030 failures. We appear to now support TLS1.1+ with Exim.
1032 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1033 lets Exim select keys and certificates based upon TLS SNI from client.
1034 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1035 before an outbound SMTP session. New log_selector, +tls_sni.
1037 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1038 NULL dereference. Report and patch from Alun Jones.
1040 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1041 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1042 Not seeing resolver debug output on NetBSD, but suspect this is a
1043 resolver implementation change.
1045 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1046 Left warnings. Added "eximon gdb" invocation mode.
1048 PP/21 Defaulting "accept_8bitmime" to true, not false.
1050 PP/22 Added -bw for inetd wait mode support.
1052 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1053 locate the relevant includes and libraries. Made this the default.
1055 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1056 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1058 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1059 This may cause build issues on older platforms.
1061 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1062 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1063 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1064 Added SNI support via GnuTLS too.
1065 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1067 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1069 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1070 Applied second patch from Janne, implementing suggestion to default
1071 multiple-strings-in-record handling to match SPF spec.
1073 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1075 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1076 read-only, out of scope).
1077 Patch from Wolfgang Breyha, report from Stuart Northfield.
1079 PP/29 Fix three issues highlighted by clang analyser static analysis.
1080 Only crash-plausible issue would require the Cambridge-specific
1081 iplookup router and a misconfiguration.
1082 Report from Marcin Mirosław.
1084 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1086 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1087 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1088 As part of this, removing so much warning spew let me fix some minor
1089 real issues in debug logging.
1091 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1092 assignment on my part. Fixed.
1094 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1095 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1096 Janne Snabb (who went above and beyond: thank you).
1098 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1099 string otherwise requires a connection and a bunch more work and it's
1100 relatively easy to get wrong. Should also expose TLS library linkage
1103 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1104 64-bit ${eval} (JH/03).
1106 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1107 GNU libc to support some of the 64-bit stuff, should not lead to
1108 conflicts. Defined before os.h is pulled in, so if a given platform
1109 needs to override this, it can.
1111 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1112 protection layer was required, which is not implemented.
1113 Bugzilla 1254, patch from Wolfgang Breyha.
1115 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1116 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1117 tls_dhparam take prime identifiers. Also unbreak combination of
1118 OpenSSL+DH_params+TLSSNI.
1120 PP/39 Disable SSLv2 by default in OpenSSL support.
1126 PP/01 Solaris build fix for Oracle's LDAP libraries.
1127 Bugzilla 1109, patch from Stephen Usher.
1129 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1131 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1134 TF/02 Fix a couple more cases where we did not log the error message
1135 when unlink() failed. See also change 4.74-TF/03.
1137 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1138 lock up or crash if it happened to be inside a call to libc when it
1139 got a SIGUSR1 from exiwhat.
1141 The SIGUSR1 handler appends the current process status to the process
1142 log which is later printed by exiwhat. It used to use the general
1143 purpose logging code to do this, but several functions it calls are
1144 not safe for signals.
1146 The new output code in the SIGUSR1 handler is specific to the process
1147 log, and simple enough that it's easy to inspect for signal safety.
1148 Removing some special cases also simplifies the general logging code.
1149 Removing the spurious timestamps from the process log simplifies
1152 TF/04 Improved ratelimit ACL condition.
1154 The /noupdate option has been deprecated in favour of /readonly which
1155 has clearer semantics. The /leaky, /strict, and /readonly update modes
1156 are mutually exclusive. The update mode is no longer included in the
1157 database key; it just determines when the database is updated. (This
1158 means that when you upgrde Exim will forget old rate measurements.)
1160 Exim now checks that the per_* options are used with an update mode that
1161 makes sense for the current ACL. For example, when Exim is processing a
1162 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1163 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1164 must specify per_mail/readonly. If you omit the update mode it defaults to
1165 /leaky where that makes sense (as before) or /readonly where required.
1167 The /noupdate option is now undocumented but still supported for
1168 backwards compatibility. It is equivalent to /readonly except that in
1169 ACLs where /readonly is required you may specify /leaky/noupdate or
1170 /strict/noupdate which are treated the same as /readonly.
1172 A useful new feature is the /count= option. This is a generalization
1173 of the per_byte option, so that you can measure the throughput of other
1174 aggregate values. For example, the per_byte option is now equivalent
1175 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1177 The per_rcpt option has been generalized using the /count= mechanism
1178 (though it's more complicated than the per_byte equivalence). When it is
1179 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1180 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1181 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1182 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1183 Note that using per_rcpt with a non-readonly update mode in more than
1184 one ACL will cause the recipients to be double-counted. (The per_mail
1185 and per_byte options don't have this problem.)
1187 The handling of very low rates has changed slightly. If the computed rate
1188 is less than the event's count (usually one) then this event is the first
1189 after a long gap. In this case the rate is set to the same as this event's
1190 count, so that the first message of a spam run is counted properly.
1192 The major new feature is a mechanism for counting the rate of unique
1193 events. The new per_addr option counts the number of different
1194 recipients that someone has sent messages to in the last time period. It
1195 behaves like per_rcpt if all the recipient addresses are different, but
1196 duplicate recipient addresses do not increase the measured rate. Like
1197 the /count= option this is a general mechanism, so the per_addr option
1198 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1199 example, measure the rate that a client uses different sender addresses
1200 with the options per_mail/unique=$sender_address. There are further
1201 details in the main documentation.
1203 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1205 TF/06 Removed a few PCRE remnants.
1207 TF/07 Automatically extract Exim's version number from tags in the git
1208 repository when doing development or release builds.
1210 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1211 Bugzilla 879. Patch from Paul Fisher.
1213 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1214 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1217 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1219 PP/05 Variable $av_failed, true if the AV scanner deferred.
1220 Bugzilla 1078. Patch from John Horne.
1222 PP/06 Stop make process more reliably on build failure.
1223 Bugzilla 1087. Patch from Heiko Schlittermann.
1225 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1226 Bugzilla 1089. Patch from Heiko Schlittermann.
1228 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1229 Bugzilla 1131. Patch from Holger Weiß.
1231 PP/09 Handle IPv6 addresses with SPF.
1232 Bugzilla 860. Patch from Wolfgang Breyha.
1234 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1236 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1239 PP/11 match_* no longer expand right-hand-side by default.
1240 New compile-time build option, EXPAND_LISTMATCH_RHS.
1241 New expansion conditions, "inlist", "inlisti".
1243 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1245 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1247 PP/14 fix log_write() format string regression from TF/03.
1248 Bugzilla 1152. Patch from Dmitry Isaikin.
1254 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1256 PP/02 Harmonised TLS library version reporting; only show if debugging.
1257 Layout now matches that introduced for other libraries in 4.74 PP/03.
1259 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1261 PP/04 New "dns_use_edns0" global option.
1263 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1266 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1267 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1269 TK/01 Updated PolarSSL code to 0.14.2.
1270 Bugzilla 1097. Patch from Andreas Metzler.
1272 PP/07 Catch divide-by-zero in ${eval:...}.
1273 Fixes bugzilla 1102.
1275 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1278 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1279 format-string attack -- SECURITY: remote arbitrary code execution.
1281 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1282 time unintentionally subject to list matching rules, letting the header
1283 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1284 arbitrary string expansion). This allowed for information disclosure.
1286 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1287 INT_MIN/-1 -- value coerced to INT_MAX.
1293 NM/01 Workround for PCRE version dependency in version reporting
1296 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1297 This fixes portability to compilers other than gcc, notably
1298 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1300 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1301 makefiles for portability to HP-UX and POSIX correctness.
1303 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1304 Also via indented variable definition in the Makefile.
1305 (Debugging by Oliver Heesakkers).
1307 PP/02 Restore caching of spamd results with expanded spamd_address.
1308 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1310 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1311 Improves build reliability. Fix from: Frank Elsner
1313 NM/02 Fix wide character breakage in the rfc2047 coding
1314 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1316 NM/03 Allow underscore in dnslist lookups
1317 Fixes bug 1026. Patch from Graeme Fowler
1319 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1320 Code patches from Adam Ciarcinski of NetBSD.
1322 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1325 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1326 is logged, to avoid truncation. Patch from John Horne.
1328 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1329 Patch from Jakob Hirsch.
1331 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1332 SQL string expansion failure details.
1333 Patch from Andrey Oktyabrski.
1335 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1336 Patch from Simon Arlott.
1338 PP/09 New lookups functionality failed to compile on old gcc which rejects
1339 extern declarations in function scope.
1340 Patch from Oliver Fleischmann
1342 PP/10 Use sig_atomic_t for flags set from signal handlers.
1343 Check getgroups() return and improve debugging.
1344 Fixed developed for diagnosis in bug 927 (which turned out to be
1347 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1348 Patch from Mark Zealey.
1350 PP/12 Bugzilla 1056: Improved spamd server selection.
1351 Patch from Mark Zealey.
1353 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1354 Based on patch from Heiko Schlittermann.
1356 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1357 Patch from Uwe Doering, sign-off by Michael Haardt.
1359 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1360 variable declaration deep within a block. Bug and patch from
1363 PP/15 lookups-Makefile IRIX compatibilty coercion.
1365 PP/16 Make DISABLE_DKIM build knob functional.
1367 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1368 Patch by Simon Arlott
1370 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1371 variable argument macros. Our copy now differs from upstream.
1377 TF/01 Failure to get a lock on a hints database can have serious
1378 consequences so log it to the panic log.
1380 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1381 controlled using the smtp_confirmation log selector.
1383 TF/03 Include the error message when we fail to unlink a spool file.
1385 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1386 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1387 for maintaining out-of-tree patches for some time.
1389 PP/01 Bugzilla 139: Documentation and portability issues.
1390 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1391 Handle per-OS dynamic-module compilation flags.
1393 PP/02 Let /dev/null have normal permissions.
1394 The 4.73 fixes were a little too stringent and complained about the
1395 permissions on /dev/null. Exempt it from some checks.
1396 Reported by Andreas M. Kirchwitz.
1398 PP/03 Report version information for many libraries, including
1399 Exim version information for dynamically loaded libraries. Created
1400 version.h, now support a version extension string for distributors
1401 who patch heavily. Dynamic module ABI change.
1403 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1404 privilege escalation vulnerability whereby the Exim run-time user
1405 can cause root to append content of the attacker's choosing to
1408 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1411 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1412 If dropping privileges for untrusted macros, we disabled normal logging
1413 on the basis that it would fail; for the Exim run-time user, this is not
1414 the case, and it resulted in successful deliveries going unlogged.
1415 Fixed. Reported by Andreas Metzler.
1421 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1422 only prepend for the Resent-* case. Fixes regression introduced in
1423 Exim 4.70 by NM/22 for Bugzilla 607.
1425 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1426 increasing numbers of administrators be bitten by this.
1428 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1430 PP/03 Bugzilla 994: added openssl_options main configuration option.
1432 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1434 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1436 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1438 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1439 without a peer certificate, leading to a segfault because of an
1440 assumption that peers always have certificates. Be a little more
1441 paranoid. Problem reported by Martin Tscholak.
1443 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1444 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1445 NB: ClamAV planning to remove STREAM in "middle of 2010".
1446 CL also introduces -bmalware, various -d+acl logging additions and
1447 more caution in buffer sizes.
1449 PP/09 Implemented reverse_ip expansion operator.
1451 PP/10 Bugzilla 937: provide a "debug" ACL control.
1453 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1455 PP/12 Bugzilla 973: Implement --version.
1457 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1459 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1461 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1463 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1464 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1465 ignore trailing whitespace.
1467 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1469 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1472 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1473 Notification from Dr Andrew Aitchison.
1475 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1476 ExtendedDetectionInfo response format.
1477 Notification from John Horne.
1479 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1482 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1483 XSL and documented dependency on system catalogs, with examples of how
1486 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1489 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1490 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1491 configuration file which is writeable by the Exim user or group.
1493 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1494 of configuration files to cover files specified with the -C option if
1495 they are going to be used with root privileges, not just the default
1498 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1499 option (effectively making it always true).
1501 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1502 files to be used while preserving root privileges.
1504 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1505 that rogue child processes cannot use them.
1507 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1508 run-time user, instead of root.
1510 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1511 Exim run-time user without dropping privileges.
1513 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1514 result string, instead of calling string_vformat() twice with the same
1517 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1518 for other users. Others should always drop root privileges if they use
1519 -C on the command line, even for a whitelisted configure file.
1521 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1523 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1529 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1530 $data_path, and $header_path variables; fixed documentation bugs and
1533 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1534 exipick to access non-standard spools, including the "frozen" queue
1537 NM/01 Bugzilla 965: Support mysql stored procedures.
1538 Patch from Alain Williams
1540 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1542 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1543 Patch from Andreas Metzler
1545 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1546 Patch from Kirill Miazine
1548 NM/05 Bugzilla 671: Added umask to procmail example.
1550 JJ/03 installed exipick 20100323.0, fixing doc bug
1552 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1553 directory. Notification and patch from Dan Rosenberg.
1555 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1557 TK/02 Improve log output when DKIM signing operation fails.
1559 MH/01 Treat the transport option dkim_domain as a colon separated
1560 list, not as a single string, and sign the message with each element,
1561 omitting multiple occurences of the same signer.
1563 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1564 Bugzilla 985, 986. Patch by Simon Arlott
1566 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1567 Patch by Simon Arlott
1569 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1570 MBX locking. Notification from Dan Rosenberg.
1576 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1578 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1580 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1582 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1584 NM/04 Bugzilla 915: Flag broken perl installation during build.
1590 TK/01 Added patch by Johannes Berg that expands the main option
1591 "spamd_address" if it starts with a dollar sign.
1593 TK/02 Write list of recipients to X-Envelope-Sender header when building
1594 the mbox-format spool file for content scanning (suggested by Jakob
1597 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1598 (http://www.dcc-servers.net/) support via dccifd. Activated by
1599 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1601 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1602 by Mark Daniel Reidel <mr@df.eu>.
1604 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1605 When building exim an external PCRE library is now needed -
1606 PCRE is a system library on the majority of modern systems.
1607 See entry on PCRE_LIBS in EDITME file.
1609 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1610 conversation. Added nologin parameter to request.
1611 Patch contributed by Kirill Miazine.
1613 TF/01 Do not log submission mode rewrites if they do not change the address.
1615 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1617 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1618 log files in place. Contributed by Roberto Lima.
1620 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1622 TF/03 Bugzilla 615: When checking the local_parts router precondition
1623 after a local_part_suffix or local_part_prefix option, Exim now
1624 does not use the address's named list lookup cache, since this
1625 contains cached lookups for the whole local part.
1627 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1628 Robert Millan. Documentation is in experimental-spec.txt.
1630 TF/04 Bugzilla 668: Fix parallel build (make -j).
1632 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1634 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1635 Patch provided by Jan Srzednicki.
1637 TF/05 Leading white space used to be stripped from $spam_report which
1638 wrecked the formatting. Now it is preserved.
1640 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1641 that they are available at delivery time.
1643 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1645 TF/08 TLS error reporting now respects the incoming_interface and
1646 incoming_port log selectors.
1648 TF/09 Produce a more useful error message if an SMTP transport's hosts
1649 setting expands to an empty string.
1651 NM/06 Bugzilla 744: EXPN did not work under TLS.
1652 Patch provided by Phil Pennock.
1654 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1655 Patch provided by Richard Godbee.
1657 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1658 acl_smtp_notquit, added index entry.
1660 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1661 Patch provided by Eugene Bujak.
1663 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1664 accept(). Patch provided by Maxim Dounin.
1666 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1667 Patch provided by Phil Pennock.
1669 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1671 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1672 Patch provided by Brad "anomie" Jorsch.
1674 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1675 Patch provided by Dean Brooks.
1677 TK/05 Add native DKIM support (does not depend on external libraries).
1679 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1680 Patch provided by Graeme Fowler.
1682 NM/16 Bugzilla 851: Documentation example syntax fix.
1684 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1686 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1689 NM/19 Bugzilla 745: TLS version reporting.
1690 Patch provided by Phil Pennock.
1692 NM/20 Bugzilla 167: bool: condition support.
1693 Patch provided by Phil Pennock.
1695 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1696 clients. Patch provided by Phil Pennock.
1698 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1699 Patch provided by Brad "anomie" Jorsch.
1701 NM/23 Bugzilla 687: Fix misparses in eximstats.
1702 Patch provided by Heiko Schlittermann.
1704 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1705 Patch provided by Heiko Schlittermann.
1707 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1708 plus update to original patch.
1710 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1712 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1713 Patch provided by David Brownlee.
1715 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1717 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1719 NM/30 Bugzilla 888: TLS documentation bugfixes.
1721 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1723 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1724 Unlike the original bugzilla I have changed all shell scripts in src tree.
1726 NM/33 Bugzilla 898: Transport filter timeout fix.
1727 Patch by Todd Rinaldo.
1729 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1730 Patch by Serge Demonchaux.
1732 NM/35 Bugzilla 39: Base64 decode bug fixes.
1733 Patch by Jakob Hirsch.
1735 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1737 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1739 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1741 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1747 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1748 ALT-N's libdkim that I have put here:
1749 http://duncanthrax.net/exim-experimental/
1751 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1752 were called 'true' and it seems that C99 defines that as a reserved
1753 keyword to be used with 'bool' variable types. That means you could
1754 not include C99-style headers which use bools without triggering
1755 build errors in sieve.c.
1757 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1758 as mailq or other aliases. Changed the --help handling significantly
1759 to do whats expected. exim_usage() emits usage/help information.
1761 SC/01 Added the -bylocaldomain option to eximstats.
1763 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1765 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1767 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1773 PH/01 Another patch from the Sieve maintainer.
1775 PH/02 When an IPv6 address is converted to a string for single-key lookup
1776 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1777 dots are used instead of colons so that keys in lsearch files need not
1778 contain colons. This was done some time before quoting was made available
1779 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1780 (notated using the quote facility) so as to distinguish them from IPv4
1781 keys. This meant that lookups for IP addresses in host lists did not work
1782 for iplsearch lookups.
1784 This has been fixed by arranging for IPv6 addresses to be expressed with
1785 colons if the lookup type is iplsearch. This is not incompatible, because
1786 previously such lookups could never work.
1788 The situation is now rather anomolous, since one *can* have colons in
1789 ordinary lsearch keys. However, making the change in all cases is
1790 incompatible and would probably break a number of configurations.
1792 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1795 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1796 conversion specification without a maximum field width, thereby enabling
1797 a rogue spamd server to cause a buffer overflow. While nobody in their
1798 right mind would setup Exim to query an untrusted spamd server, an
1799 attacker that gains access to a server running spamd could potentially
1800 exploit this vulnerability to run arbitrary code as the Exim user.
1802 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1803 $primary_hostname instead of what libspf2 thinks the hosts name is.
1805 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1806 a directory entry by the name of the lookup key. Previously, if a
1807 symlink pointed to a non-existing file or a file in a directory that
1808 Exim lacked permissions to read, a lookup for a key matching that
1809 symlink would fail. Now it is enough that a matching directory entry
1810 exists, symlink or not. (Bugzilla 503.)
1812 PH/03 The body_linecount and body_zerocount variables are now exported in the
1815 PH/04 Added the $dnslist_matched variable.
1817 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1818 This means they are set thereafter only if the connection becomes
1821 PH/06 Added the client_condition to authenticators so that some can be skipped
1822 by clients under certain conditions.
1824 PH/07 The error message for a badly-placed control=no_multiline_responses left
1825 "_responses" off the end of the name.
1827 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1829 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1830 (without spaces) instead of just copying the configuration text.
1832 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1834 PH/11 Added $max_received_linelength.
1836 PH/12 Added +ignore_defer and +include_defer to host lists.
1838 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1839 way in which PCRE > 7.0 is built.
1841 PH/14 Implemented queue_only_load_latch.
1843 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1844 MAIL command. The effect was to mangle the value on 64-bit systems.
1846 PH/16 Another patch from the Sieve maintainer.
1848 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1850 PH/18 If a system quota error occurred while trying to create the file for
1851 a maildir delivery, the message "Mailbox is full" was not appended to the
1852 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1853 only to a quota excession during the actual writing of the file.
1855 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1856 characters?) which causes problems in log lines. The DN values are now
1857 passed through string_printing() before being added to log lines.
1859 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1860 and InterBase are left for another time.)
1862 PH/21 Added message_body_newlines option.
1864 PH/22 Guard against possible overflow in moan_check_errorcopy().
1866 PH/23 POSIX allows open() to be a macro; guard against that.
1868 PH/24 If the recipient of an error message contained an @ in the local part
1869 (suitably quoted, of course), incorrect values were put in $domain and
1870 $local_part during the evaluation of errors_copy.
1876 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1877 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1880 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1881 issue a MAIL command.
1883 PH/02 In an ACL statement such as
1885 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1887 if a client was not listed at all, or was listed with a value other than
1888 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1889 the condition was not true (as it should be), so access was not denied.
1890 The bug was that the ! inversion was incorrectly passed on to the second
1891 item. This has been fixed.
1893 PH/03 Added additional dnslists conditions == and =& which are different from
1894 = and & when the dns lookup returns more than one IP address.
1896 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1897 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1899 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1900 FSYNC, which compiles an option called disable_fsync that allows for
1901 bypassing fsync(). The documentation is heavily laced with warnings.
1903 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1905 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1906 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1907 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1908 including adding "make clean"; (3) Added -fPIC when compiling the test
1909 dynamically loaded module, to get rid of a warning.
1911 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1912 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1913 = 0s. The bug is otherwise harmless.
1915 PH/07 There was a bug in the dovecot authenticator such that the value of
1916 $auth1 could be overwritten, and so not correctly preserved, after a
1917 successful authentication. This usually meant that the value preserved by
1918 the server_setid option was incorrect.
1920 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1922 PH/09 Installed PCRE release 7.0.
1924 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1925 run for batched SMTP input. It is now run at the start of every message
1926 in the batch. While fixing this I discovered that the process information
1927 (output by running exiwhat) was not always getting set for -bs and -bS
1928 input. This is fixed, and it now also says "batched" for BSMTP.
1930 PH/11 Added control=no_pipelining.
1932 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1933 patch, slightly modified), and move the expansion of helo_data till after
1934 the connection is made in the smtp transport (so it can use these
1937 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1939 PH/14 Added log_selector = +pid.
1941 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1943 PH/16 Add ${if forany and ${if forall.
1945 PH/17 Added dsn_from option to vary the From: line in DSNs.
1947 PH/18 Flush SMTP output before performing a callout, unless control =
1948 no_callout_flush is set.
1950 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1951 was true (the default) a successful delivery failed to delete the retry
1952 item, thus causing premature timeout of the address. The bug is now
1955 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1957 PH/21 Long custom messages for fakedefer and fakereject are now split up
1958 into multiline reponses in the same way that messages for "deny" and
1959 other ACL rejections are.
1961 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1962 with slight modification.
1964 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1965 draft, changing the syntax and factoring some duplicate code".
1967 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1968 for deliveries of the second and subsequent messages over the same SMTP
1971 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1972 ${reduce, with only minor "tidies".
1974 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1976 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1977 expansion side effects.
1979 PH/27 When a message times out after an over-quota error from an Exim-imposed
1980 quota, the bounce message says "mailbox is full". This message was not
1981 being given when it was a system quota that was exceeded. It now should
1984 MH/03 Made $recipients available in local_scan(). local_scan() already has
1985 better access to the recipient list through recipients_list[], but
1986 $recipients can be useful in postmaster-provided expansion strings.
1988 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1989 in the case of a MAIL command with additional options following the
1990 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1991 were accidentally chopped off.
1993 PH/29 SMTP synchronization checks are implemented when a command is read -
1994 there is a check that no more input is waiting when there shouldn't be
1995 any. However, for some commands, a delay in an ACL can mean that it is
1996 some time before the response is written. In this time, more input might
1997 arrive, invalidly. So now there are extra checks after an ACL has run for
1998 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1999 pipelining has not been advertised.
2001 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2003 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2004 returned in the SMTP response when smtp_return_error_details was false.
2005 This has been fixed.
2007 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2008 instead of the C I/O that was originally supplied, because problems were
2009 reported on Solaris.
2011 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2012 Exim which did not show up earlier: it was assuming that a call to
2013 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2014 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2015 was a macro that became an assignment, so it seemed to work. This has
2016 changed to a proper function call with a void return, hence the compile
2017 error. Exim's code has been fixed.
2019 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2022 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2024 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2026 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2027 because it turns out to be too big or there is a timeout) there is no
2028 "Completed" line in the log. When some messages of this type were
2029 selected by exigrep, they were listed as "not completed". Others were
2030 picked up by some special patterns. I have improved the selection
2031 criteria to be more general.
2033 PH/38 The host_find_failed option in the manualroute router can now be set
2034 to "ignore", to completely ignore a host whose IP address cannot be
2035 found. If all hosts are ignored, the behaviour is controlled by the new
2036 host_all_ignored option.
2038 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2039 homing or because of multiple MX records with /mx) generated more than
2040 one IP address, and the following item turned out to be the local host,
2041 all the secondary addresses of the first item were incorrectly removed
2042 from the list, along with the local host and any following hosts (which
2043 is what is supposed to happen).
2045 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2046 whoever called Exim into the -H file. In the case of the daemon it was
2047 behaving confusingly. When first started, it used values for whoever
2048 started the daemon, but after a SIGHUP it used the Exim user (because it
2049 calls itself on a restart). I have changed the code so that it now always
2052 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2053 message are rejected with the same error (e.g. no authentication or bad
2054 sender address), and a DATA command is nevertheless sent (as can happen
2055 with PIPELINING or a stupid MUA), the error message that was given to the
2056 RCPT commands is included in the rejection of the DATA command. This is
2057 intended to be helpful for MUAs that show only the final error to their
2060 PH/42 Another patch from the Sieve maintainer.
2062 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2063 Eximstats - Fixed some broken HTML links and added missing column headers
2065 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2066 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2068 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2070 PH/43 Yet another patch from the Sieve maintainer.
2072 PH/44 I found a way to check for a TCP/IP connection going away before sending
2073 the response to the final '.' that terminates a message, but only in the
2074 case where the client has not sent further data following the '.'
2075 (unfortunately, this is allowed). However, in many cases there won't be
2076 any further data because there won't be any more messages to send. A call
2077 to select() can be used: if it shows that the input is "ready", there is
2078 either input waiting, or the socket has been closed. An attempt to read
2079 the next input character can distinguish the two cases. Previously, Exim
2080 would have sent an OK response which the client would never have see.
2081 This could lead to message repetition. This fix should cure that, at
2082 least in a lot of common cases.
2084 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2085 advertised in response to EHLO.
2091 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2092 fixed by 4.65/MH/01 (is this a record?) are fixed:
2094 (i) An empty string was always treated as zero by the numeric comparison
2095 operators. This behaviour has been restored.
2097 (ii) It is documented that the numeric comparison operators always treat
2098 their arguments as decimal numbers. This was broken in that numbers
2099 starting with 0 were being interpreted as octal.
2101 While fixing these problems I realized that there was another issue that
2102 hadn't been noticed. Values of message_size_limit (both the global option
2103 and the transport option) were treated as octal if they started with 0.
2104 The documentation was vague. These values are now always treated as
2105 decimal, and I will make that clear in the documentation.
2111 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2112 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2115 MH/01 Don't check that the operands of numeric comparison operators are
2116 integers when their expansion is in "skipping" mode (fixes bug
2117 introduced by 4.64-PH/07).
2119 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2120 child addresses, Exim now panics and dies. Previously, because the count
2121 is held in a short int, deliveries were likely to be lost. As such a
2122 large number of recipients for a single message is ridiculous
2123 (performance will be very, very poor), I have chosen to impose a limit
2124 rather than extend the field.
2130 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2131 leftover -K file (the existence of which was triggered by #402).
2132 While we were at it, introduced process PID as part of the -K
2133 filename. This should rule out race conditions when creating
2136 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2137 processing considerably. Previous code took too long for large mails,
2138 triggering a timeout which in turn triggers #401.
2140 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2141 in the DK code in transports.c. sendfile() is not really portable,
2142 hence the _LINUX specificness.
2144 TF/01 In the add_headers option to the mail command in an Exim filter,
2145 there was a bug that Exim would claim a syntax error in any
2146 header after the first one which had an odd number of characters
2149 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2150 callout verification, Exim cached a "reject" for the entire domain. This
2151 is correct for most verifications, but it is not correct for a recipient
2152 verification with use_sender or use_postmaster set, because in that case
2153 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2154 case of MAIL FROM:<> rejection from other early rejections (e.g.
2155 rejection of HELO). When verifying a recipient using a non-null MAIL
2156 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2157 Whatever the result of the callout, the value of the domain cache is
2158 left unchanged (for any other kind of callout, getting as far as trying
2159 RCPT means that the domain itself is ok).
2161 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2164 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2165 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2166 session, the daemon ignores ECONNECT errors and logs others; it now
2167 ignores EPIPE as well.
2169 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2170 (quoted-printable decoding).
2172 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2173 later the small subsequent patch to fix an introduced bug.
2175 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2177 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2179 PH/08 An error is now given if message_size_limit is specified negative.
2181 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2182 to be given (somewhat) arbitrary names.
2184 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2187 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2188 miscellaneous code fixes
2190 PH/10 Added the log_reject_target ACL modifier to specify where to log
2193 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2194 hostname. This is wrong, because it relates to the incoming message (and
2195 probably the interface on which it is arriving) and not to the outgoing
2196 callout (which could be using a different interface). This has been
2197 changed to use the value of the helo_data option from the smtp transport
2198 instead - this is what is used when a message is actually being sent. If
2199 there is no remote transport (possible with a router that sets up host
2200 addresses), $smtp_active_hostname is used.
2202 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2203 tweaks were necessary in order to get it to work (see also 21 below):
2204 (a) The code assumed that strncpy() returns a negative number on buffer
2205 overflow, which isn't the case. Replaced with Exim's string_format()
2207 (b) There were several signed/unsigned issues. I just did the minimum
2208 hacking in of casts. There is scope for a larger refactoring.
2209 (c) The code used strcasecmp() which is not a standard C function.
2210 Replaced with Exim's strcmpic() function.
2211 (d) The code set only $1; it now sets $auth1 as well.
2212 (e) A simple test gave the error "authentication client didn't specify
2213 service in request". It would seem that Dovecot has changed its
2214 interface. Fortunately there's a specification; I followed it and
2215 changed what the client sends and it appears to be working now.
2217 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2220 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2221 address A is aliased to B and C, where B exists and C does not. Without
2222 -v the output is "A verified" because verification stops after a
2223 successful redirection if more than one address is generated. However,
2224 with -v the child addresses are also verified. Exim was outputting "A
2225 failed to verify" and then showing the successful verification for C,
2226 with its parentage. It now outputs "B failed to verify", showing B's
2227 parentage before showing the successful verification of C.
2229 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2230 look up a TXT record in a specific list after matching in a combined
2233 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2234 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2235 they consult the DNS. I had assumed they would set it the way they
2236 wanted; and indeed my experiments on Linux seem to show that in some
2237 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2238 To be on the safe side, however, I have now made the interface to
2239 host_find_byname() similar to host_find_bydns(), with an argument
2240 containing the DNS resolver options. The host_find_byname() function now
2241 sets these options at its start, just as host_find_bydns() does. The smtp
2242 transport options dns_qualify_single and dns_search_parents are passed to
2243 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2244 of host_find_byname() use the default settings of RES_DEFNAMES
2245 (qualify_single) but not RES_DNSRCH (search_parents).
2247 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2248 spool_read_header() do less string testing, by means of a preliminary
2249 switch on the second character of optional "-foo" lines. (This is
2250 overdue, caused by the large number of possibilities that now exist.
2251 Originally there were few.) While I was there, I also converted the
2252 str(n)cmp tests so they don't re-test the leading "-" and the first
2253 character, in the hope this might squeeze out yet more improvement.
2255 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2256 flag allowing group syntax was set by the header_syntax check but not
2257 turned off, possible causing trouble later; (2) The flag was not being
2258 set at all for the header_verify test, causing "group"-style headers to
2259 be rejected. I have now set it in this case, and also caused header_
2260 verify to ignore an empty address taken from a group. While doing this, I
2261 came across some other cases where the code for allowing group syntax
2262 while scanning a header line wasn't quite right (mostly, not resetting
2263 the flag correctly in the right place). These bugs could have caused
2264 trouble for malformed header lines. I hope it is now all correct.
2266 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2267 with the "reply" argument non-NULL. The code, however (which originally
2268 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2269 but it didn't always do it. This confused somebody who was copying the
2270 code for some other use. I have removed all the tests.
2272 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2273 feature that was used to support insecure browsers during the U.S. crypto
2274 embargo. It requires special client support, and Exim is probably the
2275 only MTA that supported it -- and would never use it because real RSA is
2276 always available. This code has been removed, because it had the bad
2277 effect of slowing Exim down by computing (never used) parameters for the
2278 RSA_EXPORT functionality.
2280 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2281 authenticator to fail if there's a tab character in the incoming data
2282 (there should never be unless someone is messing about, as it's supposed
2283 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2284 if the connection is using TLS or if the remote IP is the same as the
2285 local IP, and the "valid-client-cert option" if a client certificate has
2288 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2289 authenticators. This can be used for authorization after authentication
2290 succeeds. (In the case of plaintext, it servers for both authentication
2293 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2294 if any retry times were supplied.
2296 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2297 connection, where there is no client IP address to check. In this
2298 situation, the verify now always succeeds.
2300 PH/25 Applied John Jetmore's -Mset patch.
2302 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2304 PH/27 In a string expansion for a processed (not raw) header when multiple
2305 headers of the same name were present, leading whitespace was being
2306 removed from all of them, but trailing whitespace was being removed only
2307 from the last one. Now trailing whitespace is removed from each header
2308 before concatenation. Completely empty headers in a concatenation (as
2309 before) are ignored.
2311 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2312 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2314 PH/29 [Removed. This was a change that I later backed out, and forgot to
2315 correct the ChangeLog entry (that I had efficiently created) before
2316 committing the later change.]
2318 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2319 address errors (4xx response to RCPT) over the same connection as other
2320 messages routed to the same hosts. Such deliveries are always "forced",
2321 so retry times are not inspected. This resulted in far too many retries
2322 for the affected addresses. The effect occurred only when there were more
2323 hosts than the hosts_max_try setting in the smtp transport when it had
2324 the 4xx errors. Those hosts that it had tried were not added to the list
2325 of hosts for which the message was waiting, so if all were tried, there
2326 was no problem. Two fixes have been applied:
2328 (i) If there are any address or message errors in an SMTP delivery, none
2329 of the hosts (tried or untried) are now added to the list of hosts
2330 for which the message is waiting, so the message should not be a
2331 candidate for sending over the same connection that was used for a
2332 successful delivery of some other message. This seems entirely
2333 reasonable: after all the message is NOT "waiting for some host".
2334 This is so "obvious" that I'm not sure why it wasn't done
2335 previously. Hope I haven't missed anything, but it can't do any
2336 harm, as the worst effect is to miss an optimization.
2338 (ii) If, despite (i), such a delivery is accidentally attempted, the
2339 routing retry time is respected, so at least it doesn't keep
2340 hammering the server.
2342 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2343 in ${readsocket because some servers need this prod.
2345 PH/32 Added some extra debug output when updating a wait-xxx database.
2347 PH/33 The hint "could be header name not terminated by colon", which has been
2348 given for certain expansion errors for a long time, was not being given
2349 for the ${if def:h_colon_omitted{... case.
2351 PH/34 The spec says: "With one important exception, whenever a domain list is
2352 being scanned, $domain contains the subject domain." There was at least
2353 one case where this was not true.
2355 PH/35 The error "getsockname() failed: connection reset by peer" was being
2356 written to the panic log as well as the main log, but it isn't really
2357 panic-worthy as it just means the connection died rather early on. I have
2358 removed the panic log writing for the ECONNRESET error when getsockname()
2361 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2362 runs only) independently of the message's sender address. This meant
2363 that, if the 4xx error was in fact related to the sender, a different
2364 message to the same recipient with a different sender could confuse
2365 things. In particualar, this can happen when sending to a greylisting
2366 server, but other circumstances could also provoke similar problems.
2367 I have changed the default so that the retry time for these errors is now
2368 based a combination of the sender and recipient addresses. This change
2369 can be overridden by setting address_retry_include_sender=false in the
2372 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2373 remote server are returned as part of bounce messages. This was not
2374 happening for LMTP over a pipe (the lmtp transport), but now it is the
2375 same for both kinds of LMTP.
2377 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2378 in header lines that were in fact CNAMEs. This is no longer the case.
2380 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2381 and queue runs started by the daemon processed all messages. This has
2382 been fixed so that -R and -S can now usefully be given with -q<time>.
2384 PH/40 Import PCRE release 6.7 (fixes some bugs).
2386 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2388 PH/42 Give an error if -q is specified more than once.
2390 PH/43 Renamed the variables $interface_address and $interface_port as
2391 $received_ip_address and $received_port, to make it clear that these
2392 values apply to message reception, and not to the outgoing interface when
2393 a message is delivered. (The old names remain recognized, of course.)
2395 PH/44 There was no timeout on the connect() call when using a Unix domain
2396 socket in the ${readsocket expansion. There now is.
2398 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2399 be meaningful with "accept".
2401 SC/01 Eximstats V1.43
2402 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2404 SC/02 Eximstats V1.44
2405 Use a glob alias rather than an array ref in the generated
2406 parser. This improves both readability and performance.
2408 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2409 Collect SpamAssassin and rejection statistics.
2410 Don't display local sender or destination tables unless
2411 there is data to show.
2412 Added average volumes into the top table text output.
2414 SC/04 Eximstats V1.46
2415 Collect data on the number of addresses (recipients)
2416 as well as the number of messages.
2418 SC/05 Eximstats V1.47
2419 Added 'Message too big' to the list of mail rejection
2420 reasons (thanks to Marco Gaiarin).
2422 SC/06 Eximstats V1.48
2423 Mainlog lines which have GMT offsets and are too short to
2424 have a flag are now skipped.
2426 SC/07 Eximstats V1.49 (Alain Williams)
2427 Added the -emptyok flag.
2429 SC/08 Eximstats V1.50
2430 Fixes for obtaining the IP address from reject messages.
2432 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2433 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2434 whitesspace changes from 4.64-PH/27
2436 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2439 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2440 are found, allow negative numbers in numeric criteria)
2442 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2444 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2445 to match changes made in 4.64-PH/43
2447 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2449 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2450 "domain" is actually an IP address, and force a failure. This locks out
2451 those revolvers/nameservers that support "A-for-A" lookups, in
2452 contravention of the specifications.
2454 PH/48 When a host name was looked up from an IP address, and the subsequent
2455 forward lookup of the name timed out, the host name was left in
2456 $sender_host_name, contrary to the specification.
2458 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2459 restricted to single-key lookups, Exim was not diagnosing an error if
2460 * or *@ was used with a query-style lookup.
2462 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2464 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2465 long ago, but noone interested enough thought of it. Let's just say that
2466 the "1.1" means that there are some new functions that weren't there at
2467 some point in the past.
2469 PH/51 Error processing for expansion failure of helo_data from an smtp
2470 transport during callout processing was broken.
2472 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2473 tested/used via the -bh/-bhc/-bs options.
2475 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2476 bug, fixed in subsequent PCRE releases).
2478 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2479 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2481 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2487 SC/01 Use a glob alias rather than an array ref in eximstats generated
2488 parser. This improves both readability and performance.
2490 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2491 Don't display local sender or destination tables in eximstats unless
2492 there is data to show.
2493 Added average volumes into the eximstats top table text output.
2495 SC/03 Collect data on the number of addresses (recipients) as well
2496 as the number of messages in eximstats.
2498 TF/01 Correct an error in the documentation for the redirect router. Exim
2499 does (usually) call initgroups() when daemonizing.
2501 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2502 with consistent privilege compared to when running as a daemon.
2504 TF/03 Note in the spec that $authenticated_id is not set for local
2505 submissions from trusted users.
2507 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2508 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2510 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2511 by adding some example configuration directives to the default
2512 configuration file. A little bit of work is required to uncomment the
2513 directives and define how usernames and passwords are checked, but
2514 there is now a framework to start from.
2516 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2517 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2518 without this. I don't know how relevant this is to other LDAP libraries.
2520 PH/02 Add the verb name to the "unknown ACL verb" error.
2522 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2524 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2526 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2527 directory not expanded when it should be if an expanded home directory
2528 was set for the address (which is overridden by the transport).
2530 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2533 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2534 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2535 because it is too late at that time, and has no effect.
2537 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2538 security issue with \' (bugzilla #107). I could not use the
2539 PQescapeStringConn() function, because it needs a PGconn value as one of
2542 PH/08 When testing addresses using -bt, indicate those final addresses that
2543 are duplicates that would not cause an additional delivery. At least one
2544 person was confused, thinking that -bt output corresponded to deliveries.
2545 (Suppressing duplicates isn't a good idea as you lose the information
2546 about possibly different redirections that led to the duplicates.)
2548 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2549 systems where poll() doesn't work, in particular OS X.
2551 PH/10 Added more information to debugging output for retry time not reached.
2553 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2554 operations in malware.c.
2556 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2559 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2560 syslog_duplication set false, log lines that would normally be written
2561 both the the main log and to the reject log were not written to syslog at
2564 PH/14 In the default configuration, change the use of "message" in ACL warn
2565 statements to "add_header".
2567 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2568 not followed by a command (e.g. "seen endif").
2570 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2571 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2574 PH/17 Added extra conditions to the default value of delay_warning_condition
2578 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2579 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2580 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2583 The Auto-Submitted: and various List- headers are standardised, whereas I
2584 don't think Precedence: ever was.
2586 PH/18 Refactored debugging code in route_finduser() to show more information,
2587 in particular, the error code if getpwnam() issues one.
2589 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2590 This is apparently needed in addition to the PH/07 change above to avoid
2591 any possible encoding problems.
2593 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2594 but not after initializing Perl.
2596 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2597 output them only if debugging. By default they are written stderr,
2598 apparently, which is not desirable.
2600 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2603 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2606 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2608 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2609 authenticated or an ident call has been made. Suppress the default
2610 values for $authenticated_id and $authenticated_sender (but permit -oMai
2611 and -oMas) when testing with -bh.
2613 PH/24 Re-jigged the order of the tests in the default configuration so that the
2614 tests for valid domains and recipients precede the DNS black list and CSA
2615 tests, on the grounds that those ones are more expensive.
2617 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2618 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2619 This bug exists in every version of Exim that I still have, right back to
2622 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2623 However, an attempt to turn on case-sensitivity in a regex key by
2624 including (?-i) didn't work because the subject string was already
2625 lowercased, and the effects were non-intuitive. It turns out that a
2626 one-line patch can be used to allow (?-i) to work as expected.
2632 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2633 other effects) broke the use of negated acl sub-conditions.
2635 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2638 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2639 "Deny" causes Exim to reject the incoming connection with a 554 error.
2640 Unfortunately, if there is a major crisis, such as a disk failure,
2641 tcp-wrappers gives "deny", whereas what one would like would be some
2642 kind of temporary error. A kludge has been added to help with this.
2643 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2644 554 error is used if errno is still zero or contains ENOENT (which occurs
2645 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2648 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2650 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2651 errors. Otherwise a message that provokes a temporary error (when other
2652 messages do not) can cause a whole host to time out.
2654 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2655 addresses were routed directly to files or pipes from a redirect router.
2656 File deliveries just didn't batch; pipe deliveries might have suffered
2659 PH/06 A failure to get a lock for a hints database would erroneously always say
2660 "Failed to get write lock", even when it was really a read lock.
2662 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2663 of 0600. This has been changed to use the value of the lockfile_mode
2664 option (which defaults to 0600).
2666 PH/08 Applied small patch from the Sieve maintainer.
2668 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2669 folder from quota calculations, a direct delivery into this folder messed
2670 up the contents of the maildirsize file. This was because the regex was
2671 used only to exclude .Trash (or whatever) when the size of the mailbox
2672 was calculated. There was no check that a delivery was happening into an
2673 excluded directory. This bug has been fixed by ignoring all quota
2674 processing for deliveries into excluded directories.
2676 PH/10 Added the maildirfolder_create_regex option to appendfile.
2682 PH/01 The code for finding all the local interface addresses on a FreeBSD
2683 system running IPv6 was broken. This may well have applied to all BSD
2684 systems, as well as to others that have similar system calls. The broken
2685 code found IPv4 interfaces correctly, but gave incorrect values for the
2686 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2687 that it would not match correctly against @[] and not recognize the IPv6
2690 PH/02 The ipliteral router was not recognizing addresses of the form user@
2691 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2693 PH/03 Added disable_ipv6.
2695 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2696 decoded form, because it is most often used to construct To: headers
2697 lines in autoreplies, and the decoded form may well be syntactically
2698 invalid. However, $reply_address has leading white space removed, and all
2699 newlines turned into spaces so that the autoreply transport does not
2702 PH/05 If group was specified without a user on a router, and no group or user
2703 was specified on a transport, the group from the router was ignored.
2705 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2706 for visible compile-time settings that can be used to change these
2707 numbers, for those that want even more. Backwards compatibility with old
2708 spool files has been maintained. However, going back to a previous Exim
2709 release will lost any variables that are in spool files.
2711 PH/07 Two small changes when running in the test harness: increase delay when
2712 passing a TCP/IP connection to a new process, in case the original
2713 process has to generate a bounce, and remove special handling of
2714 127.0.0.2 (sic), which is no longer necessary.
2716 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2717 be the same on different OS.
2719 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2722 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2723 whether --show-vars was specified or not
2725 JJ/02 exipick: Added support for new ACL variable spool format introduced
2728 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2729 syntactically invalid From: or Reply-to: line, and a filter used this to
2730 generate an autoreply, and therefore failed to obtain an address for the
2731 autoreply, Exim could try to deliver to a non-existent relative file
2732 name, causing unrelated and misleading errors. What now happens is that
2733 it logs this as a hard delivery error, but does not attempt to create a
2736 PH/11 The exinext utility has a -C option for testing purposes, but although
2737 the given file was scanned by exinext itself; it wasn't being passed on
2738 when Exim was called.
2740 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2741 an end-of-file indication when reading a command response.
2743 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2744 compiled. In many other places in Exim, IPv6 addresses are always
2745 recognized, so I have changed this. It also means that IPv4 domain
2746 literals of the form [IPV4:n.n.n.n] are now always recognized.
2748 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2749 used if the router is not running as root, for example, when verifying at
2750 ACL time, or when using -bh. The debugging output from this situation was
2751 non-existent - all you got was a failure to exec. I have made two
2754 (a) Failures to set uid/gid, the current directory, or a process leader
2755 in a subprocess such as that created by queryprogram now generate
2756 suitable debugging ouput when -d is set.
2758 (b) The queryprogram router detects when it is not running as root,
2759 outputs suitable debugging information if -d is set, and then runs
2760 the subprocess without attempting to change uid/gid.
2762 PH/15 Minor change to Makefile for building test_host (undocumented testing
2765 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2766 additional section of a DNS packet that returns MX or SRV records.
2767 Instead, it always explicitly searches for A/AAAA records. This avoids
2768 major problems that occur when a DNS server includes only records of one
2769 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2770 fixed another bug: if SRV records were looked up and the corresponding
2771 address records were *not* found in the additional section, the port
2772 values from the SRV records were lost.
2774 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2775 using the correct key (the original address) when searching the retry
2776 rules in order to find which one to use for generating the retry hint.
2778 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2779 adding the default one. Similarly, if it contains a Reply-To: header, the
2780 errors_reply_to option, if set, is not used.
2782 PH/19 When calculating a retry time, Exim used to measure the "time since
2783 failure" by looking at the "first failed" field in the retry record. Now
2784 it does not use this if it is later than than the arrival time of the
2785 message. Instead it uses the arrival time. This makes for better
2786 behaviour in cases where some deliveries succeed, thus re-setting the
2787 "first failed" field. An example is a quota failure for a huge message
2788 when small messages continue to be delivered. Without this change, the
2789 "time since failure" will always be short, possible causing more frequent
2790 delivery attempts for the huge message than are intended.
2791 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2793 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2794 $1, $2, $3) because the numerical variables can be reset during some
2795 expansion items (e.g. "match"), thereby losing the authentication data.
2797 PH/21 Make -bV show the size of off_t variables so that the test suite can
2798 decide whether to run tests for quotas > 2G.
2800 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2801 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2802 filecount value is greater than 2G or if a quota value is greater than 2G
2803 on a system where the size of off_t is not greater than 4, a panic error
2806 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2807 never match. The debug and -bh output now contains an explicit error
2808 message indicating a malformed IPv4 address or mask.
2810 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2811 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2812 PH/23 above applies.
2814 PH/25 Do not write to syslog when running in the test harness. The only
2815 occasion when this arises is a failure to open the main or panic logs
2816 (for which there is an explicit test).
2818 PH/26 Added the /no_tell option to "control=freeze".
2820 PH/27 If a host name lookup failed very early in a connection, for example, if
2821 the IP address matched host_lookup and the reverse lookup yielded a name
2822 that did not have a forward lookup, an error message of the form "no IP
2823 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2824 could be logged. Now it outputs the IP address instead of "NULL".
2826 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2827 allows the sender and the authenticated sender to be set when
2828 submitting a message from within Exim. Since child_open_exim() is
2829 documented for local_scan(), the new function should be too.
2831 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2832 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2833 results in an empty string is now treated as unset.
2835 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2837 PH/31 Added sender_verify_fail logging option.
2839 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2840 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2841 tidied the source and removed it altogether.
2843 PH/33 When a queue run was abandoned because the load average was too high, a
2844 log line was always written; now it is written only if the queue_run log
2845 selector is set. In addition, the log line for abandonment now contains
2846 information about the queue run such as the pid. This is always present
2847 in "start" and "stop" lines but was omitted from the "abandon" line.
2849 PH/34 Omit spaces between a header name and the colon in the error message that
2850 is given when verify = headers_syntax fails (if there are lots of them,
2851 the message gets confusing).
2853 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2854 names, as there are now some PTR records that contain slashes. This check
2855 is only to protect against broken name servers that fall over on strange
2856 characters, so the fact that it applies to all lookups doesn't matter.
2858 PH/36 Now that the new test suite is complete, we can remove some of the
2859 special code in Exim that was needed for the old test suite. For example,
2860 sorting DNS records because real resolvers return them in an arbitrary
2861 order. The new test suite's fake resolver always returns records in the
2864 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2865 bounces) except when queue_only is set, to avoid logging races between
2866 the different processes.
2868 PH/38 Panic-die if .include specifies a non-absolute path.
2870 PH/39 A tweak to the "H" retry rule from its user.
2872 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2873 a label. They prevented compilation on older perls.
2875 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2876 a warning to be raised on newish perls.
2878 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2879 on queue. Changes to match documented behaviour of showing count of
2880 messages matching specified criteria.
2882 PH/40 Changed the default ident timeout from 30s to 5s.
2884 PH/41 Added support for the use of login_cap features, on those BSD systems
2885 that have them, for controlling the resources used by pipe deliveries.
2887 PH/42 The content-scanning code uses fopen() to create files in which to put
2888 message data. Previously it was not paying any attention to the mode of
2889 the files. Exim runs with umask(0) because the rest of the code creates
2890 files with open(), and sets the required mode explicitly. Thus, these
2891 files were ending up world-writeable. This was not a big issue, because,
2892 being within the spool directory, they were not world-accessible. I have
2893 created a function called modefopen, which takes an additional mode
2894 argument. It sets umask(777), creates the file, chmods it to the required
2895 mode, then resets the umask. All the relevant calls to fopen() in the
2896 content scanning code have been changed to use this function.
2898 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2899 to 24 hours. This avoids potential overflow problems when processing G
2900 and H retry rules. I suspect nobody ever tinkers with this value.
2902 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2904 PH/45 When the plaintext authenticator is running as a client, the server's
2905 challenges are checked to ensure they are valid base64 strings. By
2906 default, the authentication attempt is cancelled if an invalid string is
2907 received. Setting client_ignore_invalid_base64 true ignores these errors.
2908 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2909 they are received. Thus, the responses can be made to depend on the
2910 challenges. If an invalid string is ignored, an empty string is placed in
2913 PH/46 Messages that are created by the autoreply transport now contains a
2914 References: header, in accordance with RFCs 2822 and 3834.
2916 PH/47 Added authenticated_sender_force to the smtp transport.
2918 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2920 PH/49 Installed latest patch from the Sieve maintainer.
2922 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2923 was also set, the appendfile transport was unnecessarily scanning a
2924 directory of message files (e.g. for maildir delivery) to find the count
2925 of files (along with the size), even though it did not need this
2926 information. It now does the scan only if it needs to find either the
2927 size of the count of files.
2929 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2931 PH/52 Two bugs concerned with error handling when the smtp transport is
2934 (i) Exim was not creating retry information for temporary errors given
2935 for individual recipients after the DATA command when the smtp transport
2936 was used in LMTP mode. This meant that they could be retried too
2937 frequently, and not timed out correctly.
2939 (ii) Exim was setting the flag that allows error details to be returned
2940 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2941 recipients that were returned after the DATA command.
2943 PH/53 This is related to PH/52, but is more general: for any failing address,
2944 when detailed error information was permitted to be returned to the
2945 sender, but the error was temporary, then after the final timeout, only
2946 "retry timeout exceeded" was returned. Now it returns the full error as
2947 well as "retry timeout exceeded".
2949 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2950 do this, and (what is worse) MTAs that accept it.
2952 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2953 will now be deprecated.
2955 PH/56 New os.c-cygwin from the Cygwin maintainer.
2957 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2958 formats (previously only available in exim formats via -bpr, -bpru,
2959 and -bpra. Now also available in native and exiqgrep formats)
2961 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2962 with very large, slow to parse queues
2964 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2966 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2968 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2969 responses to authentication challenges, though it was showing the
2970 challenges; (ii) I've removed the CR characters from the debug output for
2973 PH/58 Allow for the insertion of a newline as well as a space when a string
2974 is turned into more than one encoded-word during RFC 2047 encoding. The
2975 Sieve code now uses this.
2977 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2978 data_4xx, lost_connection, tls_required.
2980 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2981 message was being sent as an SMTP response.
2983 PH/61 Add -l and -k options to exicyclog.
2985 PH/62 When verifying, if an address was redirected to one new address, so that
2986 verification continued, and the new address failed or deferred after
2987 having set something in $address_data, the value of $address_data was not
2988 passed back to the ACL. This was different to the case when no
2989 redirection occurred. The value is now passed back in both cases.
2991 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2992 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2993 use login_cap.h, so on its own it isn't the distinguishing feature. The
2994 new name refers directly to the setclassresources() function.
2996 PH/65 Added configuration files for NetBSD3.
2998 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3000 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3001 is preferred over IPv4.
3003 PH/68 The bounce_return_message and bounce_return_body options were not being
3004 honoured for bounces generated during the reception of non-SMTP messages.
3005 In particular, this applied to messages rejected by the ACL. This bug has
3006 been fixed. However, if bounce_return_message is true and bounce_return_
3007 body is false, the headers that are returned for a non-SMTP message
3008 include only those that have been read before the error was detected.
3009 (In the case of an ACL rejection, they have all been read.)
3011 PH/69 The HTML version of the specification is now built in a directory called
3012 spec_html instead of spec.html, because the latter looks like a path with
3013 a MIME-type, and this confuses some software.
3015 PH/70 Catch two compiler warnings in sieve.c.
3017 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3018 function verify_get_ident() calls ip_connect() to connect a socket, but
3019 if the "connect()" function timed out, ip_connect() used to close the
3020 socket. However, verify_get_ident() also closes the socket later, and in
3021 between Exim writes to the log, which may get opened at this point. When
3022 the socket was closed in ip_connect(), the log could get the same file
3023 descriptor number as the socket. This naturally causes chaos. The fix is
3024 not to close the socket in ip_connect(); the socket should be closed by
3025 the function that creates it. There was only one place in the code where
3026 this was missing, in the iplookup router, which I don't think anybody now
3027 uses, but I've fixed it anyway.
3029 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3030 well as to direct DNS lookups. Otherwise the handling of names in host
3031 lists is inconsistent and therefore confusing.
3037 PH/01 Two changes to the default runtime configuration:
3039 (1) Move the checks for relay_from_hosts and authenticated clients from
3040 after to before the (commented out) DNS black list checks.
3042 (2) Add control=submission to the relay_from_hosts and authenticated
3043 clients checks, on the grounds that messages accepted by these
3044 statements are most likely to be submissions.
3046 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3048 (1) Generate an error if the third argument for the ${prvs expansion is
3051 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3054 (3) Reset the variables that are obtained from the first argument of
3055 ${prvscheck and used in the second argument before leaving the code,
3056 because their memory is reclaimed, so using them afterwards may do
3059 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3060 one (it's much easier than Tom thought :-).
3062 (5) Because of (4), we can now allow for the use of $prvscheck_result
3063 inside the third argument.
3065 PH/03 For some reason, the default setting of PATH when running a command from
3066 a pipe transport was just "/usr/bin". I have changed it to
3069 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3070 anything to be listed in the output from -bV.
3072 PH/05 When a filter generated an autoreply, the entire To: header line was
3073 quoted in the delivery log line, like this:
3075 => >A.N.Other <ano@some.domain> <original@ddress> ...
3077 This has been changed so that it extracts the operative address. There
3078 may be more than one such address. If so, they are comma-separated, like
3081 => >ano@some.domain,ona@other.domain <original@ddress> ...
3083 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3084 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3085 not being looked up in the rDNS to get a host name, Exim was showing the
3086 IP address twice in Received: lines, even though the IP addresses were
3087 identical. For example:
3089 Received: from [1.2.3.4] (helo=[1.2.3.4])
3091 However, if the real host name was known, it was omitting the HELO data
3092 if it matched the actual IP address. This has been tidied up so that it
3093 doesn't show the same IP address twice.
3095 PH/07 When both +timestamp and +memory debugging was on, the value given by
3096 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3097 called by the debug printing, thereby overwriting the timestamp buffer.
3098 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3100 PH/08 When the original message was included in an autoreply transport, it
3101 always said "this is a copy of the message, including all the headers",
3102 even if body_only or headers_only was set. It now gives an appropriate
3105 PH/09 Applied a patch from the Sieve maintainer which:
3107 o fixes some comments
3108 o adds the (disabled) notify extension core
3109 o adds some debug output for the result of if/elsif tests
3110 o points to the current vacation draft in the documentation
3111 and documents the missing references header update
3115 o fixes a bug in processing the envelope test (when testing
3116 multiple envelope elements, the last element determinted the
3119 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3120 Electronic Mail") by including:
3122 Auto-submitted: auto-generated
3124 in the messages that it generates (bounce messages and others, such as
3125 warnings). In the case of bounce messages for non-SMTP mesages, there was
3126 also a typo: it was using "Auto_submitted" (underscore instead of
3127 hyphen). Since every message generated by Exim is necessarily in response
3128 to another message, thes have all been changed to:
3130 Auto-Submitted: auto-replied
3132 in accordance with these statements in the RFC:
3134 The auto-replied keyword:
3136 - SHOULD be used on messages sent in direct response to another
3137 message by an automatic process,
3139 - MUST NOT be used on manually-generated messages,
3141 - MAY be used on Delivery Status Notifications (DSNs) and Message
3142 Disposition Notifications (MDNs),
3144 - MUST NOT be used on messages generated by automatic or periodic
3145 processes, except for messages which are automatic responses to
3148 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3149 to the default Received: header definition.
3151 PH/12 Added log selector acl_warn_skipped (default on).
3153 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3154 variables because (a) they are in the wrong storage pool and (b) even if
3155 they were copied, it wouldn't work properly because of the caching.
3157 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3158 checking when decoding. Apparently there are clients that generate
3159 overlong encoded strings. Why am I not surprised?
3161 PH/15 If the first argument of "${if match_address" was not empty, but did not
3162 contain an "@" character, Exim crashed. Now it writes a panic log message
3163 and treats the condition as false.
3165 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3167 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3168 extension "envelope-auth". The code is finished and in agreement with
3169 other implementations, but there is no documentation so far and in fact,
3170 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3171 not changing the active code.
3173 Print executed "if" and "elsif" statements when debugging is used. This
3174 helps a great deal to understand what a filter does.
3176 Document more things not specified clearly in RFC3028. I had all this
3177 sorted out, when out of a sudden new issues came to my mind. Oops."
3179 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3182 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3183 canonical form (as documented). However, after a host name lookup from
3184 the IP address, check_host() was doing a simple string comparison with
3185 addresses acquired from the DNS when checking that the found name did
3186 have the original IP as one of its addresses. Since any found IPv6
3187 addresses are likely to be in abbreviated form, the comparison could
3188 fail. Luckily, there already exists a function for doing the comparison
3189 by converting both addresses to binary, so now that is used instead of
3190 the text comparison.
3192 PH/20 There was another similar case to PH/19, when a complete host name was
3193 given in a host list; looking up its IP address could give an abbreviated
3194 form, whereas the current host's name might or might not be abbreviated.
3195 The same fix has been applied.
3201 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3202 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3205 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3206 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3208 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3210 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3211 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3212 Also, for "Auto-Submitted", treat anything other than "no" as
3213 non-personal, in accordance with RFC 3834. (Previously it treated
3214 anything starting "auto-" as non-personal.)
3216 TF/01 The control=submission/name=... option had a problem with syntax
3217 errors if the name included a slash character. The /name= option
3218 now slurps the rest of the string, so it can include any characters
3219 but it must come last in the list of options (after /sender_retain
3222 PH/05 Some modifications to the interface to the fake nameserver for the new
3230 TK/01 Added the "success_on_redirect" address verification option. See
3231 NewStuff for rationale and an example.
3233 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3235 PH/02 Patch to exigrep to allow it to work on syslog lines.
3237 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3238 fread() to skip over the body file's header line, because in Cygwin the
3239 header line is locked and is inaccessible.
3241 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3242 co-exist for some time) to make it clear that it is the Exim ID that is
3243 referenced, not the Message-ID: header line.
3245 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3246 string_format() function, because snprintf() does not exist on all
3249 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3250 ${stat: expansion item.
3252 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3253 protocol synchronization error", to keep the pedants happy.
3255 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3256 well as for IRIX systems, when gcc is being used. See the host.c source
3259 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3261 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3264 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3265 transport and to the smtp transport in LMTP mode.
3267 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3269 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3270 run a filter in a subprocess. This could lead to confusion in subsequent
3271 lookups in the parent process. There should also be a search_tidyup() at
3272 the end of the subprocess.
3274 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3275 only if the host matched helo_try_verify_hosts, which caused the
3276 verification to occur when the EHLO/HELO command was issued. The ACL just
3277 tested the remembered result. Now, if a previous verification attempt has
3278 not happened, "verify = helo" does it there and then.
3280 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3282 TK/03 Fix log output including CR from clamd.
3284 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3285 exist provoked a memory error which could cause a segfault.
3287 PH/15 Installed PCRE 6.2
3289 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3291 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3292 of the problem. Specifically, suggested +O2 rather than +O1 for the
3295 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3297 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3298 record, the port was not correctly logged when the outgoing_port log
3299 selector was set (it logged the transort's default port).
3301 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3302 fallback_hosts, and "hosts" in the smtp transport.
3304 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3305 host errors such as "Connection refused".
3307 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3308 authentication with radiusclient 0.4.9:
3310 - Error returned from rc_read_config was caught wrongly
3311 - Username/password not passed on to radius server due to wrong length.
3313 The presumption is that some radiusclient API changes for 4.51/PH/17
3314 were not taken care of correctly. The code is still untested by me (my
3315 Linux distribution still has 0.3.2 of radiusclient), but it was
3316 contributed by a Radius user.
3318 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3319 expanding the "port" option of the smtp transport.
3321 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3322 while reading a MIME header. Thanks to Tom Hughes for a patch.
3324 PH/24 Include config.h inside local_scan.h so that configuration settings are
3327 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3328 that in an ACL for RCPT (for example), you can examine exactly what was
3331 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3332 commands, but it was not correctly comparing the address with the actual
3333 client host address. Thus, it would show the EHLO address in Received:
3334 header lines when this was not necessary.
3336 PH/27 Added the % operator to ${eval:}.
3338 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3339 it should be ignoring failures (because with -C, for example, it has lost
3340 privilege). It wasn't ignoring creation failures other than "already
3343 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3346 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3347 because an input line was too long, either on its own, or by virtue of
3348 too many continuations, the temporary file was not being removed, and the
3349 return code was incorrect.
3351 PH/31 Missing "BOOL" in function definition in filtertest.c.
3353 PH/32 Applied Sieve patches from the maintainer.
3355 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3357 PH/33 Added "verify = not_blind".
3359 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3360 Local/Makefile (with some defaults set). These are used in built scripts
3361 such as exicyclog, but they have never been used in the exim_install
3362 script (though there are many overriding facilities there). I have
3363 arranged that the exim_install script now takes note of these two
3366 PH/35 Installed configuration files for Dragonfly.
3368 PH/36 When a locally submitted message by a trusted user did not contain a
3369 From: header, and the sender address was obtained from -f or from an SMTP
3370 MAIL command, and the trusted user did not use -F to supply a sender
3371 name, $originator_name was incorrectly used when constructing a From:
3372 header. Furthermore, $originator_name was used for submission mode
3373 messages from external hosts without From: headers in a similar way,
3374 which is clearly wrong.
3376 PH/37 Added control=suppress_local_fixups.
3378 PH/38 When log_selector = +received_sender was set, and the addition of the
3379 sender made the log line's construction buffer exactly full, or one byte
3380 less than full, an overflow happened when the terminating "\n" was
3383 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3384 when the result of a list match is failure because a DNS lookup failed.
3386 PH/40 RM_COMMAND is now used in the building process.
3388 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3389 the "build-* directories that it finds.
3391 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3392 address in a domain literal was a prefix of an interface address.
3394 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3395 when verifying a sender address, unless rewrite_headers is false.
3397 PH/44 Wrote a long comment about why errors_to addresses are verified as
3398 recipients, not senders.
3400 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3401 the ratelimit ACL was added.
3403 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3405 PH/46 Added extra information about PostgreSQL errors to the error string.
3407 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3408 avoiding the need to install special zones in a real server. This is
3409 backwards compatible; if it can't find the fake resolver, it drops back.
3410 Thus, both old and new test suites can be run.
3412 TF/02 Added util/ratelimit.pl
3414 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3415 clock is set back in time.
3417 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3418 Candler <B.Candler@pobox.com>.
3420 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3421 OK for addresses that are the result of redirections.
3423 PH/48 A number of further additions for the benefit of the new test suite,
3424 including a fake gethostbyname() that interfaces to the fake DNS resolver
3427 TF/06 The fix for widen_domains has also been applied to qualify_single and
3428 search_parents which are the other dnslookup options that can cause
3431 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3434 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3436 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3437 service" if it encountered an MX record with an empty target hostname.
3438 The message is now "an MX or SRV record indicated no SMTP service".
3440 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3441 if the DNS of the sending site is misconfigured. This is quite a
3442 common situation. This change restores the behaviour of treating a
3443 helo verification defer as a failure.
3445 PH/51 If self=fail was set on a router, the bounce message did not include the
3446 actual error message.
3452 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3454 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3455 command itself ended in error, the underlying message about the transport
3456 filter timeout was being overwritten with the pipe command error. Now the
3457 underlying error message should be appended to the second error message.
3459 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3461 PH/02 Reduce the amount of output that "make" produces by default. Full output
3462 can still be requested.
3464 PH/03 The warning log line about a condition test deferring for a "warn" verb
3465 was being output only once per connection, rather than after each
3466 occurrence (because it was using the same function as for successful
3467 "warn" verbs). This seems wrong, so I have changed it.
3469 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3470 it should not have, which might have caused a crash in the right
3471 circumstances, but probably never did.
3473 PH/04 Installed a modified version of Tony Finch's patch to make submission
3474 mode fix the return path as well as the Sender: header line, and to
3475 add a /name= option so that you can make the user's friendly name appear
3478 TF/03 Added the control = fakedefer ACL modifier.
3480 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3481 Mark Lowes for thorough testing.
3483 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3485 TK/03 Merged latest SRS patch from Miles Wilton.
3487 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3488 with the definition in sysexits.h (which is #included earlier).
3489 Fortunately, Exim does not actually use EX_OK. The code used to try to
3490 preserve the sysexits.h value, by assumimg that macro definitions were
3491 scanned for macro replacements. I have been disabused of this notion,
3492 so now the code just undefines EX_OK before #including unistd.h.
3494 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3495 in the smtp transport. When a block could not be written in a single
3496 write() function, the timeout was being re-applied to each part-write.
3497 This seems wrong - if the receiver was accepting one byte at a time it
3498 would take for ever. The timeout is now adjusted when this happens. It
3499 doesn't have to be particularly precise.
3501 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3502 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3504 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3505 without a domain if the check to <postmaster@domain> fails.
3507 SC/01 Eximstats: added -xls and the ability to specify output files
3508 (patch written by Frank Heydlauf).
3510 SC/02 Eximstats: use FileHandles for outputing results.
3512 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3514 SC/04 Eximstats: fixed display of large numbers with -nvr option
3516 SC/05 Eximstats: fixed merging of reports with empty tables.
3518 SC/06 Eximstats: added the -include_original_destination flag
3520 SC/07 Eximstats: removed tabs and trailing whitespace.
3522 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3524 TK/06 MBOX spool code: Add real "From " MBOX separator line
3525 so the .eml file is really in mbox format (even though
3526 most programs do not really care). Patch from Alex Miller.
3528 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3529 The latter is generated from $received_to and is only set if the
3530 message has one envelope recipient. SA can use these headers,
3531 obviously out-of-the-box. Patch from Alex Miller.
3533 PH/08 The ${def test on a variable was returning false if the variable's
3534 value was "0", contrary to what the specification has always said!
3535 The result should be true unless the variable is empty.
3537 PH/09 The syntax error of a character other than { following "${if
3538 def:variable_name" (after optional whitespace) was not being diagnosed.
3539 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3540 accidental colon was present, for example, could give incorrect results.
3542 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3543 result is used (not including appendfile, where other changes are about
3546 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3547 This involved changing a lot of size variables from int to off_t. It
3548 should work with maildirs and everything.
3550 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3551 spamd dying while we are connected to it.
3553 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3556 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3557 function for BDB 4.3.
3559 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3561 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3562 the sources for PCRE have been re-arranged and more files are now
3565 PH/15 The code I had for printing potentially long long variables in PH/11
3566 above was not the best (it lost precision). The length of off_t variables
3567 is now inspected at build time, and an appropriate printing format (%ld
3568 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3569 to be "long long int" or "long int". This is needed for the internal
3570 formatting function string_vformat().
3572 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3573 the configuration file to be ":syslog", then the script "guesses" where
3574 the logs files are, rather than using the compiled in default. In our
3575 case the guess is not the same as the compiled default, so the script
3576 suddenly stopped working when I started to use syslog. The patch checks
3577 to see if log_file_path is "". If so, it attempts to read it from exim
3578 with no configuration file to get the compiled in version, before it
3579 falls back to the previous guessing code."
3581 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3582 implementing BATV in an Exim configuration. See NewStuff for the gory
3585 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3586 Makefile that are specific to HP-UX.
3588 PH/18 If the "use_postmaster" option was set for a recipient callout together
3589 with the "random" option, the postmaster address was used as the MAIL
3590 FROM address for the random test, but not for the subsequent recipient
3591 test. It is now used for both.
3593 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3594 patch removes a few documentation additions to RFC 3028, because the
3595 latest draft now contains them. It adds the new en;ascii-case comparator
3596 and a new error check for 8bit text in MIME parts. Comparator and
3597 require names are now matched exactly. I enabled the subaddress
3598 extension, but it is not well tested yet (read: it works for me)."
3600 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3601 rework some of the code of TK/09 above to avoid the hardwired use of
3602 "%lld" and "long long". Replaced the call to snprintf() with a call to
3605 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3606 records point to non-existent hosts", "retry timeout exceeded", and
3607 "retry time not reached for any host after a long failure period".
3609 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3610 experimental DomainKeys support:
3612 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3613 (2) On an error such as an illegally used "control", the wrong name for
3614 the control was given.
3616 These problems did NOT occur unless DomainKeys support was compiled.
3618 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3620 PH/24 Added ${if match_ip condition.
3622 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3623 databases so that it will be absolutely obvious if a crash occurs in the
3624 DB library. This is a regular occurrence (often caused by mis-matched
3627 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3628 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3629 picked up on a user's system that detects such things. There doesn't seem
3630 to be a gcc warning option for this - only an attribute that has to be
3631 put on the function's prototype. It seems that in Fedora Core 4 they have
3632 set this on a number of new functions. No doubt there will be more in due
3635 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3636 specify a transport. However, if an address that was verified by such a
3637 router was the subject of a callout, Exim crashed because it tried to
3638 read the rcpt_include_affixes from the non-existent transport. Now it
3639 just assumes that the setting of that option is false. This bug was
3640 introduced by 4.51/PH/31.
3642 PH/28 Changed -d+all to exclude +memory, because that information is very
3643 rarely of interest, but it makes the output a lot bigger. People tend to
3644 do -d+all out of habit.
3646 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3647 code in os-type was giving problems when libc.so lives in lib64, like on
3650 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3651 aren't the modern standard, and it seems that some systems' include files
3652 don't always have them. Exim was already checking for some of the newer
3653 ones like T_AAAA, and defining it itself. I've added checks for all the
3654 record types that Exim uses.
3656 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3657 not automatically generating a new one, as it is supposed to. This
3658 prevented TLS from working. If the file did exist, but contained invalid
3659 data, a new version was generated, as expected. It was only the case of a
3660 non-existent file that was broken.
3662 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3663 with a change in libdomainkeys > 0.64.
3665 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3666 from DNS. If the selector record carries the flag, it now has
3667 precedence over the domain-wide flag.
3669 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3671 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3672 the use of a port name that isn't defined in /etc/services) occurred, the
3673 message was deferred as in a normal delivery, and thus remained on the
3674 spool, instead of being failed because of the mua_wrapper setting. This
3675 is now fixed, and I tidied up some of the mua_wrapper messages at the
3678 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3679 the messages in a hash of arrays rather than using individual hashes.
3680 This is a bit cleaner and results in dramatic memory savings, albeit
3681 at a slight CPU cost.
3683 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3684 as requested by Marc Sherman.
3686 SC/10 Eximstats: added histograms for user specified patterns as requested
3689 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3691 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3692 fopen() in the content-scanning modules that did not already have it.
3698 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3699 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3701 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3703 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3705 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3706 to test Sieve filters that use "vacation".
3708 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3709 that changes the way the GnuTLS parameters are stored in the cache file.
3710 The new format can be generated externally. For backward compatibility,
3711 if the data in the cache doesn't make sense, Exim assumes it has read an
3712 old-format file, and it generates new data and writes a new file. This
3713 means that you can't go back to an older release without removing the
3716 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3717 work if there are any delivery delays because "one_time" forces the
3718 parent to be marked "delivered", so its unseen clone is never tried
3719 again. For this reason, Exim now forbids the simultaneous setting of
3722 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3723 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3724 entry if you want to know the details. The fix, however, neglected to
3725 consider the case when local delivery batching is involved. The test for
3726 "previously delivered" was not happening when checking to see if an
3727 address could be batched with a previous (undelivered) one; under
3728 certain circumstances this could lead to multiple deliveries to the same
3731 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3732 in its include files, and this causes problems building Exim.
3734 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3735 header_syntax) but Exim was just ignoring anything given after a slash.
3736 In particular, this caused confusion with an attempt to use "verify =
3737 reverse_host_lookup/defer_ok". An error is now given when options are
3738 supplied for verify items that do not have them. (Maybe reverse_host_
3739 lookup should have a defer_ok option, but that's a different point.)
3741 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3742 defined by RFC 821) to 2048, because there were problems with some AUTH
3743 commands, and RFC 1869 says the size should be increased for extended
3744 SMTP commands that take arguments.
3746 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3749 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3750 "unknown" error; now it says that the functionality isn't in the binary.
3752 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3753 an address' error message when a string expansion fails (syntax or
3754 whatever). Otherwise the password may appear in the log. Following change
3755 PH/42 below, there is no longer a chance of it appearing in a bounce
3758 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3760 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3761 of its addresses was ever tried. (Bugzilla bug #2.)
3763 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3764 the result incorrectly in the debug output. (It correctly added a newline
3765 to what was transported.)
3767 TF/01 Added $received_time.
3769 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3770 commented out examples of how to interface to a virus scanner and to
3771 SpamAssassin. Also added commented examples of av_scanner and
3772 spamd_address settings.
3774 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3775 and controls are allowed in which ACLs. There were a couple of minor
3776 errors. Some of the entries in the conditions table (which is a table of
3777 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3778 negation of where the condition IS allowed.
3780 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3782 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3783 header file does not have a version number, so I've had to invent a new
3784 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3785 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3786 radiusclient), but it was contributed by a Radius user.
3788 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3789 files or directories, for OpenSSL.
3791 PH/19 When an Exim process that is running as root has to create an Exim log
3792 file, it does so in a subprocess that runs as exim:exim so as to get the
3793 ownership right at creation (otherwise, other Exim processes might see
3794 the file with the wrong ownership). There was no test for failure of this
3795 fork() call, which would lead to the process getting stuck as it waited
3796 for a non-existent subprocess. Forks do occasionally fail when resources
3797 run out. I reviewed all the other calls to fork(); they all seem to check
3800 PH/20 When checking for unexpected SMTP input at connect time (before writing
3801 the banner), Exim was not dealing correctly with a non-positive return
3802 from the read() function. If the client had disconnected by this time,
3803 the result was a log entry for a synchronization error with an empty
3804 string after "input=" when read() returned zero. If read() returned -1
3805 (an event I could not check), uninitialized data bytes were printed.
3806 There were reports of junk text (parts of files, etc) appearing after
3809 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3811 PH/22 Added support for macro redefinition, and (re)definition in between
3812 driver and ACL definitions.
3814 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3815 forgetting to use the resulting value; it was using the unexpanded value.
3817 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3818 hadn't been configured. The fix is from Juergen Kreileder, who
3819 understands it better than I do:
3821 "Here's what I see happening with three configured cyrus_sasl
3822 authenticators configured (plain, login, cram-md5):
3824 On startup auth_cyrus_sasl_init() gets called for each of these.
3825 This means three calls to sasl_listmech() without a specified mech_list.
3826 => SASL tests which mechs of all available mechs actually work
3827 => three warnings about OTP not working
3828 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3830 With the patch, sasl_listmech() also gets called three times. But now
3831 SASL's mech_list option is set to the server_mech specified in the the
3832 authenticator. Or in other words, the answer from sasl_listmech()
3833 gets limited to just the mech you're testing for (which is different
3835 => the return list contains just 'plain' or 'login', 'cram-md5' or
3836 nothing depending on the value of ob->server_mech.
3838 I've just tested the patch: Authentication still works fine,
3839 unavailable mechs specified in the exim configuration are still
3840 caught, and the auth.log warnings about OTP are gone."
3842 PH/25 When debugging is enabled, the contents of the command line are added
3843 to the debugging output, even when log_selector=+arguments is not
3846 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3847 answer is "GNU", and only if the return is "GNU/something" is the answer
3850 PH/27 $acl_verify_message is now set immediately after the failure of a
3851 verification in an ACL, and so is available in subsequent modifiers. In
3852 particular, the message can be preserved by coding like this:
3854 warn !verify = sender
3855 set acl_m0 = $acl_verify_message
3857 Previously, $acl_verify_message was set only while expanding "message"
3858 and "log_message" when a very denied access.
3860 PH/28 Modified OS/os.c-Linux with
3862 -#ifndef OS_LOAD_AVERAGE
3863 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3865 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3866 nomenclature these days.)
3868 PH/29 Installed patch from the Sieve maintainer that adds the options
3869 sieve_useraddress and sieve_subaddress to the redirect router.
3871 PH/30 In these circumstances:
3872 . Two addresses routed to the same list of hosts;
3873 . First host does not offer TLS;
3874 . First host accepts first address;
3875 . First host gives temporary error to second address;
3876 . Second host offers TLS and a TLS session is established;
3877 . Second host accepts second address.
3878 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3879 and peerdn, if requested) that were in fact used only for the second
3882 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3883 attention to any local part prefix or suffix that was matched by the
3884 router that accepted the address. It now behaves in the same way as it
3885 does for delivery: the affixes are removed from the local part unless
3886 rcpt_include_affixes is set on the transport.
3888 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3889 timeout during the DATA phase of an incoming message.
3891 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3892 applied a patch sanctioned by the Sieve maintainer.
3894 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3895 the uid or gid is negative. A case of a negative gid caused this to be
3896 noticed. The fix allows for either to be negative.
3898 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3899 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3902 PH/36 The change PH/12 above was broken. Fixed it.
3904 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3905 the grounds that routing the same address twice would always produce the
3906 same answer. This might have been true once, but it is certainly no
3907 longer true now. Routing a child address may depend on the previous
3908 routing that produced that child. Some complicated redirection strategies
3909 went wrong when messages had multiple recipients, and made Exim's
3910 behaviour dependent on the order in which the addresses were given.
3912 I have moved the duplicate checking until after the routing is complete.
3913 Exim scans the addresses that are assigned to local and remote
3914 transports, and removes any duplicates. This means that more work will be
3915 done, as duplicates will always all be routed, but duplicates are
3916 presumably rare, so I don't expect this is of any significance.
3918 For deliveries to pipes, files, and autoreplies, the duplicate checking
3919 still happens during the routing process, since they are not going to be
3922 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3923 It corrects a timeout issue with spamd. This is Ian's comment: "The
3924 background is that sometimes spamd either never reads data from a
3925 connection it has accepted, or it never writes response data. The exiscan
3926 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3927 blindly assumes that writes won't block so it may never time out."
3929 PH/39 Allow G after quota size as well as K and M.
3931 PH/40 The value set for $authenticated_id in an authenticator may not contain
3932 binary zeroes or newlines because the value is written to log lines and
3933 to spool files. There was no check on this. Now the value is run through
3934 the string_printing() function so that such characters are converted to
3935 printable escape sequences.
3937 PH/41 $message_linecount is a new variable that contains the total number of
3938 lines in the message. Compare $body_linecount, which is the count for the
3941 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3942 bounce and delay warning messages, except in certain special cases, which
3945 (a) An SMTP error message from a remote host;
3946 (b) A message specified in a :fail: redirection;
3947 (c) A message specified in a "fail" command in a system filter;
3948 (d) A message specified in a FAIL return from the queryprogram router;
3949 (e) A message specified by the cannot_route_message router option.
3951 In these cases only, Exim does include the error details in bounce and
3952 warning messages. There are also a few cases where bland messages such
3953 as "unrouteable address" or "local delivery error" are given.
3955 PH/43 $value is now also set for the "else" part of a ${run expansion.
3957 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3958 being worked on, but at least Exim now implements the latest version to
3961 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3962 process to complete was treated as a delivery failure, a timeout while
3963 writing the message to the pipe was logged, but erroneously treated as a
3964 successful delivery. Such timeouts include transport filter timeouts. For
3965 consistency with the overall process timeout, these timeouts are now
3966 treated as errors, giving rise to delivery failures by default. However,
3967 there is now a new Boolean option for the pipe transport called
3968 timeout_defer, which, if set TRUE, converts the failures into defers for
3969 both kinds of timeout. A transport filter timeout is now identified in
3972 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3973 systems where "make" and "gmake" are different, calling "gmake" at top
3974 level broke things. I've arranged for the value of $(MAKE) to be passed
3975 from the Makefile to this script so that it can call the same version of
3979 A note about Exim versions 4.44 and 4.50
3980 ----------------------------------------
3982 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3983 changes of various kinds. As a consequence, a big documentation update was
3984 needed. This delayed the release for rather longer than seemed good, especially
3985 in the light of a couple of (minor) security issues. Therefore, the changes
3986 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3987 release. So 4.44 and 4.50 are in effect two different branches that both start
3990 I have left the 4.50 change log unchanged; it contains all the changes since
3991 4.43. The change log for 4.44 is below; many of its items are identical to
3992 those for 4.50. This seems to be the most sensible way to preserve the
3993 historical information.
3999 1. Minor wording change to the doc/README.SIEVE file.
4001 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4002 computation of the current number of files was incorrect.
4004 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4005 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4008 4. Give more explanation in the error message when the command for a transport
4009 filter fails to execute.
4011 5. There are several places where Exim runs a non-Exim command in a
4012 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4013 was being done only for the command run by the queryprogram router. It is
4014 now done for all such subprocesses. The other cases are: ${run, transport
4015 filters, and the commands run by the lmtp and pipe transports.
4017 6. Added CONFIGURE_GROUP build-time option.
4019 7. Some older OS have a limit of 256 on the maximum number of file
4020 descriptors. Exim was using setrlimit() to set 1000 as a large value
4021 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4022 systems. I've change it so that if it can't get 1000, it tries for 256.
4024 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4025 was an oversight, and furthermore, ever since the addition of extra
4026 controls (e.g. 4.43/32), the checks on when to allow different forms of
4027 "control" were broken. There should now be diagnostics for all cases when a
4028 control that does not make sense is encountered.
4030 9. Added the /retain_sender option to "control=submission".
4032 10. $recipients is now available in the predata ACL (oversight).
4034 11. Tidy the search cache before the fork to do a delivery from a message
4035 received from the command line. Otherwise the child will trigger a lookup
4036 failure and thereby defer the delivery if it tries to use (for example) a
4037 cached ldap connection that the parent has called unbind on.
4039 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4040 of $address_data from the recipient verification was clobbered by the
4041 sender verification.
4043 13. The value of address_data from a sender verification is now available in
4044 $sender_address_data in subsequent conditions in the ACL statement.
4046 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4048 15. Added a new option "connect=<time>" to callout options, to set a different
4051 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4052 was its contents. (It was OK if the option was not defined at all.)
4054 17. A "Completed" log line is now written for messages that are removed from
4055 the spool by the -Mrm option.
4057 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4058 information about exactly what failed.
4060 19. Added -dd to debug only the daemon process.
4062 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4063 handles timeouts, both on the server side and network timeouts. Renamed the
4064 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4066 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4067 It is now set to "smtps".
4069 22. $host_address is now set to the target address during the checking of
4070 ignore_target_hosts.
4072 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4073 being passed; this would have caused $sender_host_name to have been used if
4074 matching the list had actually called for a host name (not very likely,
4075 since this list is usually IP addresses). A host name is now passed as
4078 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4079 code that specifies a non-restarting handler (typically sigaction() in
4080 modern systems) in an attempt to fix a rare and obscure crash bug.
4082 25. Narrowed the window for a race in the daemon that could cause it to ignore
4083 SIGCHLD signals. This is not a major problem, because they are used only to
4084 wake it up if nothing else does.
4086 26. A malformed maildirsize file could cause Exim to calculate negative values
4087 for the mailbox size or file count. Odd effects could occur as a result.
4088 The maildirsize information is now recalculated if the size or filecount
4091 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4092 support for a long time. Removed HAVE_SYS_VFS_H.
4094 28. Installed the latest version of exipick from John Jetmore.
4096 29. In an address list, if the pattern was not a regular expression, an empty
4097 subject address (from a bounce message) matched only if the pattern was an
4098 empty string. Non-empty patterns were not even tested. This was the wrong
4099 because it is perfectly reasonable to use an empty address as part of a
4100 database query. An empty address is now tested by patterns that are
4101 lookups. However, all the other forms of pattern expect the subject to
4102 contain a local part and a domain, and therefore, for them, an empty
4103 address still always fails if the pattern is not itself empty.
4105 30. Exim went into a mad DNS loop when attempting to do a callout where the
4106 host was specified on an smtp transport, and looking it up yielded more
4107 than one IP address.
4109 31. Re-factored the code for checking spool and log partition space into a
4110 function that finds that data and another that does the check. The former
4111 is then used to implement four new variables: $spool_space, $log_space,
4112 $spool_inodes, and $log_inodes.
4114 32. The RFC2047 encoding function was originally intended for short strings
4115 such as real names; it was not keeping to the 75-character limit for
4116 encoded words that the RFC imposes. It now respects the limit, and
4117 generates multiple encoded words if necessary. To be on the safe side, I
4118 have increased the buffer size for the ${rfc2047: expansion operator from
4121 33. It is now permitted to omit both strings after an "if" condition; if the
4122 condition is true, the result is "true". As before, when the second string
4123 is omitted, a false condition yields an empty string. This makes it less
4124 cumbersome to write custom ACL and router conditions.
4126 34. Failure to deliver a bounce message always caused it to be frozen, even if
4127 there was an errors_to setting on the router. The errors_to setting is now
4130 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4131 canonical form (fully expanded) before being placed in
4132 $sender_host_address.
4134 36. The table in the code that translates DNS record types into text (T_A to
4135 "A" for instance) was missing entries for NS and CNAME. It is just possible
4136 that this could have caused confusion if both these types were looked up
4137 for the same domain, because the text type is used as part of Exim's
4138 per-process caching. But the chance of anyone hitting this buglet seems
4141 37. The dnsdb lookup has been extended in a number of ways.
4143 (1) There is a new type, "zns", which walks up the domain tree until it
4144 finds some nameserver records. It should be used with care.
4146 (2) There is a new type, "mxh", which is like "mx" except that it returns
4147 just the host names, not the priorities.
4149 (3) It is now possible to give a list of domains (or IP addresses) to be
4150 looked up. The behaviour when one of the lookups defers can be
4151 controlled by a keyword.
4153 (4) It is now possible to specify the separator character for use when
4154 multiple records are returned.
4156 38. The dnslists ACL condition has been extended: it is now possible to supply
4157 a list of IP addresses and/or domains to be looked up in a particular DNS
4160 39. Added log_selector=+queue_time_overall.
4162 40. When running the queue in the test harness, wait just a tad after forking a
4163 delivery process, to get repeatability of debugging output.
4165 41. Include certificate and key file names in error message when GnuTLS fails
4166 to set them up, because the GnuTLS error message doesn't include the name
4167 of the failing file when there is a problem reading it.
4169 42. Allow both -bf and -bF in the same test run.
4171 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4173 44. The "Exiscan patch" is now merged into the mainline Exim source.
4175 45. Sometimes the final signoff response after QUIT could fail to get
4176 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4177 < 0 before doing a fflush(). This bug looks as though it goes back to the
4178 introduction of TLS in release 3.20, but "sometimes" must have been rare
4179 because the tests only now provoked it.
4181 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4182 (this can affect the format of dates).
4184 47. exim_tidydb, when checking for the continued existence of a message for
4185 which it has found a message-specific retry record, was not finding
4186 messages that were in split spool directories. Consequently, it was
4187 deleting retry records that should have stayed in existence.
4189 48. Steve fixed some bugs in eximstats.
4191 49. The SPA authentication driver was not abandoning authentication and moving
4192 on to the next authenticator when an expansion was forced to fail,
4193 contradicting the general specification for all authenticators. Instead it
4194 was generating a temporary error. It now behaves as specified.
4196 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4197 (the order specifies the preference for clients). The order is now AES256,
4198 AES128, 3DES, ARCFOUR128.
4200 51. Small patch to Sieve code - explicitly set From: when generating an
4203 52. Exim crashed if a remote delivery caused a very long error message to be
4204 recorded - for instance if somebody sent an entire SpamAssassin report back
4205 as a large number of 550 error lines. This bug was coincidentally fixed by
4206 increasing the size of one of Exim's internal buffers (big_buffer) that
4207 happened as part of the Exiscan merge. However, to be on the safe side, I
4208 have made the code more robust (and fixed the comments that describe what
4211 53. Now that there can be additional text after "Completed" in log lines (if
4212 the queue_time_overall log selector is set), a one-byte patch to exigrep
4213 was needed to allow it to recognize "Completed" as not the last thing in
4216 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4217 patch that reportedly fixes this has been added. I am not expert enough to
4218 create a test for it. This is what the patch creator wrote:
4220 "I found a little strange behaviour of ldap code when working with
4221 Windows 2003 AD Domain, where users was placed in more than one
4222 Organization Units. When I tried to give exim partial DN, the exit code
4223 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4224 But simultaneously result of request was absolutely normal ldap result,
4225 so I produce this patch..."
4227 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4228 REFERENCE, so I have modified the code to exclude the patch when that macro
4231 55. Some experimental protocols are using DNS PTR records for new purposes. The
4232 keys for these records are domain names, not reversed IP addresses. The
4233 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4234 leaves it alone. Component reversal etc. now happens only for IP addresses.
4237 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4239 57. Double the size of the debug message buffer (to 2048) so that more of very
4240 long debug lines gets shown.
4242 58. The exicyclog utility now does better if the number of log files to keep
4243 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4245 59. Two changes related to the smtp_active_hostname option:
4247 (1) $smtp_active_hostname is now available as a variable.
4248 (2) The default for smtp_banner uses $smtp_active_hostname instead
4249 of $primary_hostname.
4251 60. The host_aton() function is supposed to be passed a string that is known
4252 to be a valid IP address. However, in the case of IPv6 addresses, it was
4253 not checking this. This is a hostage to fortune. Exim now panics and dies
4254 if the condition is not met. A case was found where this could be provoked
4255 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4256 components; fortuitously, this particular loophole had already been fixed
4257 by change 4.50/55 above.
4259 If there are any other similar loopholes, the new check in host_aton()
4260 itself should stop them being exploited. The report I received stated that
4261 data on the command line could provoke the exploit when Exim was running as
4262 exim, but did not say which command line option was involved. All I could
4263 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4264 running as the user.
4267 61. There was a buffer overflow vulnerability in the SPA authentication code
4268 (which came originally from the Samba project). I have added a test to the
4269 spa_base64_to_bits() function which I hope fixes it.
4272 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4273 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4275 63. The daemon start-up calls getloadavg() while still root for those OS that
4276 need the first call to be done as root, but it missed one case: when
4277 deliver_queue_load_max is set with deliver_drop_privilege. This is
4278 necessary for the benefit of the queue runner, because there is no re-exec
4279 when deliver_drop_privilege is set.
4281 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4282 This has been fixed.
4284 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4285 was in use, was not putting the data itself into the right store pool;
4286 consequently, it could be overwritten for a subsequent message in the same
4287 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4290 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4292 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4293 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4294 the function were treating the return as a boolean value, which happened to
4295 work because 0=false and not-0=true, but is not correct code.
4297 68. The host_aton() function was not handling scoped IPv6 addresses (those
4298 with, for example, "%eth0" on the end) correctly.
4300 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4301 negated items (that is, ~something) in unsigned ints. Some compilers
4302 apparently mutter when there is no cast.
4304 70. If an address verification called from an ACL failed, and did not produce a
4305 user-specific message (i.e. there was only a "system" message), nothing was
4306 put in $acl_verify_message. In this situation, it now puts the system
4309 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4310 session; change 4.31/43 added the unwanted input to the log line - except
4311 that it did not do this in the start of session case. It now does.
4313 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4314 This is wrong and can cause the other end to generate a synchronization
4315 error if it is another Exim or anything else that does the synchronization
4316 check. A QUIT command is no longer sent after a timeout.
4318 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4319 during host lookups.
4321 74. The defer_ok option of callout verification was not working if it was used
4322 when verifying addresses in header lines, that is, for this case:
4324 verify = header_sender/callout=defer_ok
4326 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4327 those file descriptors could be used for SMTP connections. If anything
4328 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4329 could be sent to the SMTP client, causing chaos. The daemon now opens
4330 stdin, stdout, and stderr to /dev/null when it puts itself into the
4333 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4334 log by default. The user can override this with suitable Perl magic.
4336 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4337 the log message when discard triggers, was not working for the DATA ACL or
4338 for the non-SMTP ACL.
4340 78. Error message wording change in sieve.c.
4342 79. If smtp_accept_max_per_host was set, the number of connections could be
4343 restricted to fewer than expected, because the daemon was trying to set up
4344 a new connection before checking whether the processes handling previous
4345 connections had finished. The check for completed processes is now done
4346 earlier. On busy systems, this bug wouldn't be noticed because something
4347 else would have woken the daemon, and it would have reaped the completed
4350 80. If a message was submitted locally by a user whose login name contained one
4351 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4352 It caused a spool format error. I have fixed the spool reading code. A
4353 related problem was that the "from" clause in the Received: line became
4354 illegal because of the space(s). It is now covered by ${quote_local_part.
4356 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4359 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4361 83. Patch from Sieve maintainer to fix unterminated string problem in
4362 "vacation" handling.
4364 84. Some minor changes to the Linux configuration files to help with other
4365 OS variants using glibc.
4367 85. One more patch for Sieve to update vacation handling to latest spec.
4370 ----------------------------------------------------
4371 See the note above about the 4.44 and 4.50 releases.
4372 ----------------------------------------------------
4378 1. Change 4.43/35 introduced a bug that caused file counts to be
4379 incorrectly computed when quota_filecount was set in an appendfile
4382 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4383 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4386 3. Give more explanation in the error message when the command for a transport
4387 filter fails to execute.
4389 4. There are several places where Exim runs a non-Exim command in a
4390 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4391 was being done only for the command run by the queryprogram router. It is
4392 now done for all such subprocesses. The other cases are: ${run, transport
4393 filters, and the commands run by the lmtp and pipe transports.
4395 5. Some older OS have a limit of 256 on the maximum number of file
4396 descriptors. Exim was using setrlimit() to set 1000 as a large value
4397 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4398 systems. I've change it so that if it can't get 1000, it tries for 256.
4400 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4401 was an oversight, and furthermore, ever since the addition of extra
4402 controls (e.g. 4.43/32), the checks on when to allow different forms of
4403 "control" were broken. There should now be diagnostics for all cases when a
4404 control that does not make sense is encountered.
4406 7. $recipients is now available in the predata ACL (oversight).
4408 8. Tidy the search cache before the fork to do a delivery from a message
4409 received from the command line. Otherwise the child will trigger a lookup
4410 failure and thereby defer the delivery if it tries to use (for example) a
4411 cached ldap connection that the parent has called unbind on.
4413 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4414 of $address_data from the recipient verification was clobbered by the
4415 sender verification.
4417 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4418 was its contents. (It was OK if the option was not defined at all.)
4420 11. A "Completed" log line is now written for messages that are removed from
4421 the spool by the -Mrm option.
4423 12. $host_address is now set to the target address during the checking of
4424 ignore_target_hosts.
4426 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4427 being passed; this would have caused $sender_host_name to have been used if
4428 matching the list had actually called for a host name (not very likely,
4429 since this list is usually IP addresses). A host name is now passed as
4432 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4433 code that specifies a non-restarting handler (typically sigaction() in
4434 modern systems) in an attempt to fix a rare and obscure crash bug.
4436 15. Narrowed the window for a race in the daemon that could cause it to ignore
4437 SIGCHLD signals. This is not a major problem, because they are used only to
4438 wake it up if nothing else does.
4440 16. A malformed maildirsize file could cause Exim to calculate negative values
4441 for the mailbox size or file count. Odd effects could occur as a result.
4442 The maildirsize information is now recalculated if the size or filecount
4445 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4446 support for a long time. Removed HAVE_SYS_VFS_H.
4448 18. Updated exipick to current release from John Jetmore.
4450 19. Allow an empty sender to be matched against a lookup in an address list.
4451 Previously the only cases considered were a regular expression, or an
4454 20. Exim went into a mad DNS lookup loop when doing a callout where the
4455 host was specified on the transport, if the DNS lookup yielded more than
4458 21. The RFC2047 encoding function was originally intended for short strings
4459 such as real names; it was not keeping to the 75-character limit for
4460 encoded words that the RFC imposes. It now respects the limit, and
4461 generates multiple encoded words if necessary. To be on the safe side, I
4462 have increased the buffer size for the ${rfc2047: expansion operator from
4465 22. Failure to deliver a bounce message always caused it to be frozen, even if
4466 there was an errors_to setting on the router. The errors_to setting is now
4469 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4470 canonical form (fully expanded) before being placed in
4471 $sender_host_address.
4473 24. Updated eximstats to version 1.33
4475 25. Include certificate and key file names in error message when GnuTLS fails
4476 to set them up, because the GnuTLS error message doesn't include the name
4477 of the failing file when there is a problem reading it.
4479 26. Expand error message when OpenSSL has problems setting up cert/key files.
4482 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4483 (this can affect the format of dates).
4485 28. exim_tidydb, when checking for the continued existence of a message for
4486 which it has found a message-specific retry record, was not finding
4487 messages that were in split spool directories. Consequently, it was
4488 deleting retry records that should have stayed in existence.
4490 29. eximstats updated to version 1.35
4491 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4492 1.35 - bugfix such that pie charts by volume are generated correctly
4494 30. The SPA authentication driver was not abandoning authentication and moving
4495 on to the next authenticator when an expansion was forced to fail,
4496 contradicting the general specification for all authenticators. Instead it
4497 was generating a temporary error. It now behaves as specified.
4499 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4500 (the order specifies the preference for clients). The order is now AES256,
4501 AES128, 3DES, ARCFOUR128.
4503 31. Small patch to Sieve code - explicitly set From: when generating an
4506 32. Exim crashed if a remote delivery caused a very long error message to be
4507 recorded - for instance if somebody sent an entire SpamAssassin report back
4508 as a large number of 550 error lines. This bug was coincidentally fixed by
4509 increasing the size of one of Exim's internal buffers (big_buffer) that
4510 happened as part of the Exiscan merge. However, to be on the safe side, I
4511 have made the code more robust (and fixed the comments that describe what
4514 33. Some experimental protocols are using DNS PTR records for new purposes. The
4515 keys for these records are domain names, not reversed IP addresses. The
4516 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4517 leaves it alone. Component reversal etc. now happens only for IP addresses.
4520 34. The host_aton() function is supposed to be passed a string that is known
4521 to be a valid IP address. However, in the case of IPv6 addresses, it was
4522 not checking this. This is a hostage to fortune. Exim now panics and dies
4523 if the condition is not met. A case was found where this could be provoked
4524 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4525 components; fortuitously, this particular loophole had already been fixed
4526 by change 4.50/55 or 4.44/33 above.
4528 If there are any other similar loopholes, the new check in host_aton()
4529 itself should stop them being exploited. The report I received stated that
4530 data on the command line could provoke the exploit when Exim was running as
4531 exim, but did not say which command line option was involved. All I could
4532 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4533 running as the user.
4536 35. There was a buffer overflow vulnerability in the SPA authentication code
4537 (which came originally from the Samba project). I have added a test to the
4538 spa_base64_to_bits() function which I hope fixes it.
4541 36. The daemon start-up calls getloadavg() while still root for those OS that
4542 need the first call to be done as root, but it missed one case: when
4543 deliver_queue_load_max is set with deliver_drop_privilege. This is
4544 necessary for the benefit of the queue runner, because there is no re-exec
4545 when deliver_drop_privilege is set.
4547 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4548 was in use, was not putting the data itself into the right store pool;
4549 consequently, it could be overwritten for a subsequent message in the same
4550 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4553 38. Sometimes the final signoff response after QUIT could fail to get
4554 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4555 < 0 before doing a fflush(). This bug looks as though it goes back to the
4556 introduction of TLS in release 3.20, but "sometimes" must have been rare
4557 because the tests only now provoked it.
4563 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4564 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4565 Now it returns an integer. A number of calls were still expecting a T/F
4566 return. Fortuitously, in all cases, the tests worked in OK situations,
4567 which is the norm. However, things would have gone wrong on any write
4568 failures on the smtp file descriptor. This function is used when sending
4569 messages over SMTP and also when doing verify callouts.
4571 2. When Exim is called to do synchronous delivery of a locally submitted
4572 message (the -odf or -odi options), it no longer closes stderr before doing
4575 3. Implemented the mua_wrapper option.
4577 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4579 5. Implemented the functions header_remove(), header_testname(),
4580 header_add_at_position(), and receive_remove_recipient(), and exported them
4583 6. If an ACL "warn" statement specified the addition of headers, Exim already
4584 inserted X-ACL-Warn: at the start if there was no header name. However, it
4585 was not making this test for the second and subsequent header lines if
4586 there were newlines in the string. This meant that an invalid header could
4587 be inserted if Exim was badly configured.
4589 7. Allow an ACL "warn" statement to add header lines at the start or after all
4590 the Received: headers, as well as at the end.
4592 8. Added the rcpt_4xx retry error code.
4594 9. Added postmaster_mailfrom=xxx to callout verification option.
4596 10. Added mailfrom=xxxx to the callout verification option, for verify=
4599 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4600 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4602 12. Inserted some casts to stop certain compilers warning when using pointer
4603 differences as field lengths or precisions in printf-type calls (mostly
4604 affecting debugging statements).
4606 13. Added optional readline() support for -be (dynamically loaded).
4608 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4609 same clock tick as a message's arrival, so that its received time was the
4610 same as the "first fail" time on the retry record, and that message
4611 remained on the queue past the ultimate address timeout, every queue runner
4612 would try a delivery (because it was past the ultimate address timeout) but
4613 after another failure, the ultimate address timeout, which should have then
4614 bounced the address, did not kick in. This was a "< instead of <=" error;
4615 in most cases the first failure would have been in the next clock tick
4616 after the received time, and all would be well.
4618 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4619 being recognized when the domain list was tested by the match_domain
4620 condition in an expansion string.
4622 16. Added the ${str2b64: operator.
4624 17. Exim was always calling setrlimit() to set a large limit for the number of
4625 processes, without checking whether the existing limit was already
4626 adequate. (It did check for the limit on file descriptors.) Furthermore,
4627 errors from getrlimit() and setrlimit() were being ignored. Now they are
4628 logged to the main and panic logs, but Exim does carry on, to try to do its
4629 job under whatever limits there are.
4631 18. Imported PCRE 5.0.
4633 19. Trivial typo in log message " temporarily refused connection" (the leading
4636 20. If the log selector return_path_on_delivery was set and an address was
4637 redirected to /dev/null, the delivery process crashed because it assumed
4638 that a return path would always be set for a "successful" delivery. In this
4639 case, the whole delivery is bypassed as an optimization, and therefore no
4642 21. Internal re-arrangement: the function for sending a challenge and reading
4643 a response while authentication was assuming a zero-terminated challenge
4644 string. It's now changed to take a pointer and a length, to allow for
4645 binary data in such strings.
4647 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4649 23. Exim was not respecting finduser_retries when seeking the login of the
4650 uid under which it was called; it was always trying 10 times. (The default
4651 setting of finduser_retries is zero.) Also, it was sleeping after the final
4652 failure, which is pointless.
4654 24. Implemented tls_on_connect_ports.
4656 25. Implemented acl_smtp_predata.
4658 26. If the domain in control=submission is set empty, Exim assumes that the
4659 authenticated id is a complete email address when it generates From: or
4660 Sender: header lines.
4662 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4663 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4664 chown and chgrp in /bin and hostname in /usr/bin.
4666 28. Exim was keeping the "process log" file open after each use, just as it
4667 does for the main log. This opens the possibility of it remaining open for
4668 long periods when the USR1 signal hits a daemon. Occasional processlog
4669 errors were reported, that could have been caused by this. Anyway, it seems
4670 much more sensible not to leave this file open at all, so that is what now
4673 29. The long-running daemon process does not normally write to the log once it
4674 has entered its main loop, and it closes the log before doing so. This is
4675 so that log files can straightforwardly be renamed and moved. However,
4676 there are a couple of unusual error situations where the daemon does write
4677 log entries, and I had neglected to close the log afterwards.
4679 30. The text of an SMTP error response that was received during a remote
4680 delivery was being truncated at 512 bytes. This is too short for some of
4681 the long messages that one sometimes sees. I've increased the limit to
4684 31. It is now possible to make retry rules that apply only when a message has a
4685 specific sender, in particular, an empty sender.
4687 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4688 it possible to be selective about when SMTP synchronization is enforced.
4690 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4692 32. Implemented hosts_connection_nolog.
4694 33. Added an ACL for QUIT.
4696 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4699 35. Added mailbox_size and mailbox_filecount to appendfile.
4701 36. Added control = no_multiline_responses to ACLs.
4703 37. There was a bug in the logic of the code that waits for the clock to tick
4704 in the case where the clock went backwards by a substantial amount such
4705 that the microsecond fraction of "now" was more than the microsecond
4706 fraction of "then" (but the whole seconds number was less).
4708 38. Added support for the libradius Radius client library this is found on
4709 FreeBSD (previously only the radiusclient library was supported).
4715 1. When certain lookups returned multiple values in the form name=value, the
4716 quoting of the values was not always being done properly. Specifically:
4717 (a) If the value started with a double quote, but contained no whitespace,
4719 (b) If the value contained whitespace other than a space character (i.e.
4720 tabs or newlines or carriage returns) it was not quoted.
4721 This fix has been applied to the mysql and pgsql lookups by writing a
4722 separate quoting function and calling it from the lookup code. The fix
4723 should probably also be applied to nisplus, ibase and oracle lookups, but
4724 since I cannot test any of those, I have not disturbed their existing code.
4726 2. A hit in the callout cache for a specific address caused a log line with no
4727 reason for rejecting RCPT. Now it says "Previous (cached) callout
4728 verification failure".
4730 3. There was an off-by-one bug in the queryprogram router. An over-long
4731 return line was truncated at 256 instead of 255 characters, thereby
4732 overflowing its buffer with the terminating zero. As well as fixing this, I
4733 have increased the buffer size to 1024 (and made a note to document this).
4735 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4736 when Exim is waiting for an SMTP response from a remote server, Exim
4737 restarts its select() call on the socket, thereby resetting its timeout.
4738 This is not a problem when such interrupts are rare. Somebody set up a cron
4739 job to run exiwhat every 2 minutes, which is less than the normal select()
4740 timeout (5 or 10 minutes). This meant that the select() timeout never
4741 kicked in because it was always reset. I have fixed this by comparing the
4742 time when an interrupt arrives with the time at the start of the first call
4743 to select(). If more time than the timeout has elapsed, the interrupt is
4744 treated as a timeout.
4746 5. Some internal re-factoring in preparation for the addition of Sieve
4747 extensions (by MH). In particular, the "personal" test is moved to a
4748 separate function, and given an option for scanning Cc: and Bcc: (which is
4749 not set for Exim filters).
4751 6. When Exim created an email address using the login of the caller as the
4752 local part (e.g. when creating a From: or Sender: header line), it was not
4753 quoting the local part when it contained special characters such as @.
4755 7. Installed new OpenBSD configuration files.
4757 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4758 try to make them clearer.
4760 9. Callout options, other than the timeout value, were being ignored when
4761 verifying sender addresses in header lines. For example, when using
4763 verify = header_sender/callout=no_cache
4765 the cache was (incorrectly) being used.
4767 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4768 only the Cygwin environment.
4770 11. When return_path_on_delivery was set as a log selector, if different remote
4771 addresses in the same message used different return paths and parallel
4772 remote delivery occurred, the wrong values would sometimes be logged.
4773 (Whenever a remote delivery process finished, the return path value from
4774 the most recently started remote delivery process was logged.)
4776 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4777 lines when AUTH and/or TLS are in use. This is the "received protocol"
4778 field. Exim used to use "asmtp" for authenticated SMTP, without any
4779 indication (in the protocol name) for TLS use. Now it follows the RFC and
4780 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4781 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4782 names appear in log lines as well as in Received: header lines.
4784 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4785 extensions, and comparison tests, and to fix some bugs.
4787 14. Changes to the "personal" filter test:
4789 (1) The test was buggy in that it was just doing the equivalent of
4790 "contains" tests on header lines. For example, if a user's address was
4791 anne@some.where, the "personal" test would incorrectly be true for
4793 To: susanne@some.where
4795 This test is now done by extracting each address from the header in turn,
4796 and checking the entire address. Other tests that are part of "personal"
4797 are now done using regular expressions (for example, to check local parts
4798 of addresses in From: header lines).
4800 (2) The list of non-personal local parts in From: addresses has been
4801 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4802 taken from the Sieve specification recommendations.
4804 (3) If the message contains any header line starting with "List-" it is
4805 treated as non-personal.
4807 (4) The test for "circular" in the Subject: header line has been removed
4808 because it now seems ill-conceived.
4810 15. Minor typos in src/EDITME comments corrected.
4812 16. Installed latest exipick from John Jetmore.
4814 17. If headers_add on a router specified a text string that was too long for
4815 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4816 of string_sprintf() is now avoided.
4818 18. $message_body_size was not set (it was always zero) when running the DATA
4819 ACL and the local_scan() function.
4821 19. For the "mail" command in an Exim filter, no default was being set for
4822 the once_repeat time, causing a random time value to be used if "once" was
4823 specified. (If the value happened to be <= 0, no repeat happened.) The
4824 default is now 0s, meaning "never repeat". The "vacation" command was OK
4825 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4826 (I found it when inspecting the code).
4828 20. There is now an overall timeout for performing a callout verification. It
4829 defaults to 4 times the callout timeout, which applies to individual SMTP
4830 commands during the callout. The overall timeout applies when there is more
4831 than one host that can be tried. The timeout is checked before trying the
4832 next host. This prevents very long delays if there are a large number of
4833 hosts and all are timing out (e.g. when the network connections are timing
4834 out). The value of the overall timeout can be changed by specifying an
4835 additional sub-option for "callout", called "maxwait". For example:
4837 verify = sender/callout=5s,maxwait=20s
4839 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4840 to the end before writing, but this should make it even safer).
4842 22. Exim was forgetting that it had advertised PIPELINING for the second and
4843 subsequent messages on an SMTP connection. It was also not resetting its
4844 memory on STARTTLS and an internal HELO.
4846 23. When Exim logs an SMTP synchronization error within a session, it now
4847 records whether PIPELINING has been advertised or not.
4849 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4850 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4851 rather than long int.
4853 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4855 26. Added the never_mail option to autoreply.
4861 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4862 crash if the getsockname() call failed; this can happen if a connection is
4863 closed very soon after it is established. The problem was simply in the
4864 order in which certain operations were done, causing Exim to try to write
4865 to the SMTP stream before it had set up the file descriptor. The bug has
4866 been fixed by making things happen in the correct order.
4872 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4873 before the connection was closed, thus losing the rejection response.
4875 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4876 some early Solaris releases, but causes trouble in current releases where
4877 socklen_t is defined.
4879 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4882 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4885 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4886 that releasing the top of it at the end releases what was used for sub-
4887 expansions (unless the block got too big). However, discard this block if
4888 the first thing is a variable or header, so that we can use its block when
4889 it is dynamic (useful for very large $message_headers, for example).
4891 6. Lookups now cache *every* query, not just the most recent. A new, separate
4892 store pool is used for this. It can be recovered when all lookup caches are
4893 flushed. Lookups now release memory at the end of their result strings.
4894 This has involved some general refactoring of the lookup sources.
4896 7. Some code has been added to the store_xxx() functions to reduce the amount
4897 of flapping under certain conditions.
4899 8. log_incoming_interface used to affect only the <= reception log lines. Now
4900 it causes the local interface and port to be added to several more SMTP log
4901 lines, for example "SMTP connection from", and rejection lines.
4903 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4905 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4907 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4908 could be overwritten at the end of the current message (or the start of a
4909 new message if it was set in a HELO ACL). The value is now preserved for
4910 the duration of the SMTP connection.
4912 12. If a transport had a headers_rewrite setting, and a matching header line
4913 contained an unqualified address, that address was qualified, even if it
4914 did not match any rewriting rules. The underlying bug was that the values
4915 of the flags that permit the existence of unqualified sender and recipient
4916 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4917 non-local messages, and by -bnq for local messages) were not being
4918 preserved with the message after it was received.
4920 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4921 "next input=" as part of the text comprising the host identity instead of
4922 the correct text. The code was using the same buffer for two different
4923 strings. However, depending on which order the printing function evaluated
4924 its arguments, the bug did not always show up. Under Linux, for example, my
4925 test suite worked just fine.
4927 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4928 doesn't work with some older versions of Perl. It has been changed to "my",
4929 which in any case is probably the better facility to use.
4931 15. A really picky compiler found some instances of statements for creating
4932 error messages that either had too many or two few arguments for the format
4935 16. The size of the buffer for calls to the DNS resolver has been increased
4936 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4937 for addresses that have a lot of PTR records. This alleviates a problem; it
4938 does not fully solve it.
4940 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4941 buffer now truncates the list and logs the incident, which is the same
4942 action as happens when Exim is looking up a host name and its aliases.
4943 Previously in this situation something unpredictable would happen;
4944 sometimes it was "internal error: store_reset failed".
4946 18. If a server dropped the connection unexpectedly when an Exim client was
4947 using GnuTLS and trying to read a response, the client delivery process
4948 crashed while trying to generate an error log message.
4950 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4951 string, for example:
4953 warn message = H1: something\nH2: something
4955 the text was added as a single header line from Exim's point of view
4956 though it ended up OK in the delivered message. However, searching for the
4957 second and subsequent header lines using $h_h2: did not work. This has been
4958 fixed. Similarly, if a system filter added multiple headers in this way,
4959 the routers could not see them.
4961 20. Expanded the error message when iplsearch is called with an invalid key to
4962 suggest using net-iplsearch in a host list.
4964 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4965 ACLs are no longer actually imposed (and a message to that effect is
4968 22. If a "gecos" field in a passwd entry contained escaped characters, in
4969 particular, if it contained a \" sequence, Exim got it wrong when building
4970 a From: or a Sender: header from that name. A second bug also caused
4971 incorrect handling when an unquoted " was present following a character
4972 that needed quoting.
4974 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4975 was not being matched caselessly.
4977 24. Arranged for all hyphens in the exim.8 source to be escaped with
4980 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4981 properly. Recipient callout cache records were still being keyed to include
4982 the sender, even when use_sender was set false. This led to far more
4983 callouts that were necessary. The sender is no longer included in the key
4984 when use_sender is false.
4986 26. Added "control = submission" modifier to ACLs.
4988 27. Added the ${base62d: operator to decode base 62 numbers.
4990 28. dnsdb lookups can now access SRV records.
4992 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4993 the configuration file.
4995 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4996 (-v) mode. This makes the output for a verbose queue run more intelligible.
4998 31. Added a use_postmaster feature to recipient callouts.
5000 32. Added the $body_zerocount variable, containing the number of binary zero
5001 bytes in the message body.
5003 33. The time of last modification of the "new" subdirectory is now used as the
5004 "mailbox time last read" when there is a quota error for a maildir
5007 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5009 35. Added +ignore_unknown as a special item in host lists.
5011 36. Code for decoding IPv6 addresses in host lists is now included, even if
5012 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5013 address was recognized as an IP address, but was then not correctly decoded
5014 into binary, causing unexpected and incorrect effects when compared with
5021 1. Very minor rewording of debugging text in manualroute to say "list of
5022 hosts" instead of "hostlist".
5024 2. If verify=header_syntax was set, and a header line with an unqualified
5025 address (no domain) and a large number of spaces between the end of the
5026 name and the colon was received, the reception process suffered a buffer
5027 overflow, and (when I tested it) crashed. This was caused by some obsolete
5028 code that should have been removed. The fix is to remove it!
5030 3. When running in the test harness, delay a bit after writing a bounce
5031 message to get a bit more predictability in the log output.
5033 4. Added a call to search_tidyup() just before forking a reception process. In
5034 theory, someone could use a lookup in the expansion of smtp_accept_max_
5035 per_host which, without the tidyup, could leave open a database connection.
5037 5. Added the variables $recipient_data and $sender_data which get set from a
5038 lookup success in an ACL "recipients" or "senders" condition, or a router
5039 "senders" option, similar to $domain_data and $local_part_data.
5041 6. Moved the writing of debug_print from before to after the "senders" test
5044 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5045 problems for message scanning, either using a data ACL, or using
5046 local_scan() because the Received: header was not generated till after they
5047 were called (in order to set the time as the time of reception completion).
5048 I have revised the way this works. The header is now generated after the
5049 body is received, but before the ACL or local_scan() are called. After they
5050 are run, the timestamp in the header is updated.
5056 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5057 before starting a queue runner without re-exec. This happened only when
5058 deliver_drop_privilege was set or when the Exim user was set to root. The
5059 effect of the bug was that timeouts during subsequent deliveries caused
5060 crashes instead of being properly handled. The handler is now left at its
5061 default (and expected) setting.
5063 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5064 message, again when deliver_drop_privilege is set or Exim is run as root.
5065 The bug described in (1) was not present in this case, but the tidying up
5066 of the other signals was missing. I have made the two cases consistent.
5068 3. The ignore_target_hosts setting on a manualroute router was being ignored
5069 for hosts that were looked up using the /MX notation.
5071 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5074 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5075 operated on the sender address. After changing the $sender_address to <>
5076 for the sender address verify, Exim was re-instated it as the original
5077 (before rewriting) address, but remembering that it had rewritten it, so it
5078 wasn't rewriting it again. This bug also had the effect of breaking the
5079 sender address verification caching when the sender address was rewritten.
5081 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5082 This has been changed so that if the ip literal address matches
5083 ignore_target_hosts, the router declines.
5085 7. Added expansion conditions match_domain, match_address, and match_local_
5086 part (NOT match_host).
5088 8. The placeholder for the Received: header didn't have a length field set.
5090 9. Added code to Exim itself and to exim_lock to test for a specific race
5091 condition that could lead to file corruption when using MBX delivery. The
5092 issue is with the lockfile that is created in /tmp. If this file is removed
5093 after a process has opened it but before that process has acquired a lock,
5094 there is the potential for a second process to recreate the file and also
5095 acquire a lock. This could lead to two Exim processes writing to the file
5096 at the same time. The added code performs the same test as UW imapd; it
5097 checks after acquiring the lock that its file descriptor still refers to
5098 the same named file.
5100 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5101 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5104 11. Added the smtp_active_hostname option. If used, this will typically be made
5105 to depend on the incoming interface address. Because $interface_address is
5106 not set up until the daemon has forked a reception process, error responses
5107 that can happen earlier (such as "too many connections") no longer contain
5110 12. If an expansion in a condition on a "warn" statement fails because a lookup
5111 defers, the "warn" statement is abandoned, and the next ACL statement is
5112 processed. Previously this caused the whole ACL to be aborted.
5114 13. Added the iplsearch lookup type.
5116 14. Added ident_timeout as a log selector.
5118 15. Added tls_certificate_verified as a log selector.
5120 16. Added a global option tls_require_ciphers (compare the smtp transport
5121 option of the same name). This controls incoming TLS connections.
5123 17. I finally figured out how to make tls_require_ciphers do a similar thing
5124 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5125 before starting the TLS session.
5127 18. Tabs are now shown as \t in -bP output.
5129 19. If the log selector return_path_on_delivery was set, Exim crashed when
5130 bouncing a message because it had too many Received: header lines.
5132 20. If two routers both had headers_remove settings, and the first one included
5133 a superfluous trailing colon, the final name in the first list and the
5134 first name in the second list were incorrectly joined into one item (with a
5135 colon in the middle).
5141 1. Added -C and -D options to the exinext utility, mainly to make it easier
5142 to include in the automated testing, but these could be helpful when
5143 multiple configurations are in use.
5145 2. The exinext utility was not formatting the output nicely when there was
5146 an alternate port involved in the retry record key, nor when there was a
5147 message id as well (for retries that were specific to a specific message
5148 and a specific host). It was also confused by IPv6 addresses, because of
5149 the additional colons they contain. I have fixed the IPv4 problem, and
5150 patched it up to do a reasonable job for IPv6.
5152 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5153 delivery, the log line now contains "pipelined" if PIPELINING was used.
5155 4. An SMTP transport process used to panic and die if the bind() call to set
5156 an explicit outgoing interface failed. This has been changed; it is now
5157 treated in the same way as a connect() failure.
5159 5. A reference to $sender_host_name in the part of a conditional expansion
5160 that was being skipped was still causing a DNS lookup. This no longer
5163 6. The def: expansion condition was not recognizing references to header lines
5164 that used bh_ and bheader_.
5166 7. Added the _cache feature to named lists.
5168 8. The code for checking quota_filecount in the appendfile transport was
5169 allowing one more file than it should have been.
5171 9. For compatibility with Sendmail, the command line option
5179 and sets the incoming protocol and host name (for trusted callers). The
5180 host name and its colon can be omitted when only the protocol is to be set.
5181 Note the Exim already has two private options, -pd and -ps, that refer to
5182 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5183 "s", but I don't think that's a major issue.
5185 10. A number of refactoring changes to the code, none of which should affect
5188 (a) The number of logging options was getting close to filling up the
5189 32-bit word that was used as a bit map. I have split them into two classes:
5190 those that are passed in the argument to log_write(), and those that are
5191 only ever tested independently outside of that function. These are now in
5192 separate 32-bit words, so there is plenty of room for expansion again.
5193 There is no change in the user interface or the logging behaviour.
5195 (b) When building, for example, log lines, the code previously used a
5196 macro that called string_cat() twice, in order to add two strings. This is
5197 not really sufficiently general. Furthermore, there was one instance where
5198 it was actually wrong because one of the argument was used twice, and in
5199 one call a function was used. (As it happened, calling the function twice
5200 did not affect the overall behaviour.) The macro has been replaced by a
5201 function that can join an arbitrary number of extra strings onto a growing
5204 (c) The code for expansion conditions now uses a table and a binary chop
5205 instead of a serial search (which was left over from when there were very
5206 few conditions). Also, it now recognizes conditions like "pam" even when
5207 the relevant support is not compiled in: a suitably worded error message is
5208 given if an attempt is made to use such a condition.
5210 11. Added ${time_interval:xxxxx}.
5212 12. A bug was causing one of the ddress fields not to be passed back correctly
5213 from remote delivery subprocesses. The field in question was not being
5214 subsequently used, so this caused to problems in practice.
5216 13. Added new log selectors queue_time and deliver_time.
5218 14. Might have fixed a bug in maildirsizefile handling that threw up
5219 "unexpected character" debug warnings, and recalculated the data
5220 unnecessarily. In any case, I expanded the warning message to give more
5223 15. Added the message "Restricted characters in address" to the statements in
5224 the default ACL that block characters like @ and % in local parts.
5226 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5227 Three changes have been made:
5229 (a) There was a serious bug; a negative response to MAIL caused the whole
5230 recipient domain to be cached as invalid, thereby blocking all messages
5231 to all local parts at the same domain, from all senders. This bug has
5232 been fixed. The domain is no longer cached after a negative response to
5233 MAIL if the sender used is not empty.
5235 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5238 (c) A new callout option, "use_sender" has been added for people who want
5239 the modified behaviour.
5245 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5248 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5249 indeed breaks things for older releases.
5251 3. Added additional logging to the case where there is a problem reading data
5252 from a filter that is running in a subprocess using a pipe, in order to
5253 try to track down a specific problem.
5255 4. Testing facility fudge: when running in the test harness and attempting
5256 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5257 getting "No route to host". Convert this to a timeout.
5259 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5262 6. Some OS don't have socklen_t but use size_t instead. This affects the
5263 fifth argument of getsockopt() amongst other things. This is now
5264 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5265 can be set for individual OS. I have set it for SunOS5, OSF1, and
5266 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5267 some earlier ones do not.
5269 7. Change 4.30/15 was not doing the test caselessly.
5271 8. The standard form for an IPv6 address literal was being rejected by address
5272 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5273 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5274 this, as well as the form without the "IPv6" on the front (but only when
5275 address literals are enabled, of course).
5277 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5279 10. Exim crashed if a message with an empty sender address specified by -f
5280 encountered a router with an errors_to setting. This could be provoked only
5281 by a command such as
5285 where an empty string was supplied; "<>" did not hit this bug.
5287 11. Installed PCRE release 4.5.
5289 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5290 remained set. It is now erased.
5292 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5293 times from message ids (which are base 36 rather than the normal 62).
5295 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5296 were being counted as actual protocol errors, and logged if the log
5297 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5298 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5299 and DATA following a set of rejected RCPTs do not count as protocol errors.
5300 In other words, Exim assumes they were pipelined, though this may not
5301 actually be the case. Of course, in all cases the client gets an
5302 appropriate error code.
5304 15. If a lookup fails in an ACL condition, a message about the failure may
5305 be available; it is used if testing the ACL cannot continue, because most
5306 such messages specify what the cause of the deferral is. However, some
5307 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5308 that caused an old message to be retained and used if a later statement
5309 caused a defer, replacing the real cause of the deferral.
5311 16. If an IP address had so many PTR records that the DNS lookup buffer
5312 was not large enough to hold them, Exim could crash while trying to process
5313 the truncated data. It now detects and logs this case.
5315 17. Further to 4.21/58, another change has been made: if (and only if) the
5316 first line of a message (the first header line) ends with CRLF, a bare LF
5317 in a subsequent header line has a space inserted after it, so as not to
5318 terminate the header.
5320 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5321 unnecessarily, used atoi() instead of strtol(), and didn't check the
5322 termination when getting file sizes from file names by regex.
5324 19. Completely re-implemented the support for maildirsize files, in the light
5325 of a number of problems with the previous contributed implementation
5326 (4.30/29). In particular:
5328 . If the quota is zero, the maildirsize file is maintained, but no quota is
5331 . If the maildir directory does not exist, it is created before any attempt
5332 to write a maildirsize file.
5334 . The quota value in the file is just a cache; if the quota is changed in
5335 the transport, the new value overrides.
5337 . A regular expression is available for excluding directories from the
5340 20. The autoreply transport checks the characters in options that define the
5341 message's headers; it allows continued headers, but it was checking with
5342 isspace() after an embedded newline instead of explicitly looking for a
5345 21. If all the "regular" hosts to which an address was routed had passed their
5346 expiry times, and had not reached their retry times, the address was
5347 bounced, even if fallback hosts were defined. Now Exim should go on to try
5350 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5351 equivalent code in the SMTP transport. Some hosts send humungous responses
5352 to HELO/EHLO, more than 1024 it seems.
5354 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5355 gives compiler warnings in some environments. I've now done it "properly",
5358 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5359 (because of problems with the built-in one) was declared to return uschar *
5360 instead of char *, causing compiler failure.
5362 25. Fixed a file descriptor leak when processing alias/forward files.
5364 26. Fixed a minor format string issue in dbfn.c.
5366 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5368 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5369 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5372 29. When a sender address is verified, it is cached, to save repeating the test
5373 when there is more than one recipient in a message. However, when the
5374 verification involves a callout, it is possible for different callout
5375 options to be set for different recipients. It is too complicated to keep
5376 track of this in the cache, so now Exim always runs a verification when a
5377 callout is required, relying on the callout cache for the optimization.
5378 The overhead is duplication of the address routing, but this should not be
5381 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5382 to be verified with callout=postmaster, and the main callout worked but the
5383 postmaster check failed, the verification correctly failed. However, if a
5384 subsequent RCPT command asked for sender verification *without* the
5385 postmaster check, incorrect caching caused this verification also to fail,
5388 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5389 it was not caching the DNS options (qualify_single, search_parents) that
5390 were used when the lookup failed. A subsequent lookup with different
5391 options therefore always gave the same answer, though there were cases
5392 where it should not have. (Example: a "domains = !$mx_any" option on a
5393 dnslookup router: the "domains" option is always processed without any
5394 widening, but the router might have qualify_single set.) Now Exim uses the
5395 cached value only when the same options are set.
5397 32. Added John Jetmore's "exipick" utility to the distribution.
5399 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5400 than a timeout (e.g. a certificate is required, and is not provided), an
5401 Exim server now closes the connection immediately. Previously it waited for
5402 the client to close - but if the client is SSL, it seems that they each
5403 wait for each other, leading to a delay before one of them times out.
5405 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5406 maintained 0.8.x compatibility because I don't think many are using it, and
5407 it is clearly obsolete.
5409 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5412 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5413 client certificate was expired. A simple patch fixes this, though I don't
5414 understand the full logic of why the verify callback is called multiple
5417 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5418 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5419 which causes problems with some clients (such as the Certicom SSL Plus
5420 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5421 disables the coutermeasure allowing Eudora to connect."
5423 38. Exim was not checking that a write() to a log file succeeded. This could
5424 lead to Bad Things if a log got too big, in particular if it hit a file
5425 size limit. Exim now panics and dies if it cannot write to a log file, just
5426 as it does if it cannot open a log file.
5428 39. Modified OS/Makefile-Linux so that it now contains
5430 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5432 The two -D definitions ensure that Exim is compiled with large file
5433 support, which makes it possible to handle log files that are bigger than
5436 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5437 instance) a domain was checked against a named list that involved a lookup,
5438 causing $domain_data to be set, then another domain was checked against the
5439 same list, then the first domain was re-checked, the value of $domain_data
5440 after the final check could be wrong. In particular, if the second check
5441 failed, it could be set empty. This bug probably also applied to
5444 41. The strip_trailing_dot option was not being applied to the address given
5445 with the -f command-line option.
5447 42. The code for reading a message's header from the spool was incrementing
5448 $received_count, but never initializing it. This meant that the value was
5449 incorrect (doubled) while delivering a message in the same process in which
5450 it was received. In the most common configuration of Exim, this never
5451 happens - a fresh exec is done - but it can happen when
5452 deliver_drop_privilege is set.
5454 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5455 it now includes up to 150 characters of the unexpected data in the log
5458 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5459 and building data strings. The size of both of these buffers was 10 000
5460 bytes - far larger than anybody would *ever* want, thought I. Needless to
5461 say, somebody hit the limit. I have increased the maximum line length to
5462 20 000 and the maximum data length of concatenated lines to 100 000. I have
5463 also fixed two bugs, because there was no checking on these buffers. Tsk,
5464 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5465 buffer is too small.
5467 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5468 lsearch lookups. Now it does.
5470 46. When parsing a route_list item in a manualroute router, a fixed-length
5471 buffer was used for the list of hosts. I made this 1024 bytes long,
5472 thinking that nobody would ever have a list of hosts that long. Wrong.
5473 Somebody had a whole pile of complicated expansion conditions, and the
5474 string was silently truncated, leading to an expansion error. It turns out
5475 that it is easier to change to an unlimited length (owing to other changes
5476 that have happened since this code was originally written) than to build
5477 structure for giving a limitation error. The length of the item that
5478 expands into the list of hosts is now unlimited.
5480 47. The lsearch lookup could not handle data where the length of text line was
5481 more than 4095 characters. Such lines were truncated, leading to shortened
5482 data being returned. It should now handle lines of any length.
5484 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5485 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5488 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5489 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5490 no longer happens in comment lines. A list of replacements is now placed
5491 at the head of all of the source files, except those whose only change is
5492 to replace PERL_COMMAND in the very first #! line.
5494 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5495 messages on the queue, with a bottom-up merge sort, using code contributed
5496 by Michael Haardt. This should make operations like -bp somewhat faster on
5497 large queues. It won't affect queue runners, except when queue_run_in_order
5500 51. Installed eximstats 1.31 in the distribution.
5502 52. Added support for SRV lookups to the dnslookup router.
5504 53. If an ACL referred to $message_body or $message_body_end, the value was not
5505 reset for any messages that followed in the same SMTP session.
5507 54. The store-handling optimization for building very long strings was not
5508 differentiating between the different store pools. I don't think this
5509 actually made any difference in practice, but I've tidied it.
5511 55. While running the routers to verify a sender address, $sender_address
5512 was still set to the sender address. This is wrong, because when routing to
5513 send a bounce to the sender, it would be empty. Therefore, I have changed
5514 it so that, while verifying a sender address, $sender_address is set to <>.
5515 (There is no change to what happens when verifying a recipient address.)
5517 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5518 target A or AAAA records (if not already returned) without resetting the
5519 qualify_single or search_parents options of the DNS resolver. These are
5520 inappropriate in this case because the targets of MX and SRV records must
5521 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5522 target that, when qualified, matched something in the local domain. These
5523 two options are now turned off when doing these lookups.
5525 57. It seems that at least some releases of Reiserfs (which does not have the
5526 concept of a fixed number of inodes) returns zero and not -1 for the
5527 number of available inodes. This interacted badly with check_spool_inodes,
5528 which assumed that -1 was the "no such thing" setting. What I have done is
5529 to check that the total number of inodes is greater than zero before doing
5530 the test of how many are available.
5532 58. When a "warn" ACL statement has a log_message modifier, the message is
5533 remembered, and not repeated. This is to avoid a lot of repetition when a
5534 message has many recipients that cause the same warning to be written.
5535 Howewer, Exim was preserving the list of already written lines for an
5536 entire SMTP session, which doesn't seem right. The memory is now reset if a
5537 new message is started.
5539 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5540 debugging output unless log_rewrite was also set.
5542 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5543 of (int)(handle) when we know that handle contains (void *)(-1).
5545 61. The Exim daemon panic-logs an error return when it closes the incoming
5546 connection. However "connection reset by peer" seems to be common, and
5547 isn't really an error worthy of noting specially, so that particular error
5550 62. When Exim is trying to find all the local interfaces, it used to panic and
5551 die if the ioctl to get the interface flags failed. However, it seems that
5552 on at least one OS (Solaris 9) it is possible to have an interface that is
5553 included in the list of interfaces, but for which you get a failure error
5554 for this call. This happens when the interface is not "plumbed" into a
5555 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5556 failure of the "get flags" call assumes that the interface is down.
5558 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5559 makes life easier for people who are doing arithmetic on fields extracted
5560 from dates, where you often get leading zeros that should not be
5561 interpreted as octal.
5563 64. Added qualify_domain to the redirect router, to override the global
5566 65. If a pathologically long header line contained very many addresses (the
5567 report of this problem mentioned 10 000) and each of them was rewritten,
5568 Exim could use up a very large amount of memory. (It kept on making new
5569 copies of the header line as it rewrote, and never released the old ones.)
5570 At the expense of a bit more processing, the header rewriting function has
5571 been changed so that it no longer eats memory in this way.
5573 66. The generation of the Received: header has been moved from the time that a
5574 message starts to be received, to the time that it finishes. The timestamp
5575 in the Received: header should now be very close to that of the <= log
5576 line. There are two side-effects of this change:
5578 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5579 logged header lines no longer include the local Received: line, because
5580 it has not yet been created. The same applies to a copy of the message
5581 that is returned to a non-SMTP sender when a message is rejected.
5583 (b) When a filter file is tested using -bf, no additional Received: header
5584 is added to the test message. After some thought, I decided that this
5587 This change does not affect the value of $received_for. It is still set
5588 after address rewriting, but before local_scan() is called.
5590 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5592 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5593 gave an unhelpful panic error message, and a defer error. I have managed to
5594 change this behaviour so that it now rejects any supplied certificate,
5595 which seems right, as the list of acceptable certificates is empty.
5597 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5598 gave an unhelpful defer error. I have not managed to make this reject any
5599 supplied certificates, but the error message it gives is "no certificate
5600 supplied", which is not helpful.
5602 70. exigrep's output now also includes lines that are not associated with any
5603 message, but which match the given pattern. Implemented by a patch from
5604 Martin Sluka, which also tidied up the Perl a bit.
5606 71. Recipient callout verification, like sender verification, was using <> in
5607 the MAIL FROM command. This isn't really the right thing, since the actual
5608 sender may affect whether the remote host accepts the recipient or not. I
5609 have changed it to use the actual sender in the callout; this means that
5610 the cache record is now keyed on a recipient/sender pair, not just the
5611 recipient address. There doesn't seem to be a real danger of callout loops,
5612 since a callout by the remote host to check the sender would use <>.
5613 [SEE ABOVE: changed after hitting problems.]
5615 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5616 temporary errors. However, in the case of such a code being given after
5617 the end of a data transmission (i.e. after ".") Exim was failing to write
5618 a retry record for the message. (Yes, there was some broken host that was
5619 actually sending 8xx at this point.)
5621 73. An unknown lookup type in a host list could cause Exim to panic-die when
5622 the list was checked. (An example that provoked this was putting <; in the
5623 middle of a list instead of at the start.) If this happened during a DATA
5624 ACL check, a -D file could be left lying around. This kind of configuration
5625 error no longer causes Exim to die; instead it causes a defer errror. The
5626 incident is still logged to the main and panic logs.
5628 74. Buglet left over from Exim 3 conversion. The message "too many messages
5629 in one connection" was written to the rejectlog but not the mainlog, except
5630 when address rewriting (yes!) was being logged.
5632 75. Added write_rejectlog option.
5634 76. When a system filter was run not as root (that is, when system_filter_user
5635 was set), the values of the $n variables were not being returned to the
5636 main process; thus, they were not subsequently available in the $sn
5639 77. Added +return_path_on_delivery log selector.
5641 78. A connection timeout was being treated differently from recipients deferred
5642 when testing hosts_max_try with a message that was older than the host's
5643 retry timeout. (The host should not be counted, thus allowing all hosts to
5644 be tried at least once before bouncing.) This may have been the cause of an
5645 occasionally reported bug whereby a message would remain on the queue
5646 longer than the retry timeout, but would be bounced if a delivery was
5647 forced. I say "may" because I never totally pinned down the problem;
5648 setting up timeout/retry tests is difficult. See also the next item.
5650 79. The ultimate address timeout was not being applied to errors that involved
5651 a combination of host plus message (for example, a timeout on a MAIL
5652 command). When an address resolved to a number of possible hosts, and they
5653 were not all tried for each delivery (e.g. because of hosts_max_try), a
5654 message could remain on the queue longer than the retry timeout.
5656 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5659 81. Fixed an obscure SMTP outgoing bug which required at least the following
5660 conditions: (a) there was another message waiting for the same server;
5661 (b) the server returned 5xx to all RCPT commands in the first message so
5662 that the message was not completed; (c) the server dropped the connection
5663 or gave a negative response to the RSET that Exim sends to abort the
5664 transaction. The observed case was a dropped connection after DATA that had
5665 been sent in pipelining mode. That is, the server had advertised PIPELINING
5666 but was not implementing it correctly. The effect of the bug was incorrect
5667 behaviour, such as trying another host, and this could lead to a crash.
5673 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5674 and daemon.c were passed as pointers to ints; they should have been
5675 pointers to socklen_t variables (which are typically unsigned ints).
5677 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5680 3. Fixed a really odd bug that affected only the testing scheme; patching a
5681 certain fixed string in the binary changed the value of another string that
5682 happened to be identical to the end of the original first string.
5684 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5685 name", it returns that address as the IP address. On some operating
5686 systems (e.g. Solaris), it also passes back the IP address string as the
5687 "host name". However, on others (e.g. Linux), it passes back an empty
5688 string. Exim wasn't checking for this, and was changing the host name to an
5689 empty string, assuming it had been canonicized.
5691 5. Although rare, it is permitted to have more than one PTR record for a given
5692 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5693 all the names associated with an address, because they do in Solaris.
5694 However, it seems that they do not in Linux for data that comes from the
5695 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5696 I found this out when I moved to a new Linux workstation and tried to run
5697 the Exim test suite.
5699 To get round this problem I have changed the code so that it now does its
5700 own call to the DNS to look up PTR records when searching for a host name.
5701 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5702 addresses that are only in /etc/hosts are still found.
5704 This behaviour is, however, controlled by an option called host_lookup_
5705 order, which defaults to "bydns:byaddr". If people want to use the other
5706 order, or indeed, just use one or the other means of lookup, they can
5707 specify it in this variable.
5709 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5710 some operating systems, this comes back from gethostbyaddr() as an empty
5711 string, and this is what Exim used to test for. However, it seems that in
5712 other systems, "." is yielded. Exim now tests for this case too.
5714 7. The values of check_spool_space and check_log_space are now held internally
5715 as a number of kilobytes instead of an absolute number of bytes. If a
5716 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5717 kilobyte. This means that much larger values can be stored.
5719 8. Exim monitor: an attempt to get the action menu when not actually pointing
5720 at a message produces an empty menu entitled "No message selected". This
5721 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5722 no entries in it ("Shell widget menu has zero width and/or height"). So I
5723 have added a single, blank menu entry in this case.
5725 9. Added ${quote_local_part.
5727 10. MIME decoding is now applied to the contents of Subject: header lines when
5730 11. Now that a reference to $sender_host_address automatically causes a reverse
5731 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5732 host lookup before query-style lookups in lists that might use this
5733 variable. This has therefore been abolished, and the "net-" prefix is no
5734 longer necessary for query-style lookups.
5736 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5737 have been a typo for LFLAGS, so it has been changed.
5739 13. The install script calls Exim with "-C /dev/null" in order to find the
5740 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5741 to be output. Howeve, since Exim outputs its version number before the
5742 error, it didn't break the script. It just looked ugly. I fixed this by
5743 always allowing "-C /dev/null" if the caller is root.
5745 14. Ignore overlarge ACL variable number when reading spool file - insurance
5746 against a later release with more variables having written the file.
5748 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5749 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5750 this, as well as the form without the "IPv6" on the front.
5752 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5753 OS/Makefile-Darwin file.
5755 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5756 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5758 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5759 variable. However, some OS use ints for inodes. I've added cast to long int
5760 to get rid of the compiler warning.
5762 19. I had forgotten to lock out "/../" in configuration file names when
5763 ALT_CONFIG_PREFIX was set.
5765 20. Routers used for verification do not need to specify transports. However,
5766 if such a router generated a host list, and callout was configured, Exim
5767 crashed, because it could not find a port number from the (non-existent)
5768 transport. It now assumes port 25 in this circumstance.
5770 21. Added the -t option to exigrep.
5772 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5773 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5774 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5775 them is set, LOOKUP_LSEARCH is forced.
5777 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5779 24. Added sender and host information to the "rejected by local_scan()" log
5780 line; previously there was no indication of these.
5782 25. Added .include_if_exists.
5784 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5785 Linux. It turns out that not all file systems support this. Apparently some
5786 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5787 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5788 not supported on the file descriptor, is now ignored when Exim is trying to
5789 sync a directory. This applies only to Linux.
5791 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5793 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5794 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5797 29. Added support for maildirsize files from supplied patch (modified a bit).
5799 30. The use of :fail: followed by an empty string could lead Exim to respond to
5800 sender verification failures with (e.g.):
5802 550 Verification failed for <xxx>
5803 550 Sender verify failed
5805 where the first response line was missing the '-' that indicates it is not
5806 the final line of the response.
5808 31. The loop for finding the name of the user that called Exim had a hardwired
5809 limit of 10; it now uses the value of finduser_retries, which is used for
5810 all other user lookups.
5812 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5815 33. Exim was neglecting to zero errno before one call of strtol() when
5816 expanding a string and expecting an integer value. On some systems this
5817 resulted in spurious "integer overflow" errors. Also, it was casting the
5818 result into an int without checking.
5820 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5821 did not work. The code looks as if it has *never* worked, though it appears
5822 to have been documented since at least releast 1.62. I have made it work.
5824 35. The "timeout_DNS" error in retry rules, also documented since at least
5825 1.62, also never worked. As it isn't clear exactly what this means, and
5826 clearly it isn't a major issue, I have abolished the feature by treating it
5827 as "timeout", and writing a warning to the main and panic logs.
5829 36. The display of retry rules for -brt wasn't always showing the error code
5832 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5833 timeout_connect_A, timeout_connect_MX.
5835 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5836 to the empty sender.
5838 39. The daemon was not analysing the content of -oX till after it had closed
5839 stderr and disconnected from the controlling terminal. This meant that any
5840 syntax errors were only noted on the panic log, and the return code from
5841 the command was 0. By re-arranging the code a little, I've made the
5842 decoding happen first, so such errors now appear on stderr, and the return
5843 code is 1. However, the actual setting up of the sockets still happens in
5844 the disconnected process, so errors there are still only recorded on the
5847 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5848 connections (as happens on some IP stacks) was logged at start up time as
5849 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5850 it from "IPv6 and IPv4", which means that two separate sockets are being
5853 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5854 says whether AF_INET or AF_INET6 was passed as an argument.
5856 42. Exiwhat output was messed up when time zones were included in log
5859 43. Exiwhat now gives more information about the daemon's listening ports,
5860 and whether -tls-on-connect was used.
5862 44. The "port" option of the smtp transport is now expanded.
5864 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5865 silently ignored. Now an error message is written to the main and panic
5868 46. There's a new ACL modifier called "logwrite" which writes to a log file
5869 as soon as it is encountered.
5871 47. Added $local_user_uid and $local_user_gid at routing time.
5873 48. Exim crashed when trying to verify a sender address that was being
5876 49. Exim was recognizing only a space character after ".include". It now also
5877 recognizes a tab character.
5879 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5880 extracting the relevant information from the specification. The man page no
5881 longer contains scrambled data for the -d option, and I've added a section
5882 at the front about calling Exim under different names.
5884 51. Added "extra_headers" argument to the "mail" command in filter files.
5886 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5889 53. Installed eximstats 1.29.
5891 54. Added transport_filter_timeout as a generic transport option.
5893 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5894 Cc: header lines. This was required by RFC 822, but it not required by RFC
5897 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5898 incoming messages that did not have them. Now it does so only if the
5899 message originates locally, that is, if there is no associated remote host
5900 address. When Resent- header lines are present, this applies to the Resent-
5901 lines rather than the non-Resent- lines.
5903 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5904 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5906 58. Messages for configuration errors now include the name of the main
5907 configuration file - useful now that there may be more than one file in a
5908 list (.included file names were always shown).
5910 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5911 for those rare installations that do not start the daemon as root or run it
5912 setuid root. I've cut out the call to initgroups() if the daemon is not
5915 60. The Exim user and group can now be bound into the binary as text strings
5916 that are looked up at the start of Exim's processing.
5918 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5920 62. Added $mailstore_basename variable.
5922 63. Installed patch to sieve.c from Michael Haardt.
5924 64. When Exim failed to open the panic log after failing to open the main log,
5925 the original message it was trying to log was written to stderr and debug
5926 output, but if they were not available (the usual case in production), it
5927 was lost. Now it is written to syslog before the two lines that record the
5928 failures to open the logs.
5930 65. Users' Exim filters run in subprocesses under the user's uid. It is
5931 possible for a "deliver" command or an alias in a "personal" command to
5932 provoke an address rewrite. If logging of address rewriting is configured,
5933 this fails because the process is not running as root or exim. There may be
5934 a better way of dealing with this, but for the moment (because 4.30 needs
5935 to be released), I have disabled address rewrite logging when running a
5936 filter in a non-root, non-exim process.
5942 1. The buildconfig auxiliary program wasn't quoting the value set for
5943 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5944 not defined. This bug was masked in 4.22 by the effect that was fixed in
5947 2. Some messages that were rejected after a message id was allocated were
5948 shown as "incomplete" by exigrep. It no longer does this for messages that
5949 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5951 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5952 have allow_domain_literals set, the ID did not get logged in the <= line.
5953 Domain literals are now always recognized in Message-ID: header lines.
5955 4. The first argument for a ${extract expansion item is the key name or field
5956 number. Leading and trailing spaces in this item were not being ignored,
5957 causing some misleading effects.
5959 5. When deliver_drop_privilege was set, single queue runner processes started
5960 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5961 same command in the process it spins off) were not dropping privilege.
5963 6. When the daemon running as "exim" started a queue runner, it always
5964 re-executed Exim in the spun-off process. This is a waste of effort when
5965 deliver_drop_privilege is set. The new process now just calls the
5966 queue-runner function directly.
5972 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5975 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5976 ignored. Though the use of -oP was forcing the writing of a pid file, it
5977 was always written to the default place.
5979 3. If the message "no IP address found for host xxxx" is generated during
5980 incoming verification, it is now followed by identification of the incoming
5981 connection (so you can more easily find what provoked it).
5983 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5985 5. Added some features to "harden" Exim a bit more against certain attacks:
5987 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5988 be put in Local/Makefile. This is like the never_users runtime option,
5989 but it cannot be overridden. The default setting is "root".
5991 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5992 prefix string with which any file named in a -C command line option
5995 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5996 is retained for -C and -D only if the caller of Exim is root. Without
5997 it, the exim user may also use -C and -D and retain privilege.
5999 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6000 command line option is disabled.
6002 6. Macro names set by the -D option must start with an upper case letter, just
6003 like macro names defined in the configuration file.
6005 7. Added "dereference=" facility to LDAP.
6007 8. Two instances of the typo "uknown" in the source files are fixed.
6009 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6010 the Configure-Makefile script screwed up while processing it.
6012 10. Incorporated PCRE 4.4.
6014 11. The SMTP synchronization check was not operating right at the start of an
6015 SMTP session. For example, it could not catch a HELO sent before the client
6016 waited for the greeting. There is now a check for outstanding input at the
6017 point when the greeting is written. Because of the duplex, asynchronous
6018 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6019 way, but not yet received, when the check is performed.
6021 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6022 on TCP/IP sockets, because this apparently causes some broken clients to
6025 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6026 unchanged) from the Cygwin maintainer.
6028 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6029 was supported instead of testing for mbx. Effectively a typo.
6031 15. The spa authenticator server code was not checking that the input it
6032 received was valid base64.
6034 16. The debug output line for the "set" modifier in ACLs was not showing the
6035 name of the variable that was being set.
6037 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6039 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6040 lookup on its own. Something else was needed to trigger the lookup. For
6041 example, a match in host_lookup or the need for a host name in a host list.
6042 Now, if $sender_host_name is referenced and the host name has not yet been
6043 looked up, a lookup is performed. If the lookup fails, the variable remains
6044 empty, and $host_lookup_failed is set to "1".
6046 19. Added "eqi" as a case-independent comparison operator.
6048 20. The saslauthd authentication condition could segfault if neither service
6049 nor realm was specified.
6051 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6052 error message that was logged was misleading, and incoming SMTP
6053 connections were dropped. The message is now more accurate, and temporary
6054 errors are given to SMTP connections.
6056 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6057 (except RSET) with a 421 error, until QUIT is received. However, it was
6058 failing to send a response to QUIT.
6060 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6061 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6062 ACL is now run after the helo_try_verify_hosts code.
6064 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6065 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6066 case-independent, so other case variants are also recognized). Apparently
6067 some systems use these upper case variants.
6069 25. If more than two messages were waiting for the same host, and a transport
6070 filter was specified for the transport, Exim sent two messages over the
6071 same TCP/IP connection, and then failed with "socket operation on non-
6072 socket" when it tried to send the third.
6074 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6076 27. The extern definition of crypt16() in expand.c was not being excluded when
6077 the OS had its own crypt16() function.
6079 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6080 as a preferred synonym for return_size_limit, both as an option and as an
6083 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6085 30. Changed the default configuration ACL to relax the local part checking rule
6086 for addresses that are not in any local domains. For these addresses,
6087 slashes and pipe symbols are allowed within local parts, but the sequence
6088 /../ is explicitly forbidden.
6090 31. SPA server authentication was not clearing the challenge buffer before
6093 32. log_message in a "warn" ACL statement was writing to the reject log as
6094 well as to the main log, which contradicts the documentation and doesn't
6095 seem right (because no rejection is happening). So I have stopped it.
6097 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6098 However, I am unable to do any testing of this.
6100 34. Fixed an infelicity in the appendfile transport. When checking directories
6101 for a mailbox, to see if any needed to be created, it was accidentally
6102 using path names with one or more superfluous leading slashes; tracing
6103 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6105 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6106 contents are added to the log line that is written for every discarded
6107 recipient. (Previously a log_message setting was ignored.)
6109 36. The ${quote: operator now quotes the string if it is empty.
6111 37. The install script runs exim in order to find its version number. If for
6112 some reason other than non-existence or emptiness, which it checks, it
6113 could not run './exim', it was installing it with an empty version number,
6114 i.e. as "exim-". This error state is now caught, and the installation is
6117 38. An argument was missing from the function that creates an error message
6118 when Exim fails to connect to the socket for saslauthd authentication.
6119 This could cause Exim to crash, or give a corrupted message.
6121 39. Added isip, isip4, and isip6 to ${if conditions.
6123 40. The ACL variables $acl_xx are now saved with the message, and can be
6124 accessed later in routers, transports, and filters.
6126 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6127 strings in the file are not string-expanded.
6129 42. If a MAIL command specified a SIZE value that was too large to fit into an
6130 int variable, the check against message_size_limit failed. Such values are
6131 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6132 day this will have to be increased, but I don't think I want to be around
6133 when emails are that large.
6140 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6141 iconv() is not standard in FreeBSD.
6143 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6144 IPv6 enabled. The observed symptom was a segmentation fault on return from
6145 the function os_common_find_running_interfaces() in src/os.c.
6147 3. In the check_special_case() function in daemon.c I had used "errno" as an
6148 argument name, which causes warnings on some systems. This was basically a
6149 typo, since it was named "eno" in the comments!
6151 4. The code that waits for the clock to tick (at a resolution of some fraction
6152 of a second) so as to ensure message-id uniqueness was always waiting for
6153 at least one whole tick, when it could have waited for less. [This is
6154 almost certainly not relevant at current processor speeds, where it is
6155 unlikely to ever wait at all. But we try to future-proof.]
6157 5. The function that sleeps for a time interval that includes fractions of a
6158 second contained a race. It did not block SIGALRM between setting the
6159 timer, and suspending (a couple of lines later). If the interval was short
6160 and the sigsuspend() was delayed until after it had expired, the suspension
6161 never ended. On busy systems this could lead to processes getting stuck for
6164 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6165 process, before it forks any delivery processes. The open lookup caching
6166 mechanism meant that the open file or database connection was passed into
6167 the delivery process. The problem was that delivery processes always tidy
6168 up cached lookup data. This could cause a problem for the next delivery
6169 process started by the queue runner, because the external queue runner
6170 process does not know about the closure. So the next delivery process
6171 still has data in the lookup cache. In the case of a file lookup, there was
6172 no problem because closing a file descriptor in a subprocess doesn't affect
6173 the parent. However, if the lookup was caching a connection to a database,
6174 the connection was closed, and the second delivery process was likely to
6175 see errors such as "PGSQL: query failed: server closed the connection
6176 unexpectedly". The problem has been fixed by closing all cached lookups
6177 in a queue runner before running a delivery process.
6179 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6180 seem to have the "const" qualifier which it has on other OS. I've
6183 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6184 *of the same type* (client or server) with the same public name that an
6185 error should be diagnosed.
6187 9. When Exim looked up a host name for an IP address, but failed to find the
6188 original IP address when looking up the host name (a safety check), it
6189 output the message "<ip address> does not match any IP for NULL", which was
6190 confusing, to say the least. The bug was that the host name should have
6191 appeared instead of "NULL".
6193 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6194 user that is built into the binary, and the -C or -D options is used, root
6195 privilege is dropped before the configuration file is read. In addition,
6196 logging is switched to stderr instead of the normal log files. If the
6197 configuration then re-defines the Exim user, the unprivileged environment
6198 is probably not what is expected, so Exim logs a panic warning message (but
6201 However, if deliver_drop_privilege is set, the unprivileged state may well
6202 be exactly what is intended, so the warning has been cut out in that case,
6203 and Exim is allowed to try to write to its normal log files.
6209 1. smtp_return_error_details was not giving details for temporary sender
6210 or receiver verification errors.
6212 2. Diagnose a configuration error if two authenticators have the same public
6215 3. Exim used not to create the message log file for a message until the first
6216 delivery attempt. This could be confusing when incoming messages were held
6217 for policy or load reasons. The message log file is now created at the time
6218 the message is received, and an initial "Received" line is written to it.
6220 4. The automatically generated man page for command line options had a minor
6221 bug that caused no ill effects; however, a more serious problem was that
6222 the procedure for building the man page automatically didn't always
6223 operate. Consequently, release 4.20 contains an out-of-date version. This
6224 shouldn't happen again.
6226 5. When building Exim with embedded Perl support, the script that builds the
6227 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6228 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6230 6. The freeze_tell option was not being used for messages that were frozen on
6231 arrival, either by an ACL or by local_scan().
6233 7. Added the smtp_incomplete_transaction log selector.
6235 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6236 was accepting AUTH without a new EHLO.
6238 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6239 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6242 10. Logging of TCP/IP connections (when configured) now happens in the main
6243 daemon process instead of the child process, so that the TCP/IP connection
6244 count is more accurate (but it can never be perfect).
6246 11. The use of "drop" in a nested ACL was not being handled correctly in the
6247 outer ACL. Now, if condition failure induced by the nested "drop" causes
6248 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6249 or "require"), the connection is dropped.
6251 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6252 that yield "discard" can now be used with an "accept" or a "discard" verb,
6253 but an error is generated for any others (because I can't see a useful way
6254 to define what should happen).
6256 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6257 are now processed in the same way as lines in the Exim configuration file.
6258 In particular, continuation lines are supported.
6260 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6262 15. Added -ti meaning -t -i.
6264 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6265 domains, and warn by logging if others are found.
6267 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6268 returned from GIFCONF when Exim is trying to find the list of interfaces on
6269 a host. The code in os.c has been modified to copy each ifreq to an aligned
6270 structure in all cases.
6272 Also, in some cases, the returned ifreq's were being copied to a 'struct
6273 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6274 means the last couple of bytes of an IPv6 address could be chopped if the
6275 ifreq contained only a normal sockaddr (14 bytes storage).
6277 18. Named domain lists were not supported in the hosts_treat_as_local option.
6278 An entry such as +xxxx was not recognized, and was treated as a literal
6281 19. Ensure that header lines added by a DATA ACL are included in the reject log
6282 if the ACL subsequently rejects the message.
6284 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6285 MD5 (which is deprecated).
6287 21. When testing a filter file using -bf, Exim was writing a message when it
6288 took the sender from a "From " line in the message, but it was not doing so
6289 when it took $return_path from a Return-Path: header line. It now does.
6291 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6292 with a valid header line field name (a series of printing characters
6293 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6295 23. Changed "disc" in the source to "disk" to conform to the documentation and
6296 the book and for uniformity.
6298 24. Ignore Sendmail's -Ooption=value command line item.
6300 25. When execve() failed while trying to run a command in a pipe transport,
6301 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6302 could be confused with a return value of 69 from the command itself. This
6303 has been changed to 127, the value the shell returns if it is asked to run
6304 a non-existent command. The wording for the related log line suggests a
6305 non-existent command as the problem.
6307 26. If received_header_text expands to an empty string, do not add a Received:
6308 header line to the message. (Well, it adds a token one on the spool, but
6309 marks it "old" so that it doesn't get used or transmitted.)
6311 27. Installed eximstats 1.28 (addition of -nt option).
6313 28. There was no check for failure on the call to getsockname() in the daemon
6314 code. This can fail if there is a shortage of resources on the system, with
6315 ENOMEM, for example. A temporary error is now given on failure.
6317 29. Contrary to the C standard, it seems that in some environments, the
6318 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6319 program. Exim now does this explicitly; it affects the formatting of
6320 timestamps using strftime().
6322 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6323 complaints. I've now initialized all the variables, to avoid this.
6325 32. Header lines added by a system filter were not being "seen" during
6326 transport-time rewrites.
6328 33. The info_callback() function passed to OpenSSL is set up with type void
6329 (*)(SSL *, int, int), as described somewhere. However, when calling the
6330 function (actually a macro) that sets it up, the type void(*)() is
6331 expected. I've put in a cast to prevent warnings from picky compilers.
6333 34. If a DNS black list lookup found a CNAME record, but there were no A
6334 records associated with the domain it pointed at, Exim crashed.
6336 35. If a DNS black list lookup returned more than one A record, Exim ignored
6337 all but the first. It now scans all returned addresses if a particular IP
6338 value is being sought. In this situation, the contents of the
6339 $dnslist_value variable are a list of all the addresses, separated by a
6342 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6343 to accept a host name and all its aliases if the forward lookup for any of
6344 them yielded the IP address of the incoming connection. Now it accepts only
6345 those names whose forward lookup yields the correct IP address. Any other
6346 names are discarded. This closes a loophole whereby a rogue DNS
6347 administrator could create reverse DNS records to break through a
6348 wildcarded host restriction in an ACL.
6350 37. If a user filter or a system filter that ran in a subprocess used any of
6351 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6352 the wrong values were passed to the pipe command ($thisaddress had the
6353 value of $0, $0 had the value of $1, etc). This bug was introduced by
6354 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6356 38. Improved the line breaking for long SMTP error messages from ACLs.
6357 Previously, if there was no break point between 40 and 75 characters, Exim
6358 left the rest of the message alone. Two changes have been made: (a) I've
6359 reduced the minimum length to 35 characters; (b) if it can't find a break
6360 point between 35 and 75 characters, it looks ahead and uses the first one
6361 that it finds. This may give the occasional overlong line, but at least the
6362 remaining text gets split now.
6364 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6365 file descriptors might be low, and that setting 1000 would always raise it.
6366 It turns out that in some environments, the limit is already over 1000 and
6367 that lowering it causes trouble. So now Exim takes care not to decrease it.
6369 40. When delivering a message, the value of $return_path is set to $sender_
6370 address at the start of routing (routers may change the value). By an
6371 oversight, this default was not being set up when an address was tested by
6372 -bt or -bv, which affected the outcome if any router or filter referred to
6375 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6376 log only when "message" or "log_message" are set. However, if one of the
6377 conditions was an address verification, or a call to a nested ACL, the
6378 messages generated by the underlying test were being passed through. This
6379 no longer happens. The underlying message is available in $acl_verify_
6380 message for both "message" and "log_message" expansions, so it can be
6381 passed through if needed.
6383 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6384 new expansion $bh_ to give the encoded byte string without charset
6385 translation. Translation happens only if iconv() is available; HAVE_ICONV
6386 indicates this at build time. HEADERS_CHARSET gives the charset to
6387 translate to; headers_charset can change it in the configuration, and
6388 "headers charset" can change it in an individual filter file.
6390 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6391 that creates RFC 2047 encoded "words" labels them as that charset instead
6392 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6393 expansion operator; (ii) when Exim creates a From: line for a local
6394 message; (iii) when a header line is rewritten to include a "phrase" part.
6396 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6397 buggy, causing it to skip the first lines of messages whose message ID
6398 ended in 'D'. This would not have bitten before Exim release 4.14, because
6399 message IDs were unlikely to end in 'D' before then. The effect was to have
6400 incorrect size information for certain domains.
6402 45. #include "config.h" was missing at the start of the crypt16.c module. This
6403 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6406 46. If there was a timeout during a "random" callout check, Exim treated it as
6407 a failure of the random address, and carried on sending RSET and the real
6408 address. If the delay was just some slowness somewhere, the response to the
6409 original RCPT would be taken as a response to RSET and so on, causing
6410 mayhem of various kinds.
6412 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6413 when I implemented it. It didn't allow for the fact that some option values
6414 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6415 the right test for positive values.
6417 48. Domain names in DNS records are case-independent. Exim always looks them up
6418 in lower case. Some resolvers return domain names in exactly the case they
6419 appear in the zone file, that is, they may contain uppercase letters. Not
6420 all resolvers do this - some return always lower case. Exim was treating a
6421 change of case by a resolver as a change of domain, similar to a widening
6422 of a domain abbreviation. This triggered its re-routing code and so it was
6423 trying to route what was effectively the same domain again. This normally
6424 caused routing to fail (because the router wouldn't handle the domain
6425 twice). Now Exim checks for this case specially, and just changes the
6426 casing of the domain that it ultimately uses when it transmits the message
6429 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6432 50. If a filter generated a file delivery with a non-absolute name (possible if
6433 no home directory exists for the router), the forbid_file option was not
6436 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6437 the existing equality matching.
6439 52. Exim was using ints instead of ino_t variables in some places where it was
6440 dealing with inode numbers.
6442 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6443 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6444 called TMPDIR, and if it finds it is different, it changes its value.
6446 54. The smtp_printf() function is now made available to local_scan() so
6447 additional output lines can be written before returning. There is also an
6448 smtp_fflush() function to enable the detection of a dropped connection.
6449 The variables smtp_input and smtp_batched_input are exported to
6452 55. Changed the default runtime configuration: the message "Unknown user"
6453 has been removed from the ACL, and instead placed on the localuser router,
6454 using the cannot_route_message feature. This means that any verification
6455 failures that generate their own messages won't get overridden. Similarly,
6456 the "Unrouteable address" message that was in the ACL for unverifiable
6457 relay addresses has also been removed.
6459 56. Added hosts_avoid_esmtp to the smtp transport.
6461 57. The exicyclog script was not checking for the esoteric option
6462 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6463 will work only if exicyclog is run under the appropriate euid.
6465 58. Following a discussion on the list, the rules by which Exim recognises line
6466 endings on incoming messages have been changed. The -dropcr and drop_cr
6467 options are now no-ops, retained only for backwards compatibility. The
6468 following line terminators are recognized: LF CRLF CR. However, special
6469 processing applies to CR:
6471 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6472 nor a local message in the state where . is a terminator.
6474 (ii) If a bare CR is encountered in a header line, an extra space is added
6475 after the line terminator so as not to end the header. The reasoning
6476 behind this is that bare CRs in header lines are most likely either
6477 to be mistakes, or people trying to play silly games.
6479 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6480 was being incorrectly given as 18 bytes larger than it should have been.
6481 This is a VOB (very old bug).
6483 60. This may never have affected anything current, but just in case it has:
6484 When the local host is found other than at the start of a list of hosts,
6485 the local host, those with the same MX, and any that follow, are discarded.
6486 When the list in question was part of a longer list of hosts, the following
6487 hosts (not currently being processed) were also being discarded. This no
6488 longer happens. I'm not sure if this situation could ever has previously
6491 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6494 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6495 header. This is something that is recommended in a new Internet Draft, and
6496 is something that is documented as being done by Sendmail. There are two
6497 possible values. For messages generated by the autoreply transport, Exim
6500 Auto-Submitted: auto-replied
6502 whereas for all other generated messages (e.g. bounces) it adds
6504 Auto-Submitted: auto-generated
6506 63. The "personal" condition in filters now includes a test for the
6507 Auto-Submitted: header. If it contains the string "auto-" the message it
6508 not considered personal.
6510 64. Added rcpt_include_affixes as a generic transport option.
6512 65. Added queue_only_override (default true).
6514 66. Added the syslog_duplication option.
6516 67. If what should have been the first header line of a message consisted of
6517 a space followed by a colon, Exim was mis-interpreting it as a header line.
6518 It isn't of course - it is syntactically invalid and should therefore be
6519 treated as the start of the message body. The misbehaviour could have
6520 caused a number of strange effects, including loss of data in subsequent
6521 header lines, and spool format errors.
6523 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6524 client host had authenticated. This control can now be exercised by an ACL
6525 for more flexibility.
6527 69. By default, callouts do not happen when testing with -bh. There is now a
6528 variant, -bhc, which does actually run the callout code, including
6529 consulting and updating the callout cache.
6531 70. Added support for saslauthd authentication, courtesy of Alexander
6534 71. If statvfs() failed on the spool or log directories while checking their
6535 size for availability, Exim confusingly gave the error "space shortage".
6536 Furthermore, in debugging mode it crashed with a floating point exception.
6537 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6538 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6539 serious problem, Exim now writes to the main and panic logs when this
6540 happens, with details of the failure. It then refuses to accept the
6541 incoming message, giving the message "spool directory problem" or "log
6542 directory problem" with a 421 code for SMTP messages.
6544 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6545 0, 1, and 2 exist, because some OS complain for execs without them (see
6546 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6547 descriptors. However, the code omitted to check that the open succeeded,
6548 causing mysterious errors if for some reason the permissions on /dev/null
6549 got screwed. Now Exim writes a message to the main and panic logs, and
6550 bombs out if it can't open /dev/null.
6552 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6553 interact so that it is all more flexible. It is supposed to remain
6554 backwards compatible. Also added extra_local_interfaces.
6556 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6557 to bomb out with an assertion failure - to the client this appears as a
6558 connection drop. This problem occurs in the part of the code that was taken
6559 from the Samba project. Fortunately, the assertion is in a very simple
6560 function, so I have fixed this by reproducing the function inline in the
6561 one place where it is called, and arranging for authentication to fail
6562 instead of killing the process with assert().
6564 75. The SPA client code was not working when the server requested OEM rather
6565 than Unicode encoding.
6567 76. Added code to make require_files with a specific uid setting more usable in
6568 the case where statting the file as root fails - usually a non-root-mounted
6569 NFS file system. When this happens and the failure is EACCES, Exim now
6570 forks a subprocess and does the per-uid checking as the relevant uid.
6572 77. Added process_log_path.
6574 78. If log_file_path was not explicitly set, a setting of check_log_space or
6575 check_log_inodes was ignored.
6577 79. If a space check for the spool or log partitions fails, the incident is now
6578 logged. Of course, in the latter case the data may get lost...
6580 80. Added the %p formatting code to string_format() so that it can be used to
6581 print addresses in debug_print(). Adjusted all the address printing in the
6582 debugging in store.c to use %p rather than %d.
6584 81. There was a concern that a line of code in smtp_in.c could overflow a
6585 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6586 initially expressed, the concern was not well-founded, because trailing
6587 spaces are removed early. However, if the trailing spaces were followed by
6588 a NULL, they did not get removed, so the overflow was possible. Two fixes
6591 (a) I re-wrote the offending code in a cleaner fashion.
6592 (b) If an incoming SMTP command contains a NULL character, it is rejected
6595 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6596 runs initgroups(), so that if the Exim user is in any additional groups,
6597 they will be used during message reception.
6603 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob