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 Buf 1850: support Radius libraries that return REJECT_RC
79 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
80 and 3.4.4 - once the server is enabled to respond to an OCSP request
81 it does even when not requested, resulting in a stapling non-aware
82 client dropping the TLS connection.
84 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
85 support variable-length bit vectors. No functional change.
87 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
88 The I= interface field on outgoing lines is now after the H= remote
89 host field, same as incoming lines. There is a separate
90 outgoing_interface log selector which allows you to disable the
93 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
94 If not running log_selector +smtp_connection the mainlog would be held
95 open indefinitely after a "too many connections" event, including to a
96 deleted file after a log rotate. Leave the per net connection logging
97 leaving it open for efficiency as that will be quickly detected by the
98 check on the next write.
100 HS/01 Bug 1671: Fix post transport crash.
101 Processing the wait-<transport> messages could crash the delivery
102 process if the message IDs didn't exist for some reason. When
103 using 'split_spool_directory=yes' the construction of the spool
104 file name failed already, exposing the same netto behaviour.
106 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
107 mime_regex ACL conditions.
109 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
110 to DSN fail messages (bounces): remote IP, remote greeting, remote response
111 to HELO, local diagnostic string.
113 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
114 log line to debug. Even when configured with a tls authenticator many
115 client connections are expected to not authenticate in this way, so
116 an authenticate fail is not an error.
118 HS/02 Add the Exim version string to the process info. This way exiwhat
119 gives some more detail about the running daemon.
121 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
122 matter for fast-change records such as DNSBLs.
124 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
125 retry record, even if constant. There may be multiple transports with
126 different interface settings and the retry behaviour needs to be kept
129 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
131 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
133 JH/10 Bug 840: fix log_defer_output option of pipe transport
135 JH/11 Bug 830: use same host for all RCPTS of a message, even under
136 hosts_randomize. This matters a lot when combined with mua_wrapper.
138 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
139 ${quote_pgsql:<string>} operator.
141 JH/13 Bug 1708: avoid misaligned access in cached lookup.
143 JH/14 Change header file name for freeradius-client. Relevant if compiling
144 with Radius support; from the Gentoo tree and checked under Fedora.
146 JH/15 Bug 1712: Introduce $prdr_requested flag variable
148 JH/16 Bug 1714: Permit an empty string as expansion result for transport
149 option transport_filter, meaning no filtering.
151 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
153 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
154 defaults to "*" (all hosts). The variable is now available when not built
155 with TLS, default unset, mainly to enable keeping the testuite sane.
156 If a server certificate is not supplied (via tls_certificate) an error is
157 logged, and clients will find TLS connections fail on startup. Presumably
158 they will retry in-clear.
159 Packagers of Exim are strongly encouraged to create a server certificate
160 at installation time.
162 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
163 with the $config_file variable.
165 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
166 in transport context, after the attempt, and per-recipient. The latter type
167 is per host attempted. The event data is the error message, and the errno
168 information encodes the lookup type (A vs. MX) used for the (first) host,
169 and the trailing two digits of the smtp 4xx reponse.
171 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
172 to write to mainlog (or rejectlog, paniclog) in the window between file
173 creation and permissions/ownership being changed. Particularly affects
174 installations where exicyclog is run as root, rather than exim user;
175 result is that the running daemon panics and dies.
177 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
179 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
180 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
181 "pri" and "weight". Note that the previous implicit priority given by the
182 list order is no longer honoured.
184 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
187 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
188 by defining SUPPORT_SOCKS.
190 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
191 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
192 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
193 variables are renamed to proxy_{local,external}_{address,port}.
195 JH/27 Move Internationalisation support from Experimental to mainline, enabled
196 for a build by defining SUPPORT_I18N
198 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
199 of the query string, and make ${quote_redis:} do that quoting.
201 JH/29 Move Events support from Experimental to mainline, enabled by default
202 and removable for a build by defining DISABLE_EVENT.
204 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
206 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
207 cached by the daemon.
209 JH/32 Move Redis support from Experimental to mainline, enabled for a build
210 by defining LOOKUP_REDIS. The libhiredis library is required.
212 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
213 keys are given for lookup.
215 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
216 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
217 only supported when built with TLS support. The PolarSSL SHA routines
218 are still used when the TLS library is too old for convenient support.
220 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
221 openssl_options), for security. OpenSSL forces this from version 1.1.0
222 server-side so match that on older versions.
224 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
225 allocation for $value could be released as the expansion processing
226 concluded, but leaving the global pointer active for it.
228 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
229 and to use the domains and local_parts ACL conditions.
231 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
232 incorrectly not doubled on cutthrough transmission, hence seen as a
233 body-termination at the receiving system - resulting in truncated mails.
234 Commonly the sender saw a TCP-level error, and retransmitted the message
235 via the normal store-and-forward channel. This could result in duplicates
236 received - but deduplicating mailstores were liable to retain only the
237 initial truncated version.
239 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
241 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
243 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
244 we're in there, support oversigning also; bug 1309.
246 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
248 HS/04 Add support for keep_environment and add_environment options.
250 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
251 either intentional arithmetic overflow during PRNG, or testing config-
254 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
255 delivery resulted in actual delivery. Cancel cutthrough before DATA
258 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
259 rejects a recipient: pass the reject to the originator.
261 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
262 Many were false-positives and ignorable, but it's worth fixing the
265 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
266 for the new environment-manipulation done at startup. Move the routines
267 from being local to tls.c to being global via the os.c file.
269 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
270 an extract embedded as result-arg for a map, the first arg for extract
271 is unavailable so we cannot tell if this is a numbered or keyed
272 extraction. Accept either.
278 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
281 JH/02 The smtp transport option "multi_domain" is now expanded.
283 JH/03 The smtp transport now requests PRDR by default, if the server offers
286 JH/04 Certificate name checking on server certificates, when exim is a client,
287 is now done by default. The transport option tls_verify_cert_hostnames
288 can be used to disable this per-host. The build option
289 EXPERIMENTAL_CERTNAMES is withdrawn.
291 JH/05 The value of the tls_verify_certificates smtp transport and main options
292 default to the word "system" to access the system default CA bundle.
293 For GnuTLS, only version 3.0.20 or later.
295 JH/06 Verification of the server certificate for a TLS connection is now tried
296 (but not required) by default. The verification status is now logged by
297 default, for both outbound TLS and client-certificate supplying inbound
300 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
303 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
304 Status Notification (bounce) messages are now MIME format per RFC 3464.
305 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
306 under the control of the dsn_advertise_hosts option, and routers may
307 have a dsn_lasthop option.
309 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
310 default, modifiable by a malware= option. The list separator for
311 the options can now be changed in the usual way. Bug 68.
313 JH/10 The smtp_receive_timeout main option is now expanded before use.
315 JH/11 The incoming_interface log option now also enables logging of the
316 local interface on delivery outgoing connections.
318 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
319 if the interface and destination host and port all match.
321 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
324 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
325 Patch from Andrew Lewis.
327 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
328 now supports optional time-restrictions, weighting, and priority
329 modifiers per server. Patch originally by <rommer@active.by>.
331 JH/16 The spamd_address main option now supports a mixed list of local
332 and remote servers. Remote servers can be IPv6 addresses, and
333 specify a port-range.
335 JH/17 Bug 68: The spamd_address main option now supports an optional
336 timeout value per server.
338 JH/18 Bug 1581: Router and transport options headers_add/remove can
339 now have the list separator specified.
341 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
344 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
347 JH/21 Support for the A6 type of dns record is withdrawn.
349 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
350 rather than the verbs used.
352 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
353 from 255 to 1024 chars.
355 JH/24 Verification callouts now attempt to use TLS by default.
357 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
358 are generic router options now. The defaults didn't change.
360 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
361 Original patch from Alexander Shikoff, worked over by JH.
363 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
364 files if ZCAT_COMMAND is not executable.
366 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
368 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
370 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
371 Normally benign, it bites when the pair was led to by a CNAME;
372 modern usage is to not canoicalize the domain to a CNAME target
373 (and we were inconsistent anyway for A-only vs AAAA+A).
375 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
377 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
378 when evaluating $sender_host_dnssec.
380 JH/31 Check the HELO verification lookup for DNSSEC, adding new
381 $sender_helo_dnssec variable.
383 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
385 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
387 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
389 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
390 documented as working, but never had. Support all but $spam_report.
392 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
393 added for tls authenticator.
395 HS/03 Add perl_taintmode main config option
400 TL/01 When running the test suite, the README says that variables such as
401 no_msglog_check are global and can be placed anywhere in a specific
402 test's script, however it was observed that placement needed to be near
403 the beginning for it to behave that way. Changed the runtest perl
404 script to read through the entire script once to detect and set these
405 variables, reset to the beginning of the script, and then run through
406 the script parsing/test process like normal.
408 TL/02 The BSD's have an arc4random API. One of the functions to induce
409 adding randomness was arc4random_stir(), but it has been removed in
410 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
411 function when detected.
413 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
414 cause callback expansion.
416 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
417 syntax errors in an expansion can be treated as a string instead of
418 logging or causing an error, due to the internal use of bool_lax
419 instead of bool when processing it.
421 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
422 server certificates when making smtp deliveries.
424 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
426 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
428 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
430 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
431 Merged patch from Sebastian Wiedenroth.
433 JH/05 Fix results-pipe from transport process. Several recipients, combined
434 with certificate use, exposed issues where response data items split
435 over buffer boundaries were not parsed properly. This eventually
436 resulted in duplicates being sent. This issue only became common enough
437 to notice due to the introduction of conection certificate information,
438 the item size being so much larger. Found and fixed by Wolfgang Breyha.
440 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
441 size buffer was used, resulting in syntax errors when an expansion
444 JH/07 Add support for directories of certificates when compiled with a GnuTLS
445 version 3.3.6 or later.
447 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
448 is EXPERIMENTAL_EVENT, the main-configuration and transport options
449 both become "event_action", the variables become $event_name, $event_data
450 and $event_defer_errno. There is a new variable $verify_mode, usable in
451 routers, transports and related events. The tls:cert event is now also
452 raised for inbound connections, if the main configuration event_action
455 TL/06 In test suite, disable OCSP for old versions of openssl which contained
456 early OCSP support, but no stapling (appears to be less than 1.0.0).
458 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
459 server certificate names available under the smtp transport option
460 "tls_verify_cert_hostname" now do not permit multi-component wildcard
463 JH/10 Time-related extraction expansions from certificates now use the main
464 option "timezone" setting for output formatting, and are consistent
465 between OpenSSL and GnuTLS compilations. Bug 1541.
467 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
468 encoded parameter in the incoming message. Bug 1558.
470 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
471 include certificate info, eximon was claiming there were spoolfile
474 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
476 JH/14 Log delivery-related information more consistently, using the sequence
477 "H=<name> [<ip>]" wherever possible.
479 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
480 are problematic for Debian distribution, omit them from the release
483 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
485 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
488 JH/17 Fix a null-indirection in certextract expansions when a nondefault
489 output list separator was used.
494 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
495 checkers that were complaining about end of non-void function with no
498 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
499 This was a regression intruduced in 4.83 by another bugfix.
501 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
503 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
504 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
510 TF/01 Correctly close the server side of TLS when forking for delivery.
512 When a message was received over SMTP with TLS, Exim failed to clear up
513 the incoming connection properly after forking off the child process to
514 deliver the message. In some situations the subsequent outgoing
515 delivery connection happened to have the same fd number as the incoming
516 connection previously had. Exim would try to use TLS and fail, logging
517 a "Bad file descriptor" error.
519 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
520 utilities have not been installed.
522 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
523 temporary space as the ACL may create new global variables.
525 TL/01 LDAP support uses per connection or global context settings, depending
526 upon the detected version of the libraries at build time.
528 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
529 to extract and use the src ip:port in logging and expansions as if it
530 were a direct connection from the outside internet. PPv2 support was
531 updated based on HAProxy spec change in May 2014.
533 JH/02 Add ${listextract {number}{list}{success}{fail}}.
535 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
536 Properly escape header and check for NULL return.
538 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
541 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
543 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
544 characters in header names, implemented as a verify condition.
545 Contributed by Michael Fischer v. Mollard.
547 TL/05 Rename SPF condition results err_perm and err_temp to standardized
548 results permerror and temperror. Previous values are deprecated but
549 still accepted. In a future release, err_perm and err_temp will be
550 completely removed, which will be a backward incompatibility if the
551 ACL tests for either of these two old results. Patch contributed by
552 user bes-internal on the mailing list.
554 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
556 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
557 selectors, in both main and reject logs.
559 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
562 JH/07 Add malware type "sock" for talking to simple daemon.
564 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
566 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
567 routers/transports under cutthrough routing.
569 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
570 numbers. Touch up "bool" conditional to keep the same definition.
572 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
574 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
576 JH/12 Expand items in router/transport headers_add or headers_remove lists
577 individually rather than the list as a whole. Bug 1452.
579 Required for reasonable handling of multiple headers_ options when
580 they may be empty; requires that headers_remove items with embedded
581 colons must have them doubled (or the list-separator changed).
583 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
584 view the policy declared in the DMARC record. Currently, $dmarc_status
585 is a combined value of both the record presence and the result of the
588 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
590 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
591 dnslookup router and the smtp transport (applying to the forward
594 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
595 of ldap servers used for a specific lookup. Patch provided by Heiko
598 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
599 New variable $lookup_dnssec_authenticated for observability.
601 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
602 Patch submitted by Lars Timman.
604 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
606 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
607 Requires trusted mode and valid format message id, aborts otherwise.
608 Patch contributed by Heiko Schlichting.
610 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
611 certextract with support for various fields. Bug 1358.
613 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
614 is requested by default, modifiable by smtp transport option
617 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
618 operate on certificate variables to give certificate fingerprints
619 Also new ${sha256:cert_variable}.
621 JH/23 The PRDR feature is moved from being Experimental into the mainline.
623 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
624 Christian Aistleitner.
626 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
628 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
629 file. Patch from Wolfgang Breyha.
631 JH/25 Expand the coverage of the delivery $host and $host_address to
632 client authenticators run in verify callout. Bug 1476.
634 JH/26 Port service names are now accepted for tls_on_connect_ports, to
635 align with daemon_smtp_ports. Bug 72.
637 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
638 support and error reporting did not work properly.
640 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
641 and is readable. Patch from Andrew Colin Kissa.
643 TL/14 Enhance documentation of ${run expansion and how it parses the
644 commandline after expansion, particularly in the case when an
645 unquoted variable expansion results in an empty value.
647 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
649 PP/02 Fix internal collision of T_APL on systems which support RFC3123
650 by renaming away from it. Addresses GH issue 15, reported by
653 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
655 TL/15 SECURITY: prevent double expansion in math comparison functions
656 (can expand unsanitized data). Not remotely exploitable.
663 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
665 PP/02 Make -n do something, by making it not do something.
666 When combined with -bP, the name of an option is not output.
668 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
671 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
672 $sender_host_name and config options to manage this, and basic check
675 PP/05 DSCP support for outbound connections and control modifier for inbound.
677 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
678 (Only plugin which currently uses this is kerberos4, which nobody should
679 be using, but we should make it available and other future plugins might
680 conceivably use it, even though it would break NAT; stuff *should* be
681 using channel bindings instead).
683 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
684 name; added for Sendmail compatibility; requires admin caller.
685 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
686 just ignore it); requires trusted caller.
687 Also parse but ignore: -Ac -Am -X<logfile>
690 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
692 TL/02 Add +smtp_confirmation as a default logging option.
694 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
695 Patch by Magnus Holmgren from 2007-02-20.
697 TL/04 Bugzilla 1281 - Spec typo.
698 Bugzilla 1283 - Spec typo.
699 Bugzilla 1290 - Spec grammar fixes.
701 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
703 TL/06 Add Experimental DMARC support using libopendmarc libraries.
705 TL/07 Fix an out of order global option causing a segfault. Reported to dev
706 mailing list by by Dmitry Isaikin.
708 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
710 JH/02 Support "G" suffix to numbers in ${if comparisons.
712 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
714 NM/01 Bugzilla 1197 - Spec typo
715 Bugzilla 1196 - Spec examples corrections
717 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
719 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
720 gnutls_enable_pkcs11, but renamed to more accurately indicate its
723 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
724 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
726 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
727 "acl {{name}{arg}...}", and optional args on acl condition
730 JH/05 Permit multiple router/transport headers_add/remove lines.
732 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
734 JH/07 Avoid using a waiting database for a single-message-only transport.
735 Performance patch from Paul Fisher. Bugzilla 1262.
737 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
740 JH/09 Add $headers_added variable, with content from use of ACL modifier
741 add_header (but not yet added to the message). Bugzilla 199.
743 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
744 Pulled from Bugzilla 817 by Wolfgang Breyha.
746 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
748 (nb: this is the same fix as in Exim 4.80.1)
750 JH/11 Add A= logging on delivery lines, and a client_set_id option on
753 JH/12 Add optional authenticated_sender logging to A= and a log_selector
756 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
758 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
759 advertise SMTP AUTH mechanism to us, instead of a generic
760 protocol violation error. Also, make Exim more robust to bad
761 data from the Dovecot auth socket.
763 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
765 When a queue runner is handling a message, Exim first routes the
766 recipient addresses, during which it prunes them based on the retry
767 hints database. After that it attempts to deliver the message to
768 any remaining recipients. It then updates the hints database using
771 So if a recipient address works intermittently, it can get repeatedly
772 deferred at routing time. The retry hints record remains fresh so the
773 address never reaches the final cutoff time.
775 This is a fairly common occurrence when a user is bumping up against
776 their storage quota. Exim had some logic in its local delivery code
777 to deal with this. However it did not apply to per-recipient defers
778 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
780 This change adds a proper retry rule check during routing so that the
781 final cutoff time is checked against the message's age. We only do
782 this check if there is an address retry record and there is not a
783 domain retry record; this implies that previous attempts to handle
784 the address had the retry_use_local_parts option turned on. We use
785 this as an approximation for the destination being like a local
786 delivery, as in LMTP.
788 I suspect this new check makes the old local delivery cutoff check
789 redundant, but I have not verified this so I left the code in place.
791 TF/02 Correct gecos expansion when From: is a prefix of the username.
793 Test 0254 submits a message to Exim with the header
797 When I ran the test suite under the user fanf2, Exim expanded
798 the header to contain my full name, whereas it should have added
799 a Resent-Sender: header. It erroneously treats any prefix of the
800 username as equal to the username.
802 This change corrects that bug.
804 GF/01 DCC debug and logging tidyup
805 Error conditions log to paniclog rather than rejectlog.
806 Debug lines prefixed by "DCC: " to remove any ambiguity.
808 TF/03 Avoid unnecessary rebuilds of lookup-related code.
810 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
811 Bug spotted by Jeremy Harris; was flawed since initial commit.
812 Would have resulted in OCSP responses post-SNI triggering an Exim
813 NULL dereference and crash.
815 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
817 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
818 Bug detection, analysis and fix by Samuel Thibault.
819 Bugzilla 1331, Debian bug #698092.
821 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
823 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
824 Server implementation by Todd Lyons, client by JH.
825 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
826 config variable "prdr_enable" controls whether the server
827 advertises the facility. If the client requests PRDR a new
828 acl_data_smtp_prdr ACL is called once for each recipient, after
829 the body content is received and before the acl_smtp_data ACL.
830 The client is controlled by bolth of: a hosts_try_prdr option
831 on the smtp transport, and the server advertisement.
832 Default client logging of deliveries and rejections involving
833 PRDR are flagged with the string "PRDR".
835 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
836 fclose(). Diagnosis by Todd Lyons.
838 PP/17 Update configure.default to handle IPv6 localhost better.
839 Patch by Alain Williams (plus minor tweaks).
842 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
843 This is now consistent with GnuTLS, and is now documented: the
844 previous undocumented portable approach to treating the option as
845 unset was to force an expansion failure. That still works, and
846 an empty string is now equivalent.
848 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
849 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
850 not performing validation itself.
852 PP/20 Added force_command boolean option to pipe transport.
853 Patch from Nick Koston, of cPanel Inc.
855 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
858 TF/04 Added udpsend ACL modifer and hexquote expansion operator
860 PP/21 Fix eximon continuous updating with timestamped log-files.
861 Broken in a format-string cleanup in 4.80, missed when I repaired the
862 other false fix of the same issue.
863 Report and fix from Heiko Schlichting.
866 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
867 Report from Prashanth Katuri.
869 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
870 It's SecureTransport, so affects any MacOS clients which use the
871 system-integrated TLS libraries, including email clients.
873 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
874 using a MIME ACL for non-SMTP local injection.
875 Report and assistance in diagnosis by Warren Baker.
877 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
879 JH/16 Fix comparisons for 64b. Bugzilla 1385.
881 TL/09 Add expansion variable $authenticated_fail_id to keep track of
882 last id that failed so it may be referenced in subsequent ACL's.
884 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
887 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
888 ldap library initialization, allowing self-signed CA's to be
889 used. Also properly sets require_cert option later in code by
890 using NULL (global ldap config) instead of ldap handle (per
891 session). Bug diagnosis and testing by alxgomz.
893 TL/12 Enhanced documentation in the ratelimit.pl script provided in
894 the src/util/ subdirectory.
896 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
897 renamed to Transport Post Delivery Action by Jeremy Harris, as
900 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
901 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
902 redis_servers = needs to be configured which will be used by the redis
903 lookup. Patch from Warren Baker, of The Packet Hub.
905 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
907 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
908 hostname or reverse DNS when processing a host list. Used suggestions
909 from multiple comments on this bug.
911 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
913 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
914 Missed a few lines, added it to make the runtest require no keyboard
917 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
918 contains upper case chars. Make router use caseful_local_part.
920 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
921 support when GnuTLS has been built with p11-kit.
927 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
929 This, or similar/improved, will also be change PP/11 of 4.82.
935 PP/01 Handle short writes when writing local log-files.
936 In practice, only affects FreeBSD (8 onwards).
937 Bugzilla 1053, with thanks to Dmitry Isaikin.
939 NM/01 Bugzilla 949 - Documentation tweak
941 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
944 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
946 PP/02 Implemented gsasl authenticator.
948 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
950 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
951 `pkg-config foo` for cflags/libs.
953 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
954 with rest of GSASL and with heimdal_gssapi.
956 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
957 `pkg-config foo` for cflags/libs for the TLS implementation.
959 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
960 properties get this fed in as external SSF. A number of robustness
961 and debugging improvements to the cyrus_sasl authenticator.
963 PP/08 cyrus_sasl server now expands the server_realm option.
965 PP/09 Bugzilla 1214 - Log authentication information in reject log.
966 Patch by Jeremy Harris.
968 PP/10 Added dbmjz lookup type.
970 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
972 PP/12 MAIL args handles TAB as well as SP, for better interop with
973 non-compliant senders.
974 Analysis and variant patch by Todd Lyons.
976 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
977 Bug report from Lars Müller <lars@samba.org> (via SUSE),
978 Patch from Dirk Mueller <dmueller@suse.com>
980 PP/13 tls_peerdn now print-escaped for spool files.
981 Observed some $tls_peerdn in wild which contained \n, which resulted
982 in spool file corruption.
984 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
985 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
986 or write after TLS renegotiation, which otherwise led to messages
989 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
990 as a tracking header (ie: a signed header comes before the signature).
991 Patch from Wolfgang Breyha.
993 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
994 comma-sep list; embedded commas doubled.
996 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
998 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1000 Report and patch from Dmitry Banschikov.
1002 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1003 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1004 failures. We appear to now support TLS1.1+ with Exim.
1006 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1007 lets Exim select keys and certificates based upon TLS SNI from client.
1008 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1009 before an outbound SMTP session. New log_selector, +tls_sni.
1011 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1012 NULL dereference. Report and patch from Alun Jones.
1014 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1015 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1016 Not seeing resolver debug output on NetBSD, but suspect this is a
1017 resolver implementation change.
1019 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1020 Left warnings. Added "eximon gdb" invocation mode.
1022 PP/21 Defaulting "accept_8bitmime" to true, not false.
1024 PP/22 Added -bw for inetd wait mode support.
1026 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1027 locate the relevant includes and libraries. Made this the default.
1029 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1030 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1032 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1033 This may cause build issues on older platforms.
1035 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1036 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1037 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1038 Added SNI support via GnuTLS too.
1039 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1041 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1043 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1044 Applied second patch from Janne, implementing suggestion to default
1045 multiple-strings-in-record handling to match SPF spec.
1047 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1049 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1050 read-only, out of scope).
1051 Patch from Wolfgang Breyha, report from Stuart Northfield.
1053 PP/29 Fix three issues highlighted by clang analyser static analysis.
1054 Only crash-plausible issue would require the Cambridge-specific
1055 iplookup router and a misconfiguration.
1056 Report from Marcin Mirosław.
1058 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1060 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1061 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1062 As part of this, removing so much warning spew let me fix some minor
1063 real issues in debug logging.
1065 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1066 assignment on my part. Fixed.
1068 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1069 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1070 Janne Snabb (who went above and beyond: thank you).
1072 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1073 string otherwise requires a connection and a bunch more work and it's
1074 relatively easy to get wrong. Should also expose TLS library linkage
1077 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1078 64-bit ${eval} (JH/03).
1080 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1081 GNU libc to support some of the 64-bit stuff, should not lead to
1082 conflicts. Defined before os.h is pulled in, so if a given platform
1083 needs to override this, it can.
1085 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1086 protection layer was required, which is not implemented.
1087 Bugzilla 1254, patch from Wolfgang Breyha.
1089 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1090 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1091 tls_dhparam take prime identifiers. Also unbreak combination of
1092 OpenSSL+DH_params+TLSSNI.
1094 PP/39 Disable SSLv2 by default in OpenSSL support.
1100 PP/01 Solaris build fix for Oracle's LDAP libraries.
1101 Bugzilla 1109, patch from Stephen Usher.
1103 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1105 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1108 TF/02 Fix a couple more cases where we did not log the error message
1109 when unlink() failed. See also change 4.74-TF/03.
1111 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1112 lock up or crash if it happened to be inside a call to libc when it
1113 got a SIGUSR1 from exiwhat.
1115 The SIGUSR1 handler appends the current process status to the process
1116 log which is later printed by exiwhat. It used to use the general
1117 purpose logging code to do this, but several functions it calls are
1118 not safe for signals.
1120 The new output code in the SIGUSR1 handler is specific to the process
1121 log, and simple enough that it's easy to inspect for signal safety.
1122 Removing some special cases also simplifies the general logging code.
1123 Removing the spurious timestamps from the process log simplifies
1126 TF/04 Improved ratelimit ACL condition.
1128 The /noupdate option has been deprecated in favour of /readonly which
1129 has clearer semantics. The /leaky, /strict, and /readonly update modes
1130 are mutually exclusive. The update mode is no longer included in the
1131 database key; it just determines when the database is updated. (This
1132 means that when you upgrde Exim will forget old rate measurements.)
1134 Exim now checks that the per_* options are used with an update mode that
1135 makes sense for the current ACL. For example, when Exim is processing a
1136 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1137 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1138 must specify per_mail/readonly. If you omit the update mode it defaults to
1139 /leaky where that makes sense (as before) or /readonly where required.
1141 The /noupdate option is now undocumented but still supported for
1142 backwards compatibility. It is equivalent to /readonly except that in
1143 ACLs where /readonly is required you may specify /leaky/noupdate or
1144 /strict/noupdate which are treated the same as /readonly.
1146 A useful new feature is the /count= option. This is a generalization
1147 of the per_byte option, so that you can measure the throughput of other
1148 aggregate values. For example, the per_byte option is now equivalent
1149 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1151 The per_rcpt option has been generalized using the /count= mechanism
1152 (though it's more complicated than the per_byte equivalence). When it is
1153 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1154 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1155 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1156 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1157 Note that using per_rcpt with a non-readonly update mode in more than
1158 one ACL will cause the recipients to be double-counted. (The per_mail
1159 and per_byte options don't have this problem.)
1161 The handling of very low rates has changed slightly. If the computed rate
1162 is less than the event's count (usually one) then this event is the first
1163 after a long gap. In this case the rate is set to the same as this event's
1164 count, so that the first message of a spam run is counted properly.
1166 The major new feature is a mechanism for counting the rate of unique
1167 events. The new per_addr option counts the number of different
1168 recipients that someone has sent messages to in the last time period. It
1169 behaves like per_rcpt if all the recipient addresses are different, but
1170 duplicate recipient addresses do not increase the measured rate. Like
1171 the /count= option this is a general mechanism, so the per_addr option
1172 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1173 example, measure the rate that a client uses different sender addresses
1174 with the options per_mail/unique=$sender_address. There are further
1175 details in the main documentation.
1177 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1179 TF/06 Removed a few PCRE remnants.
1181 TF/07 Automatically extract Exim's version number from tags in the git
1182 repository when doing development or release builds.
1184 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1185 Bugzilla 879. Patch from Paul Fisher.
1187 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1188 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1191 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1193 PP/05 Variable $av_failed, true if the AV scanner deferred.
1194 Bugzilla 1078. Patch from John Horne.
1196 PP/06 Stop make process more reliably on build failure.
1197 Bugzilla 1087. Patch from Heiko Schlittermann.
1199 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1200 Bugzilla 1089. Patch from Heiko Schlittermann.
1202 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1203 Bugzilla 1131. Patch from Holger Weiß.
1205 PP/09 Handle IPv6 addresses with SPF.
1206 Bugzilla 860. Patch from Wolfgang Breyha.
1208 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1210 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1213 PP/11 match_* no longer expand right-hand-side by default.
1214 New compile-time build option, EXPAND_LISTMATCH_RHS.
1215 New expansion conditions, "inlist", "inlisti".
1217 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1219 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1221 PP/14 fix log_write() format string regression from TF/03.
1222 Bugzilla 1152. Patch from Dmitry Isaikin.
1228 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1230 PP/02 Harmonised TLS library version reporting; only show if debugging.
1231 Layout now matches that introduced for other libraries in 4.74 PP/03.
1233 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1235 PP/04 New "dns_use_edns0" global option.
1237 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1240 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1241 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1243 TK/01 Updated PolarSSL code to 0.14.2.
1244 Bugzilla 1097. Patch from Andreas Metzler.
1246 PP/07 Catch divide-by-zero in ${eval:...}.
1247 Fixes bugzilla 1102.
1249 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1252 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1253 format-string attack -- SECURITY: remote arbitrary code execution.
1255 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1256 time unintentionally subject to list matching rules, letting the header
1257 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1258 arbitrary string expansion). This allowed for information disclosure.
1260 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1261 INT_MIN/-1 -- value coerced to INT_MAX.
1267 NM/01 Workround for PCRE version dependency in version reporting
1270 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1271 This fixes portability to compilers other than gcc, notably
1272 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1274 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1275 makefiles for portability to HP-UX and POSIX correctness.
1277 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1278 Also via indented variable definition in the Makefile.
1279 (Debugging by Oliver Heesakkers).
1281 PP/02 Restore caching of spamd results with expanded spamd_address.
1282 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1284 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1285 Improves build reliability. Fix from: Frank Elsner
1287 NM/02 Fix wide character breakage in the rfc2047 coding
1288 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1290 NM/03 Allow underscore in dnslist lookups
1291 Fixes bug 1026. Patch from Graeme Fowler
1293 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1294 Code patches from Adam Ciarcinski of NetBSD.
1296 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1299 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1300 is logged, to avoid truncation. Patch from John Horne.
1302 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1303 Patch from Jakob Hirsch.
1305 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1306 SQL string expansion failure details.
1307 Patch from Andrey Oktyabrski.
1309 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1310 Patch from Simon Arlott.
1312 PP/09 New lookups functionality failed to compile on old gcc which rejects
1313 extern declarations in function scope.
1314 Patch from Oliver Fleischmann
1316 PP/10 Use sig_atomic_t for flags set from signal handlers.
1317 Check getgroups() return and improve debugging.
1318 Fixed developed for diagnosis in bug 927 (which turned out to be
1321 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1322 Patch from Mark Zealey.
1324 PP/12 Bugzilla 1056: Improved spamd server selection.
1325 Patch from Mark Zealey.
1327 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1328 Based on patch from Heiko Schlittermann.
1330 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1331 Patch from Uwe Doering, sign-off by Michael Haardt.
1333 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1334 variable declaration deep within a block. Bug and patch from
1337 PP/15 lookups-Makefile IRIX compatibilty coercion.
1339 PP/16 Make DISABLE_DKIM build knob functional.
1341 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1342 Patch by Simon Arlott
1344 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1345 variable argument macros. Our copy now differs from upstream.
1351 TF/01 Failure to get a lock on a hints database can have serious
1352 consequences so log it to the panic log.
1354 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1355 controlled using the smtp_confirmation log selector.
1357 TF/03 Include the error message when we fail to unlink a spool file.
1359 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1360 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1361 for maintaining out-of-tree patches for some time.
1363 PP/01 Bugzilla 139: Documentation and portability issues.
1364 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1365 Handle per-OS dynamic-module compilation flags.
1367 PP/02 Let /dev/null have normal permissions.
1368 The 4.73 fixes were a little too stringent and complained about the
1369 permissions on /dev/null. Exempt it from some checks.
1370 Reported by Andreas M. Kirchwitz.
1372 PP/03 Report version information for many libraries, including
1373 Exim version information for dynamically loaded libraries. Created
1374 version.h, now support a version extension string for distributors
1375 who patch heavily. Dynamic module ABI change.
1377 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1378 privilege escalation vulnerability whereby the Exim run-time user
1379 can cause root to append content of the attacker's choosing to
1382 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1385 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1386 If dropping privileges for untrusted macros, we disabled normal logging
1387 on the basis that it would fail; for the Exim run-time user, this is not
1388 the case, and it resulted in successful deliveries going unlogged.
1389 Fixed. Reported by Andreas Metzler.
1395 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1396 only prepend for the Resent-* case. Fixes regression introduced in
1397 Exim 4.70 by NM/22 for Bugzilla 607.
1399 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1400 increasing numbers of administrators be bitten by this.
1402 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1404 PP/03 Bugzilla 994: added openssl_options main configuration option.
1406 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1408 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1410 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1412 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1413 without a peer certificate, leading to a segfault because of an
1414 assumption that peers always have certificates. Be a little more
1415 paranoid. Problem reported by Martin Tscholak.
1417 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1418 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1419 NB: ClamAV planning to remove STREAM in "middle of 2010".
1420 CL also introduces -bmalware, various -d+acl logging additions and
1421 more caution in buffer sizes.
1423 PP/09 Implemented reverse_ip expansion operator.
1425 PP/10 Bugzilla 937: provide a "debug" ACL control.
1427 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1429 PP/12 Bugzilla 973: Implement --version.
1431 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1433 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1435 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1437 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1438 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1439 ignore trailing whitespace.
1441 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1443 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1446 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1447 Notification from Dr Andrew Aitchison.
1449 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1450 ExtendedDetectionInfo response format.
1451 Notification from John Horne.
1453 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1456 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1457 XSL and documented dependency on system catalogs, with examples of how
1460 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1463 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1464 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1465 configuration file which is writeable by the Exim user or group.
1467 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1468 of configuration files to cover files specified with the -C option if
1469 they are going to be used with root privileges, not just the default
1472 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1473 option (effectively making it always true).
1475 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1476 files to be used while preserving root privileges.
1478 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1479 that rogue child processes cannot use them.
1481 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1482 run-time user, instead of root.
1484 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1485 Exim run-time user without dropping privileges.
1487 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1488 result string, instead of calling string_vformat() twice with the same
1491 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1492 for other users. Others should always drop root privileges if they use
1493 -C on the command line, even for a whitelisted configure file.
1495 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1497 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1503 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1504 $data_path, and $header_path variables; fixed documentation bugs and
1507 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1508 exipick to access non-standard spools, including the "frozen" queue
1511 NM/01 Bugzilla 965: Support mysql stored procedures.
1512 Patch from Alain Williams
1514 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1516 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1517 Patch from Andreas Metzler
1519 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1520 Patch from Kirill Miazine
1522 NM/05 Bugzilla 671: Added umask to procmail example.
1524 JJ/03 installed exipick 20100323.0, fixing doc bug
1526 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1527 directory. Notification and patch from Dan Rosenberg.
1529 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1531 TK/02 Improve log output when DKIM signing operation fails.
1533 MH/01 Treat the transport option dkim_domain as a colon separated
1534 list, not as a single string, and sign the message with each element,
1535 omitting multiple occurences of the same signer.
1537 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1538 Bugzilla 985, 986. Patch by Simon Arlott
1540 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1541 Patch by Simon Arlott
1543 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1544 MBX locking. Notification from Dan Rosenberg.
1550 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1552 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1554 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1556 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1558 NM/04 Bugzilla 915: Flag broken perl installation during build.
1564 TK/01 Added patch by Johannes Berg that expands the main option
1565 "spamd_address" if it starts with a dollar sign.
1567 TK/02 Write list of recipients to X-Envelope-Sender header when building
1568 the mbox-format spool file for content scanning (suggested by Jakob
1571 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1572 (http://www.dcc-servers.net/) support via dccifd. Activated by
1573 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1575 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1576 by Mark Daniel Reidel <mr@df.eu>.
1578 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1579 When building exim an external PCRE library is now needed -
1580 PCRE is a system library on the majority of modern systems.
1581 See entry on PCRE_LIBS in EDITME file.
1583 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1584 conversation. Added nologin parameter to request.
1585 Patch contributed by Kirill Miazine.
1587 TF/01 Do not log submission mode rewrites if they do not change the address.
1589 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1591 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1592 log files in place. Contributed by Roberto Lima.
1594 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1596 TF/03 Bugzilla 615: When checking the local_parts router precondition
1597 after a local_part_suffix or local_part_prefix option, Exim now
1598 does not use the address's named list lookup cache, since this
1599 contains cached lookups for the whole local part.
1601 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1602 Robert Millan. Documentation is in experimental-spec.txt.
1604 TF/04 Bugzilla 668: Fix parallel build (make -j).
1606 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1608 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1609 Patch provided by Jan Srzednicki.
1611 TF/05 Leading white space used to be stripped from $spam_report which
1612 wrecked the formatting. Now it is preserved.
1614 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1615 that they are available at delivery time.
1617 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1619 TF/08 TLS error reporting now respects the incoming_interface and
1620 incoming_port log selectors.
1622 TF/09 Produce a more useful error message if an SMTP transport's hosts
1623 setting expands to an empty string.
1625 NM/06 Bugzilla 744: EXPN did not work under TLS.
1626 Patch provided by Phil Pennock.
1628 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1629 Patch provided by Richard Godbee.
1631 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1632 acl_smtp_notquit, added index entry.
1634 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1635 Patch provided by Eugene Bujak.
1637 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1638 accept(). Patch provided by Maxim Dounin.
1640 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1641 Patch provided by Phil Pennock.
1643 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1645 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1646 Patch provided by Brad "anomie" Jorsch.
1648 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1649 Patch provided by Dean Brooks.
1651 TK/05 Add native DKIM support (does not depend on external libraries).
1653 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1654 Patch provided by Graeme Fowler.
1656 NM/16 Bugzilla 851: Documentation example syntax fix.
1658 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1660 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1663 NM/19 Bugzilla 745: TLS version reporting.
1664 Patch provided by Phil Pennock.
1666 NM/20 Bugzilla 167: bool: condition support.
1667 Patch provided by Phil Pennock.
1669 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1670 clients. Patch provided by Phil Pennock.
1672 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1673 Patch provided by Brad "anomie" Jorsch.
1675 NM/23 Bugzilla 687: Fix misparses in eximstats.
1676 Patch provided by Heiko Schlittermann.
1678 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1679 Patch provided by Heiko Schlittermann.
1681 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1682 plus update to original patch.
1684 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1686 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1687 Patch provided by David Brownlee.
1689 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1691 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1693 NM/30 Bugzilla 888: TLS documentation bugfixes.
1695 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1697 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1698 Unlike the original bugzilla I have changed all shell scripts in src tree.
1700 NM/33 Bugzilla 898: Transport filter timeout fix.
1701 Patch by Todd Rinaldo.
1703 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1704 Patch by Serge Demonchaux.
1706 NM/35 Bugzilla 39: Base64 decode bug fixes.
1707 Patch by Jakob Hirsch.
1709 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1711 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1713 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1715 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1721 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1722 ALT-N's libdkim that I have put here:
1723 http://duncanthrax.net/exim-experimental/
1725 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1726 were called 'true' and it seems that C99 defines that as a reserved
1727 keyword to be used with 'bool' variable types. That means you could
1728 not include C99-style headers which use bools without triggering
1729 build errors in sieve.c.
1731 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1732 as mailq or other aliases. Changed the --help handling significantly
1733 to do whats expected. exim_usage() emits usage/help information.
1735 SC/01 Added the -bylocaldomain option to eximstats.
1737 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1739 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1741 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1747 PH/01 Another patch from the Sieve maintainer.
1749 PH/02 When an IPv6 address is converted to a string for single-key lookup
1750 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1751 dots are used instead of colons so that keys in lsearch files need not
1752 contain colons. This was done some time before quoting was made available
1753 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1754 (notated using the quote facility) so as to distinguish them from IPv4
1755 keys. This meant that lookups for IP addresses in host lists did not work
1756 for iplsearch lookups.
1758 This has been fixed by arranging for IPv6 addresses to be expressed with
1759 colons if the lookup type is iplsearch. This is not incompatible, because
1760 previously such lookups could never work.
1762 The situation is now rather anomolous, since one *can* have colons in
1763 ordinary lsearch keys. However, making the change in all cases is
1764 incompatible and would probably break a number of configurations.
1766 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1769 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1770 conversion specification without a maximum field width, thereby enabling
1771 a rogue spamd server to cause a buffer overflow. While nobody in their
1772 right mind would setup Exim to query an untrusted spamd server, an
1773 attacker that gains access to a server running spamd could potentially
1774 exploit this vulnerability to run arbitrary code as the Exim user.
1776 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1777 $primary_hostname instead of what libspf2 thinks the hosts name is.
1779 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1780 a directory entry by the name of the lookup key. Previously, if a
1781 symlink pointed to a non-existing file or a file in a directory that
1782 Exim lacked permissions to read, a lookup for a key matching that
1783 symlink would fail. Now it is enough that a matching directory entry
1784 exists, symlink or not. (Bugzilla 503.)
1786 PH/03 The body_linecount and body_zerocount variables are now exported in the
1789 PH/04 Added the $dnslist_matched variable.
1791 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1792 This means they are set thereafter only if the connection becomes
1795 PH/06 Added the client_condition to authenticators so that some can be skipped
1796 by clients under certain conditions.
1798 PH/07 The error message for a badly-placed control=no_multiline_responses left
1799 "_responses" off the end of the name.
1801 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1803 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1804 (without spaces) instead of just copying the configuration text.
1806 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1808 PH/11 Added $max_received_linelength.
1810 PH/12 Added +ignore_defer and +include_defer to host lists.
1812 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1813 way in which PCRE > 7.0 is built.
1815 PH/14 Implemented queue_only_load_latch.
1817 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1818 MAIL command. The effect was to mangle the value on 64-bit systems.
1820 PH/16 Another patch from the Sieve maintainer.
1822 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1824 PH/18 If a system quota error occurred while trying to create the file for
1825 a maildir delivery, the message "Mailbox is full" was not appended to the
1826 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1827 only to a quota excession during the actual writing of the file.
1829 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1830 characters?) which causes problems in log lines. The DN values are now
1831 passed through string_printing() before being added to log lines.
1833 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1834 and InterBase are left for another time.)
1836 PH/21 Added message_body_newlines option.
1838 PH/22 Guard against possible overflow in moan_check_errorcopy().
1840 PH/23 POSIX allows open() to be a macro; guard against that.
1842 PH/24 If the recipient of an error message contained an @ in the local part
1843 (suitably quoted, of course), incorrect values were put in $domain and
1844 $local_part during the evaluation of errors_copy.
1850 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1851 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1854 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1855 issue a MAIL command.
1857 PH/02 In an ACL statement such as
1859 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1861 if a client was not listed at all, or was listed with a value other than
1862 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1863 the condition was not true (as it should be), so access was not denied.
1864 The bug was that the ! inversion was incorrectly passed on to the second
1865 item. This has been fixed.
1867 PH/03 Added additional dnslists conditions == and =& which are different from
1868 = and & when the dns lookup returns more than one IP address.
1870 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1871 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1873 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1874 FSYNC, which compiles an option called disable_fsync that allows for
1875 bypassing fsync(). The documentation is heavily laced with warnings.
1877 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1879 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1880 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1881 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1882 including adding "make clean"; (3) Added -fPIC when compiling the test
1883 dynamically loaded module, to get rid of a warning.
1885 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1886 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1887 = 0s. The bug is otherwise harmless.
1889 PH/07 There was a bug in the dovecot authenticator such that the value of
1890 $auth1 could be overwritten, and so not correctly preserved, after a
1891 successful authentication. This usually meant that the value preserved by
1892 the server_setid option was incorrect.
1894 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1896 PH/09 Installed PCRE release 7.0.
1898 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1899 run for batched SMTP input. It is now run at the start of every message
1900 in the batch. While fixing this I discovered that the process information
1901 (output by running exiwhat) was not always getting set for -bs and -bS
1902 input. This is fixed, and it now also says "batched" for BSMTP.
1904 PH/11 Added control=no_pipelining.
1906 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1907 patch, slightly modified), and move the expansion of helo_data till after
1908 the connection is made in the smtp transport (so it can use these
1911 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1913 PH/14 Added log_selector = +pid.
1915 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1917 PH/16 Add ${if forany and ${if forall.
1919 PH/17 Added dsn_from option to vary the From: line in DSNs.
1921 PH/18 Flush SMTP output before performing a callout, unless control =
1922 no_callout_flush is set.
1924 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1925 was true (the default) a successful delivery failed to delete the retry
1926 item, thus causing premature timeout of the address. The bug is now
1929 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1931 PH/21 Long custom messages for fakedefer and fakereject are now split up
1932 into multiline reponses in the same way that messages for "deny" and
1933 other ACL rejections are.
1935 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1936 with slight modification.
1938 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1939 draft, changing the syntax and factoring some duplicate code".
1941 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1942 for deliveries of the second and subsequent messages over the same SMTP
1945 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1946 ${reduce, with only minor "tidies".
1948 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1950 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1951 expansion side effects.
1953 PH/27 When a message times out after an over-quota error from an Exim-imposed
1954 quota, the bounce message says "mailbox is full". This message was not
1955 being given when it was a system quota that was exceeded. It now should
1958 MH/03 Made $recipients available in local_scan(). local_scan() already has
1959 better access to the recipient list through recipients_list[], but
1960 $recipients can be useful in postmaster-provided expansion strings.
1962 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1963 in the case of a MAIL command with additional options following the
1964 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1965 were accidentally chopped off.
1967 PH/29 SMTP synchronization checks are implemented when a command is read -
1968 there is a check that no more input is waiting when there shouldn't be
1969 any. However, for some commands, a delay in an ACL can mean that it is
1970 some time before the response is written. In this time, more input might
1971 arrive, invalidly. So now there are extra checks after an ACL has run for
1972 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1973 pipelining has not been advertised.
1975 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1977 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1978 returned in the SMTP response when smtp_return_error_details was false.
1979 This has been fixed.
1981 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1982 instead of the C I/O that was originally supplied, because problems were
1983 reported on Solaris.
1985 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1986 Exim which did not show up earlier: it was assuming that a call to
1987 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1988 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1989 was a macro that became an assignment, so it seemed to work. This has
1990 changed to a proper function call with a void return, hence the compile
1991 error. Exim's code has been fixed.
1993 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1996 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1998 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2000 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2001 because it turns out to be too big or there is a timeout) there is no
2002 "Completed" line in the log. When some messages of this type were
2003 selected by exigrep, they were listed as "not completed". Others were
2004 picked up by some special patterns. I have improved the selection
2005 criteria to be more general.
2007 PH/38 The host_find_failed option in the manualroute router can now be set
2008 to "ignore", to completely ignore a host whose IP address cannot be
2009 found. If all hosts are ignored, the behaviour is controlled by the new
2010 host_all_ignored option.
2012 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2013 homing or because of multiple MX records with /mx) generated more than
2014 one IP address, and the following item turned out to be the local host,
2015 all the secondary addresses of the first item were incorrectly removed
2016 from the list, along with the local host and any following hosts (which
2017 is what is supposed to happen).
2019 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2020 whoever called Exim into the -H file. In the case of the daemon it was
2021 behaving confusingly. When first started, it used values for whoever
2022 started the daemon, but after a SIGHUP it used the Exim user (because it
2023 calls itself on a restart). I have changed the code so that it now always
2026 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2027 message are rejected with the same error (e.g. no authentication or bad
2028 sender address), and a DATA command is nevertheless sent (as can happen
2029 with PIPELINING or a stupid MUA), the error message that was given to the
2030 RCPT commands is included in the rejection of the DATA command. This is
2031 intended to be helpful for MUAs that show only the final error to their
2034 PH/42 Another patch from the Sieve maintainer.
2036 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2037 Eximstats - Fixed some broken HTML links and added missing column headers
2039 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2040 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2042 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2044 PH/43 Yet another patch from the Sieve maintainer.
2046 PH/44 I found a way to check for a TCP/IP connection going away before sending
2047 the response to the final '.' that terminates a message, but only in the
2048 case where the client has not sent further data following the '.'
2049 (unfortunately, this is allowed). However, in many cases there won't be
2050 any further data because there won't be any more messages to send. A call
2051 to select() can be used: if it shows that the input is "ready", there is
2052 either input waiting, or the socket has been closed. An attempt to read
2053 the next input character can distinguish the two cases. Previously, Exim
2054 would have sent an OK response which the client would never have see.
2055 This could lead to message repetition. This fix should cure that, at
2056 least in a lot of common cases.
2058 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2059 advertised in response to EHLO.
2065 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2066 fixed by 4.65/MH/01 (is this a record?) are fixed:
2068 (i) An empty string was always treated as zero by the numeric comparison
2069 operators. This behaviour has been restored.
2071 (ii) It is documented that the numeric comparison operators always treat
2072 their arguments as decimal numbers. This was broken in that numbers
2073 starting with 0 were being interpreted as octal.
2075 While fixing these problems I realized that there was another issue that
2076 hadn't been noticed. Values of message_size_limit (both the global option
2077 and the transport option) were treated as octal if they started with 0.
2078 The documentation was vague. These values are now always treated as
2079 decimal, and I will make that clear in the documentation.
2085 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2086 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2089 MH/01 Don't check that the operands of numeric comparison operators are
2090 integers when their expansion is in "skipping" mode (fixes bug
2091 introduced by 4.64-PH/07).
2093 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2094 child addresses, Exim now panics and dies. Previously, because the count
2095 is held in a short int, deliveries were likely to be lost. As such a
2096 large number of recipients for a single message is ridiculous
2097 (performance will be very, very poor), I have chosen to impose a limit
2098 rather than extend the field.
2104 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2105 leftover -K file (the existence of which was triggered by #402).
2106 While we were at it, introduced process PID as part of the -K
2107 filename. This should rule out race conditions when creating
2110 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2111 processing considerably. Previous code took too long for large mails,
2112 triggering a timeout which in turn triggers #401.
2114 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2115 in the DK code in transports.c. sendfile() is not really portable,
2116 hence the _LINUX specificness.
2118 TF/01 In the add_headers option to the mail command in an Exim filter,
2119 there was a bug that Exim would claim a syntax error in any
2120 header after the first one which had an odd number of characters
2123 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2124 callout verification, Exim cached a "reject" for the entire domain. This
2125 is correct for most verifications, but it is not correct for a recipient
2126 verification with use_sender or use_postmaster set, because in that case
2127 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2128 case of MAIL FROM:<> rejection from other early rejections (e.g.
2129 rejection of HELO). When verifying a recipient using a non-null MAIL
2130 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2131 Whatever the result of the callout, the value of the domain cache is
2132 left unchanged (for any other kind of callout, getting as far as trying
2133 RCPT means that the domain itself is ok).
2135 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2138 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2139 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2140 session, the daemon ignores ECONNECT errors and logs others; it now
2141 ignores EPIPE as well.
2143 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2144 (quoted-printable decoding).
2146 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2147 later the small subsequent patch to fix an introduced bug.
2149 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2151 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2153 PH/08 An error is now given if message_size_limit is specified negative.
2155 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2156 to be given (somewhat) arbitrary names.
2158 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2161 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2162 miscellaneous code fixes
2164 PH/10 Added the log_reject_target ACL modifier to specify where to log
2167 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2168 hostname. This is wrong, because it relates to the incoming message (and
2169 probably the interface on which it is arriving) and not to the outgoing
2170 callout (which could be using a different interface). This has been
2171 changed to use the value of the helo_data option from the smtp transport
2172 instead - this is what is used when a message is actually being sent. If
2173 there is no remote transport (possible with a router that sets up host
2174 addresses), $smtp_active_hostname is used.
2176 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2177 tweaks were necessary in order to get it to work (see also 21 below):
2178 (a) The code assumed that strncpy() returns a negative number on buffer
2179 overflow, which isn't the case. Replaced with Exim's string_format()
2181 (b) There were several signed/unsigned issues. I just did the minimum
2182 hacking in of casts. There is scope for a larger refactoring.
2183 (c) The code used strcasecmp() which is not a standard C function.
2184 Replaced with Exim's strcmpic() function.
2185 (d) The code set only $1; it now sets $auth1 as well.
2186 (e) A simple test gave the error "authentication client didn't specify
2187 service in request". It would seem that Dovecot has changed its
2188 interface. Fortunately there's a specification; I followed it and
2189 changed what the client sends and it appears to be working now.
2191 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2194 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2195 address A is aliased to B and C, where B exists and C does not. Without
2196 -v the output is "A verified" because verification stops after a
2197 successful redirection if more than one address is generated. However,
2198 with -v the child addresses are also verified. Exim was outputting "A
2199 failed to verify" and then showing the successful verification for C,
2200 with its parentage. It now outputs "B failed to verify", showing B's
2201 parentage before showing the successful verification of C.
2203 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2204 look up a TXT record in a specific list after matching in a combined
2207 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2208 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2209 they consult the DNS. I had assumed they would set it the way they
2210 wanted; and indeed my experiments on Linux seem to show that in some
2211 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2212 To be on the safe side, however, I have now made the interface to
2213 host_find_byname() similar to host_find_bydns(), with an argument
2214 containing the DNS resolver options. The host_find_byname() function now
2215 sets these options at its start, just as host_find_bydns() does. The smtp
2216 transport options dns_qualify_single and dns_search_parents are passed to
2217 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2218 of host_find_byname() use the default settings of RES_DEFNAMES
2219 (qualify_single) but not RES_DNSRCH (search_parents).
2221 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2222 spool_read_header() do less string testing, by means of a preliminary
2223 switch on the second character of optional "-foo" lines. (This is
2224 overdue, caused by the large number of possibilities that now exist.
2225 Originally there were few.) While I was there, I also converted the
2226 str(n)cmp tests so they don't re-test the leading "-" and the first
2227 character, in the hope this might squeeze out yet more improvement.
2229 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2230 flag allowing group syntax was set by the header_syntax check but not
2231 turned off, possible causing trouble later; (2) The flag was not being
2232 set at all for the header_verify test, causing "group"-style headers to
2233 be rejected. I have now set it in this case, and also caused header_
2234 verify to ignore an empty address taken from a group. While doing this, I
2235 came across some other cases where the code for allowing group syntax
2236 while scanning a header line wasn't quite right (mostly, not resetting
2237 the flag correctly in the right place). These bugs could have caused
2238 trouble for malformed header lines. I hope it is now all correct.
2240 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2241 with the "reply" argument non-NULL. The code, however (which originally
2242 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2243 but it didn't always do it. This confused somebody who was copying the
2244 code for some other use. I have removed all the tests.
2246 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2247 feature that was used to support insecure browsers during the U.S. crypto
2248 embargo. It requires special client support, and Exim is probably the
2249 only MTA that supported it -- and would never use it because real RSA is
2250 always available. This code has been removed, because it had the bad
2251 effect of slowing Exim down by computing (never used) parameters for the
2252 RSA_EXPORT functionality.
2254 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2255 authenticator to fail if there's a tab character in the incoming data
2256 (there should never be unless someone is messing about, as it's supposed
2257 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2258 if the connection is using TLS or if the remote IP is the same as the
2259 local IP, and the "valid-client-cert option" if a client certificate has
2262 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2263 authenticators. This can be used for authorization after authentication
2264 succeeds. (In the case of plaintext, it servers for both authentication
2267 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2268 if any retry times were supplied.
2270 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2271 connection, where there is no client IP address to check. In this
2272 situation, the verify now always succeeds.
2274 PH/25 Applied John Jetmore's -Mset patch.
2276 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2278 PH/27 In a string expansion for a processed (not raw) header when multiple
2279 headers of the same name were present, leading whitespace was being
2280 removed from all of them, but trailing whitespace was being removed only
2281 from the last one. Now trailing whitespace is removed from each header
2282 before concatenation. Completely empty headers in a concatenation (as
2283 before) are ignored.
2285 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2286 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2288 PH/29 [Removed. This was a change that I later backed out, and forgot to
2289 correct the ChangeLog entry (that I had efficiently created) before
2290 committing the later change.]
2292 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2293 address errors (4xx response to RCPT) over the same connection as other
2294 messages routed to the same hosts. Such deliveries are always "forced",
2295 so retry times are not inspected. This resulted in far too many retries
2296 for the affected addresses. The effect occurred only when there were more
2297 hosts than the hosts_max_try setting in the smtp transport when it had
2298 the 4xx errors. Those hosts that it had tried were not added to the list
2299 of hosts for which the message was waiting, so if all were tried, there
2300 was no problem. Two fixes have been applied:
2302 (i) If there are any address or message errors in an SMTP delivery, none
2303 of the hosts (tried or untried) are now added to the list of hosts
2304 for which the message is waiting, so the message should not be a
2305 candidate for sending over the same connection that was used for a
2306 successful delivery of some other message. This seems entirely
2307 reasonable: after all the message is NOT "waiting for some host".
2308 This is so "obvious" that I'm not sure why it wasn't done
2309 previously. Hope I haven't missed anything, but it can't do any
2310 harm, as the worst effect is to miss an optimization.
2312 (ii) If, despite (i), such a delivery is accidentally attempted, the
2313 routing retry time is respected, so at least it doesn't keep
2314 hammering the server.
2316 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2317 in ${readsocket because some servers need this prod.
2319 PH/32 Added some extra debug output when updating a wait-xxx database.
2321 PH/33 The hint "could be header name not terminated by colon", which has been
2322 given for certain expansion errors for a long time, was not being given
2323 for the ${if def:h_colon_omitted{... case.
2325 PH/34 The spec says: "With one important exception, whenever a domain list is
2326 being scanned, $domain contains the subject domain." There was at least
2327 one case where this was not true.
2329 PH/35 The error "getsockname() failed: connection reset by peer" was being
2330 written to the panic log as well as the main log, but it isn't really
2331 panic-worthy as it just means the connection died rather early on. I have
2332 removed the panic log writing for the ECONNRESET error when getsockname()
2335 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2336 runs only) independently of the message's sender address. This meant
2337 that, if the 4xx error was in fact related to the sender, a different
2338 message to the same recipient with a different sender could confuse
2339 things. In particualar, this can happen when sending to a greylisting
2340 server, but other circumstances could also provoke similar problems.
2341 I have changed the default so that the retry time for these errors is now
2342 based a combination of the sender and recipient addresses. This change
2343 can be overridden by setting address_retry_include_sender=false in the
2346 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2347 remote server are returned as part of bounce messages. This was not
2348 happening for LMTP over a pipe (the lmtp transport), but now it is the
2349 same for both kinds of LMTP.
2351 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2352 in header lines that were in fact CNAMEs. This is no longer the case.
2354 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2355 and queue runs started by the daemon processed all messages. This has
2356 been fixed so that -R and -S can now usefully be given with -q<time>.
2358 PH/40 Import PCRE release 6.7 (fixes some bugs).
2360 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2362 PH/42 Give an error if -q is specified more than once.
2364 PH/43 Renamed the variables $interface_address and $interface_port as
2365 $received_ip_address and $received_port, to make it clear that these
2366 values apply to message reception, and not to the outgoing interface when
2367 a message is delivered. (The old names remain recognized, of course.)
2369 PH/44 There was no timeout on the connect() call when using a Unix domain
2370 socket in the ${readsocket expansion. There now is.
2372 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2373 be meaningful with "accept".
2375 SC/01 Eximstats V1.43
2376 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2378 SC/02 Eximstats V1.44
2379 Use a glob alias rather than an array ref in the generated
2380 parser. This improves both readability and performance.
2382 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2383 Collect SpamAssassin and rejection statistics.
2384 Don't display local sender or destination tables unless
2385 there is data to show.
2386 Added average volumes into the top table text output.
2388 SC/04 Eximstats V1.46
2389 Collect data on the number of addresses (recipients)
2390 as well as the number of messages.
2392 SC/05 Eximstats V1.47
2393 Added 'Message too big' to the list of mail rejection
2394 reasons (thanks to Marco Gaiarin).
2396 SC/06 Eximstats V1.48
2397 Mainlog lines which have GMT offsets and are too short to
2398 have a flag are now skipped.
2400 SC/07 Eximstats V1.49 (Alain Williams)
2401 Added the -emptyok flag.
2403 SC/08 Eximstats V1.50
2404 Fixes for obtaining the IP address from reject messages.
2406 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2407 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2408 whitesspace changes from 4.64-PH/27
2410 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2413 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2414 are found, allow negative numbers in numeric criteria)
2416 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2418 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2419 to match changes made in 4.64-PH/43
2421 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2423 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2424 "domain" is actually an IP address, and force a failure. This locks out
2425 those revolvers/nameservers that support "A-for-A" lookups, in
2426 contravention of the specifications.
2428 PH/48 When a host name was looked up from an IP address, and the subsequent
2429 forward lookup of the name timed out, the host name was left in
2430 $sender_host_name, contrary to the specification.
2432 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2433 restricted to single-key lookups, Exim was not diagnosing an error if
2434 * or *@ was used with a query-style lookup.
2436 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2438 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2439 long ago, but noone interested enough thought of it. Let's just say that
2440 the "1.1" means that there are some new functions that weren't there at
2441 some point in the past.
2443 PH/51 Error processing for expansion failure of helo_data from an smtp
2444 transport during callout processing was broken.
2446 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2447 tested/used via the -bh/-bhc/-bs options.
2449 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2450 bug, fixed in subsequent PCRE releases).
2452 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2453 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2455 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2461 SC/01 Use a glob alias rather than an array ref in eximstats generated
2462 parser. This improves both readability and performance.
2464 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2465 Don't display local sender or destination tables in eximstats unless
2466 there is data to show.
2467 Added average volumes into the eximstats top table text output.
2469 SC/03 Collect data on the number of addresses (recipients) as well
2470 as the number of messages in eximstats.
2472 TF/01 Correct an error in the documentation for the redirect router. Exim
2473 does (usually) call initgroups() when daemonizing.
2475 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2476 with consistent privilege compared to when running as a daemon.
2478 TF/03 Note in the spec that $authenticated_id is not set for local
2479 submissions from trusted users.
2481 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2482 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2484 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2485 by adding some example configuration directives to the default
2486 configuration file. A little bit of work is required to uncomment the
2487 directives and define how usernames and passwords are checked, but
2488 there is now a framework to start from.
2490 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2491 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2492 without this. I don't know how relevant this is to other LDAP libraries.
2494 PH/02 Add the verb name to the "unknown ACL verb" error.
2496 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2498 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2500 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2501 directory not expanded when it should be if an expanded home directory
2502 was set for the address (which is overridden by the transport).
2504 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2507 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2508 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2509 because it is too late at that time, and has no effect.
2511 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2512 security issue with \' (bugzilla #107). I could not use the
2513 PQescapeStringConn() function, because it needs a PGconn value as one of
2516 PH/08 When testing addresses using -bt, indicate those final addresses that
2517 are duplicates that would not cause an additional delivery. At least one
2518 person was confused, thinking that -bt output corresponded to deliveries.
2519 (Suppressing duplicates isn't a good idea as you lose the information
2520 about possibly different redirections that led to the duplicates.)
2522 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2523 systems where poll() doesn't work, in particular OS X.
2525 PH/10 Added more information to debugging output for retry time not reached.
2527 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2528 operations in malware.c.
2530 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2533 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2534 syslog_duplication set false, log lines that would normally be written
2535 both the the main log and to the reject log were not written to syslog at
2538 PH/14 In the default configuration, change the use of "message" in ACL warn
2539 statements to "add_header".
2541 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2542 not followed by a command (e.g. "seen endif").
2544 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2545 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2548 PH/17 Added extra conditions to the default value of delay_warning_condition
2552 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2553 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2554 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2557 The Auto-Submitted: and various List- headers are standardised, whereas I
2558 don't think Precedence: ever was.
2560 PH/18 Refactored debugging code in route_finduser() to show more information,
2561 in particular, the error code if getpwnam() issues one.
2563 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2564 This is apparently needed in addition to the PH/07 change above to avoid
2565 any possible encoding problems.
2567 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2568 but not after initializing Perl.
2570 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2571 output them only if debugging. By default they are written stderr,
2572 apparently, which is not desirable.
2574 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2577 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2580 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2582 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2583 authenticated or an ident call has been made. Suppress the default
2584 values for $authenticated_id and $authenticated_sender (but permit -oMai
2585 and -oMas) when testing with -bh.
2587 PH/24 Re-jigged the order of the tests in the default configuration so that the
2588 tests for valid domains and recipients precede the DNS black list and CSA
2589 tests, on the grounds that those ones are more expensive.
2591 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2592 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2593 This bug exists in every version of Exim that I still have, right back to
2596 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2597 However, an attempt to turn on case-sensitivity in a regex key by
2598 including (?-i) didn't work because the subject string was already
2599 lowercased, and the effects were non-intuitive. It turns out that a
2600 one-line patch can be used to allow (?-i) to work as expected.
2606 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2607 other effects) broke the use of negated acl sub-conditions.
2609 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2612 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2613 "Deny" causes Exim to reject the incoming connection with a 554 error.
2614 Unfortunately, if there is a major crisis, such as a disk failure,
2615 tcp-wrappers gives "deny", whereas what one would like would be some
2616 kind of temporary error. A kludge has been added to help with this.
2617 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2618 554 error is used if errno is still zero or contains ENOENT (which occurs
2619 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2622 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2624 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2625 errors. Otherwise a message that provokes a temporary error (when other
2626 messages do not) can cause a whole host to time out.
2628 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2629 addresses were routed directly to files or pipes from a redirect router.
2630 File deliveries just didn't batch; pipe deliveries might have suffered
2633 PH/06 A failure to get a lock for a hints database would erroneously always say
2634 "Failed to get write lock", even when it was really a read lock.
2636 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2637 of 0600. This has been changed to use the value of the lockfile_mode
2638 option (which defaults to 0600).
2640 PH/08 Applied small patch from the Sieve maintainer.
2642 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2643 folder from quota calculations, a direct delivery into this folder messed
2644 up the contents of the maildirsize file. This was because the regex was
2645 used only to exclude .Trash (or whatever) when the size of the mailbox
2646 was calculated. There was no check that a delivery was happening into an
2647 excluded directory. This bug has been fixed by ignoring all quota
2648 processing for deliveries into excluded directories.
2650 PH/10 Added the maildirfolder_create_regex option to appendfile.
2656 PH/01 The code for finding all the local interface addresses on a FreeBSD
2657 system running IPv6 was broken. This may well have applied to all BSD
2658 systems, as well as to others that have similar system calls. The broken
2659 code found IPv4 interfaces correctly, but gave incorrect values for the
2660 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2661 that it would not match correctly against @[] and not recognize the IPv6
2664 PH/02 The ipliteral router was not recognizing addresses of the form user@
2665 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2667 PH/03 Added disable_ipv6.
2669 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2670 decoded form, because it is most often used to construct To: headers
2671 lines in autoreplies, and the decoded form may well be syntactically
2672 invalid. However, $reply_address has leading white space removed, and all
2673 newlines turned into spaces so that the autoreply transport does not
2676 PH/05 If group was specified without a user on a router, and no group or user
2677 was specified on a transport, the group from the router was ignored.
2679 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2680 for visible compile-time settings that can be used to change these
2681 numbers, for those that want even more. Backwards compatibility with old
2682 spool files has been maintained. However, going back to a previous Exim
2683 release will lost any variables that are in spool files.
2685 PH/07 Two small changes when running in the test harness: increase delay when
2686 passing a TCP/IP connection to a new process, in case the original
2687 process has to generate a bounce, and remove special handling of
2688 127.0.0.2 (sic), which is no longer necessary.
2690 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2691 be the same on different OS.
2693 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2696 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2697 whether --show-vars was specified or not
2699 JJ/02 exipick: Added support for new ACL variable spool format introduced
2702 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2703 syntactically invalid From: or Reply-to: line, and a filter used this to
2704 generate an autoreply, and therefore failed to obtain an address for the
2705 autoreply, Exim could try to deliver to a non-existent relative file
2706 name, causing unrelated and misleading errors. What now happens is that
2707 it logs this as a hard delivery error, but does not attempt to create a
2710 PH/11 The exinext utility has a -C option for testing purposes, but although
2711 the given file was scanned by exinext itself; it wasn't being passed on
2712 when Exim was called.
2714 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2715 an end-of-file indication when reading a command response.
2717 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2718 compiled. In many other places in Exim, IPv6 addresses are always
2719 recognized, so I have changed this. It also means that IPv4 domain
2720 literals of the form [IPV4:n.n.n.n] are now always recognized.
2722 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2723 used if the router is not running as root, for example, when verifying at
2724 ACL time, or when using -bh. The debugging output from this situation was
2725 non-existent - all you got was a failure to exec. I have made two
2728 (a) Failures to set uid/gid, the current directory, or a process leader
2729 in a subprocess such as that created by queryprogram now generate
2730 suitable debugging ouput when -d is set.
2732 (b) The queryprogram router detects when it is not running as root,
2733 outputs suitable debugging information if -d is set, and then runs
2734 the subprocess without attempting to change uid/gid.
2736 PH/15 Minor change to Makefile for building test_host (undocumented testing
2739 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2740 additional section of a DNS packet that returns MX or SRV records.
2741 Instead, it always explicitly searches for A/AAAA records. This avoids
2742 major problems that occur when a DNS server includes only records of one
2743 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2744 fixed another bug: if SRV records were looked up and the corresponding
2745 address records were *not* found in the additional section, the port
2746 values from the SRV records were lost.
2748 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2749 using the correct key (the original address) when searching the retry
2750 rules in order to find which one to use for generating the retry hint.
2752 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2753 adding the default one. Similarly, if it contains a Reply-To: header, the
2754 errors_reply_to option, if set, is not used.
2756 PH/19 When calculating a retry time, Exim used to measure the "time since
2757 failure" by looking at the "first failed" field in the retry record. Now
2758 it does not use this if it is later than than the arrival time of the
2759 message. Instead it uses the arrival time. This makes for better
2760 behaviour in cases where some deliveries succeed, thus re-setting the
2761 "first failed" field. An example is a quota failure for a huge message
2762 when small messages continue to be delivered. Without this change, the
2763 "time since failure" will always be short, possible causing more frequent
2764 delivery attempts for the huge message than are intended.
2765 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2767 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2768 $1, $2, $3) because the numerical variables can be reset during some
2769 expansion items (e.g. "match"), thereby losing the authentication data.
2771 PH/21 Make -bV show the size of off_t variables so that the test suite can
2772 decide whether to run tests for quotas > 2G.
2774 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2775 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2776 filecount value is greater than 2G or if a quota value is greater than 2G
2777 on a system where the size of off_t is not greater than 4, a panic error
2780 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2781 never match. The debug and -bh output now contains an explicit error
2782 message indicating a malformed IPv4 address or mask.
2784 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2785 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2786 PH/23 above applies.
2788 PH/25 Do not write to syslog when running in the test harness. The only
2789 occasion when this arises is a failure to open the main or panic logs
2790 (for which there is an explicit test).
2792 PH/26 Added the /no_tell option to "control=freeze".
2794 PH/27 If a host name lookup failed very early in a connection, for example, if
2795 the IP address matched host_lookup and the reverse lookup yielded a name
2796 that did not have a forward lookup, an error message of the form "no IP
2797 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2798 could be logged. Now it outputs the IP address instead of "NULL".
2800 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2801 allows the sender and the authenticated sender to be set when
2802 submitting a message from within Exim. Since child_open_exim() is
2803 documented for local_scan(), the new function should be too.
2805 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2806 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2807 results in an empty string is now treated as unset.
2809 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2811 PH/31 Added sender_verify_fail logging option.
2813 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2814 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2815 tidied the source and removed it altogether.
2817 PH/33 When a queue run was abandoned because the load average was too high, a
2818 log line was always written; now it is written only if the queue_run log
2819 selector is set. In addition, the log line for abandonment now contains
2820 information about the queue run such as the pid. This is always present
2821 in "start" and "stop" lines but was omitted from the "abandon" line.
2823 PH/34 Omit spaces between a header name and the colon in the error message that
2824 is given when verify = headers_syntax fails (if there are lots of them,
2825 the message gets confusing).
2827 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2828 names, as there are now some PTR records that contain slashes. This check
2829 is only to protect against broken name servers that fall over on strange
2830 characters, so the fact that it applies to all lookups doesn't matter.
2832 PH/36 Now that the new test suite is complete, we can remove some of the
2833 special code in Exim that was needed for the old test suite. For example,
2834 sorting DNS records because real resolvers return them in an arbitrary
2835 order. The new test suite's fake resolver always returns records in the
2838 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2839 bounces) except when queue_only is set, to avoid logging races between
2840 the different processes.
2842 PH/38 Panic-die if .include specifies a non-absolute path.
2844 PH/39 A tweak to the "H" retry rule from its user.
2846 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2847 a label. They prevented compilation on older perls.
2849 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2850 a warning to be raised on newish perls.
2852 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2853 on queue. Changes to match documented behaviour of showing count of
2854 messages matching specified criteria.
2856 PH/40 Changed the default ident timeout from 30s to 5s.
2858 PH/41 Added support for the use of login_cap features, on those BSD systems
2859 that have them, for controlling the resources used by pipe deliveries.
2861 PH/42 The content-scanning code uses fopen() to create files in which to put
2862 message data. Previously it was not paying any attention to the mode of
2863 the files. Exim runs with umask(0) because the rest of the code creates
2864 files with open(), and sets the required mode explicitly. Thus, these
2865 files were ending up world-writeable. This was not a big issue, because,
2866 being within the spool directory, they were not world-accessible. I have
2867 created a function called modefopen, which takes an additional mode
2868 argument. It sets umask(777), creates the file, chmods it to the required
2869 mode, then resets the umask. All the relevant calls to fopen() in the
2870 content scanning code have been changed to use this function.
2872 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2873 to 24 hours. This avoids potential overflow problems when processing G
2874 and H retry rules. I suspect nobody ever tinkers with this value.
2876 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2878 PH/45 When the plaintext authenticator is running as a client, the server's
2879 challenges are checked to ensure they are valid base64 strings. By
2880 default, the authentication attempt is cancelled if an invalid string is
2881 received. Setting client_ignore_invalid_base64 true ignores these errors.
2882 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2883 they are received. Thus, the responses can be made to depend on the
2884 challenges. If an invalid string is ignored, an empty string is placed in
2887 PH/46 Messages that are created by the autoreply transport now contains a
2888 References: header, in accordance with RFCs 2822 and 3834.
2890 PH/47 Added authenticated_sender_force to the smtp transport.
2892 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2894 PH/49 Installed latest patch from the Sieve maintainer.
2896 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2897 was also set, the appendfile transport was unnecessarily scanning a
2898 directory of message files (e.g. for maildir delivery) to find the count
2899 of files (along with the size), even though it did not need this
2900 information. It now does the scan only if it needs to find either the
2901 size of the count of files.
2903 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2905 PH/52 Two bugs concerned with error handling when the smtp transport is
2908 (i) Exim was not creating retry information for temporary errors given
2909 for individual recipients after the DATA command when the smtp transport
2910 was used in LMTP mode. This meant that they could be retried too
2911 frequently, and not timed out correctly.
2913 (ii) Exim was setting the flag that allows error details to be returned
2914 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2915 recipients that were returned after the DATA command.
2917 PH/53 This is related to PH/52, but is more general: for any failing address,
2918 when detailed error information was permitted to be returned to the
2919 sender, but the error was temporary, then after the final timeout, only
2920 "retry timeout exceeded" was returned. Now it returns the full error as
2921 well as "retry timeout exceeded".
2923 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2924 do this, and (what is worse) MTAs that accept it.
2926 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2927 will now be deprecated.
2929 PH/56 New os.c-cygwin from the Cygwin maintainer.
2931 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2932 formats (previously only available in exim formats via -bpr, -bpru,
2933 and -bpra. Now also available in native and exiqgrep formats)
2935 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2936 with very large, slow to parse queues
2938 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2940 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2942 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2943 responses to authentication challenges, though it was showing the
2944 challenges; (ii) I've removed the CR characters from the debug output for
2947 PH/58 Allow for the insertion of a newline as well as a space when a string
2948 is turned into more than one encoded-word during RFC 2047 encoding. The
2949 Sieve code now uses this.
2951 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2952 data_4xx, lost_connection, tls_required.
2954 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2955 message was being sent as an SMTP response.
2957 PH/61 Add -l and -k options to exicyclog.
2959 PH/62 When verifying, if an address was redirected to one new address, so that
2960 verification continued, and the new address failed or deferred after
2961 having set something in $address_data, the value of $address_data was not
2962 passed back to the ACL. This was different to the case when no
2963 redirection occurred. The value is now passed back in both cases.
2965 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2966 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2967 use login_cap.h, so on its own it isn't the distinguishing feature. The
2968 new name refers directly to the setclassresources() function.
2970 PH/65 Added configuration files for NetBSD3.
2972 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2974 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2975 is preferred over IPv4.
2977 PH/68 The bounce_return_message and bounce_return_body options were not being
2978 honoured for bounces generated during the reception of non-SMTP messages.
2979 In particular, this applied to messages rejected by the ACL. This bug has
2980 been fixed. However, if bounce_return_message is true and bounce_return_
2981 body is false, the headers that are returned for a non-SMTP message
2982 include only those that have been read before the error was detected.
2983 (In the case of an ACL rejection, they have all been read.)
2985 PH/69 The HTML version of the specification is now built in a directory called
2986 spec_html instead of spec.html, because the latter looks like a path with
2987 a MIME-type, and this confuses some software.
2989 PH/70 Catch two compiler warnings in sieve.c.
2991 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2992 function verify_get_ident() calls ip_connect() to connect a socket, but
2993 if the "connect()" function timed out, ip_connect() used to close the
2994 socket. However, verify_get_ident() also closes the socket later, and in
2995 between Exim writes to the log, which may get opened at this point. When
2996 the socket was closed in ip_connect(), the log could get the same file
2997 descriptor number as the socket. This naturally causes chaos. The fix is
2998 not to close the socket in ip_connect(); the socket should be closed by
2999 the function that creates it. There was only one place in the code where
3000 this was missing, in the iplookup router, which I don't think anybody now
3001 uses, but I've fixed it anyway.
3003 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3004 well as to direct DNS lookups. Otherwise the handling of names in host
3005 lists is inconsistent and therefore confusing.
3011 PH/01 Two changes to the default runtime configuration:
3013 (1) Move the checks for relay_from_hosts and authenticated clients from
3014 after to before the (commented out) DNS black list checks.
3016 (2) Add control=submission to the relay_from_hosts and authenticated
3017 clients checks, on the grounds that messages accepted by these
3018 statements are most likely to be submissions.
3020 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3022 (1) Generate an error if the third argument for the ${prvs expansion is
3025 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3028 (3) Reset the variables that are obtained from the first argument of
3029 ${prvscheck and used in the second argument before leaving the code,
3030 because their memory is reclaimed, so using them afterwards may do
3033 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3034 one (it's much easier than Tom thought :-).
3036 (5) Because of (4), we can now allow for the use of $prvscheck_result
3037 inside the third argument.
3039 PH/03 For some reason, the default setting of PATH when running a command from
3040 a pipe transport was just "/usr/bin". I have changed it to
3043 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3044 anything to be listed in the output from -bV.
3046 PH/05 When a filter generated an autoreply, the entire To: header line was
3047 quoted in the delivery log line, like this:
3049 => >A.N.Other <ano@some.domain> <original@ddress> ...
3051 This has been changed so that it extracts the operative address. There
3052 may be more than one such address. If so, they are comma-separated, like
3055 => >ano@some.domain,ona@other.domain <original@ddress> ...
3057 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3058 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3059 not being looked up in the rDNS to get a host name, Exim was showing the
3060 IP address twice in Received: lines, even though the IP addresses were
3061 identical. For example:
3063 Received: from [1.2.3.4] (helo=[1.2.3.4])
3065 However, if the real host name was known, it was omitting the HELO data
3066 if it matched the actual IP address. This has been tidied up so that it
3067 doesn't show the same IP address twice.
3069 PH/07 When both +timestamp and +memory debugging was on, the value given by
3070 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3071 called by the debug printing, thereby overwriting the timestamp buffer.
3072 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3074 PH/08 When the original message was included in an autoreply transport, it
3075 always said "this is a copy of the message, including all the headers",
3076 even if body_only or headers_only was set. It now gives an appropriate
3079 PH/09 Applied a patch from the Sieve maintainer which:
3081 o fixes some comments
3082 o adds the (disabled) notify extension core
3083 o adds some debug output for the result of if/elsif tests
3084 o points to the current vacation draft in the documentation
3085 and documents the missing references header update
3089 o fixes a bug in processing the envelope test (when testing
3090 multiple envelope elements, the last element determinted the
3093 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3094 Electronic Mail") by including:
3096 Auto-submitted: auto-generated
3098 in the messages that it generates (bounce messages and others, such as
3099 warnings). In the case of bounce messages for non-SMTP mesages, there was
3100 also a typo: it was using "Auto_submitted" (underscore instead of
3101 hyphen). Since every message generated by Exim is necessarily in response
3102 to another message, thes have all been changed to:
3104 Auto-Submitted: auto-replied
3106 in accordance with these statements in the RFC:
3108 The auto-replied keyword:
3110 - SHOULD be used on messages sent in direct response to another
3111 message by an automatic process,
3113 - MUST NOT be used on manually-generated messages,
3115 - MAY be used on Delivery Status Notifications (DSNs) and Message
3116 Disposition Notifications (MDNs),
3118 - MUST NOT be used on messages generated by automatic or periodic
3119 processes, except for messages which are automatic responses to
3122 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3123 to the default Received: header definition.
3125 PH/12 Added log selector acl_warn_skipped (default on).
3127 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3128 variables because (a) they are in the wrong storage pool and (b) even if
3129 they were copied, it wouldn't work properly because of the caching.
3131 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3132 checking when decoding. Apparently there are clients that generate
3133 overlong encoded strings. Why am I not surprised?
3135 PH/15 If the first argument of "${if match_address" was not empty, but did not
3136 contain an "@" character, Exim crashed. Now it writes a panic log message
3137 and treats the condition as false.
3139 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3141 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3142 extension "envelope-auth". The code is finished and in agreement with
3143 other implementations, but there is no documentation so far and in fact,
3144 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3145 not changing the active code.
3147 Print executed "if" and "elsif" statements when debugging is used. This
3148 helps a great deal to understand what a filter does.
3150 Document more things not specified clearly in RFC3028. I had all this
3151 sorted out, when out of a sudden new issues came to my mind. Oops."
3153 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3156 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3157 canonical form (as documented). However, after a host name lookup from
3158 the IP address, check_host() was doing a simple string comparison with
3159 addresses acquired from the DNS when checking that the found name did
3160 have the original IP as one of its addresses. Since any found IPv6
3161 addresses are likely to be in abbreviated form, the comparison could
3162 fail. Luckily, there already exists a function for doing the comparison
3163 by converting both addresses to binary, so now that is used instead of
3164 the text comparison.
3166 PH/20 There was another similar case to PH/19, when a complete host name was
3167 given in a host list; looking up its IP address could give an abbreviated
3168 form, whereas the current host's name might or might not be abbreviated.
3169 The same fix has been applied.
3175 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3176 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3179 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3180 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3182 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3184 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3185 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3186 Also, for "Auto-Submitted", treat anything other than "no" as
3187 non-personal, in accordance with RFC 3834. (Previously it treated
3188 anything starting "auto-" as non-personal.)
3190 TF/01 The control=submission/name=... option had a problem with syntax
3191 errors if the name included a slash character. The /name= option
3192 now slurps the rest of the string, so it can include any characters
3193 but it must come last in the list of options (after /sender_retain
3196 PH/05 Some modifications to the interface to the fake nameserver for the new
3204 TK/01 Added the "success_on_redirect" address verification option. See
3205 NewStuff for rationale and an example.
3207 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3209 PH/02 Patch to exigrep to allow it to work on syslog lines.
3211 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3212 fread() to skip over the body file's header line, because in Cygwin the
3213 header line is locked and is inaccessible.
3215 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3216 co-exist for some time) to make it clear that it is the Exim ID that is
3217 referenced, not the Message-ID: header line.
3219 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3220 string_format() function, because snprintf() does not exist on all
3223 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3224 ${stat: expansion item.
3226 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3227 protocol synchronization error", to keep the pedants happy.
3229 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3230 well as for IRIX systems, when gcc is being used. See the host.c source
3233 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3235 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3238 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3239 transport and to the smtp transport in LMTP mode.
3241 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3243 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3244 run a filter in a subprocess. This could lead to confusion in subsequent
3245 lookups in the parent process. There should also be a search_tidyup() at
3246 the end of the subprocess.
3248 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3249 only if the host matched helo_try_verify_hosts, which caused the
3250 verification to occur when the EHLO/HELO command was issued. The ACL just
3251 tested the remembered result. Now, if a previous verification attempt has
3252 not happened, "verify = helo" does it there and then.
3254 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3256 TK/03 Fix log output including CR from clamd.
3258 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3259 exist provoked a memory error which could cause a segfault.
3261 PH/15 Installed PCRE 6.2
3263 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3265 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3266 of the problem. Specifically, suggested +O2 rather than +O1 for the
3269 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3271 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3272 record, the port was not correctly logged when the outgoing_port log
3273 selector was set (it logged the transort's default port).
3275 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3276 fallback_hosts, and "hosts" in the smtp transport.
3278 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3279 host errors such as "Connection refused".
3281 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3282 authentication with radiusclient 0.4.9:
3284 - Error returned from rc_read_config was caught wrongly
3285 - Username/password not passed on to radius server due to wrong length.
3287 The presumption is that some radiusclient API changes for 4.51/PH/17
3288 were not taken care of correctly. The code is still untested by me (my
3289 Linux distribution still has 0.3.2 of radiusclient), but it was
3290 contributed by a Radius user.
3292 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3293 expanding the "port" option of the smtp transport.
3295 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3296 while reading a MIME header. Thanks to Tom Hughes for a patch.
3298 PH/24 Include config.h inside local_scan.h so that configuration settings are
3301 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3302 that in an ACL for RCPT (for example), you can examine exactly what was
3305 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3306 commands, but it was not correctly comparing the address with the actual
3307 client host address. Thus, it would show the EHLO address in Received:
3308 header lines when this was not necessary.
3310 PH/27 Added the % operator to ${eval:}.
3312 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3313 it should be ignoring failures (because with -C, for example, it has lost
3314 privilege). It wasn't ignoring creation failures other than "already
3317 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3320 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3321 because an input line was too long, either on its own, or by virtue of
3322 too many continuations, the temporary file was not being removed, and the
3323 return code was incorrect.
3325 PH/31 Missing "BOOL" in function definition in filtertest.c.
3327 PH/32 Applied Sieve patches from the maintainer.
3329 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3331 PH/33 Added "verify = not_blind".
3333 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3334 Local/Makefile (with some defaults set). These are used in built scripts
3335 such as exicyclog, but they have never been used in the exim_install
3336 script (though there are many overriding facilities there). I have
3337 arranged that the exim_install script now takes note of these two
3340 PH/35 Installed configuration files for Dragonfly.
3342 PH/36 When a locally submitted message by a trusted user did not contain a
3343 From: header, and the sender address was obtained from -f or from an SMTP
3344 MAIL command, and the trusted user did not use -F to supply a sender
3345 name, $originator_name was incorrectly used when constructing a From:
3346 header. Furthermore, $originator_name was used for submission mode
3347 messages from external hosts without From: headers in a similar way,
3348 which is clearly wrong.
3350 PH/37 Added control=suppress_local_fixups.
3352 PH/38 When log_selector = +received_sender was set, and the addition of the
3353 sender made the log line's construction buffer exactly full, or one byte
3354 less than full, an overflow happened when the terminating "\n" was
3357 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3358 when the result of a list match is failure because a DNS lookup failed.
3360 PH/40 RM_COMMAND is now used in the building process.
3362 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3363 the "build-* directories that it finds.
3365 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3366 address in a domain literal was a prefix of an interface address.
3368 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3369 when verifying a sender address, unless rewrite_headers is false.
3371 PH/44 Wrote a long comment about why errors_to addresses are verified as
3372 recipients, not senders.
3374 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3375 the ratelimit ACL was added.
3377 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3379 PH/46 Added extra information about PostgreSQL errors to the error string.
3381 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3382 avoiding the need to install special zones in a real server. This is
3383 backwards compatible; if it can't find the fake resolver, it drops back.
3384 Thus, both old and new test suites can be run.
3386 TF/02 Added util/ratelimit.pl
3388 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3389 clock is set back in time.
3391 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3392 Candler <B.Candler@pobox.com>.
3394 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3395 OK for addresses that are the result of redirections.
3397 PH/48 A number of further additions for the benefit of the new test suite,
3398 including a fake gethostbyname() that interfaces to the fake DNS resolver
3401 TF/06 The fix for widen_domains has also been applied to qualify_single and
3402 search_parents which are the other dnslookup options that can cause
3405 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3408 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3410 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3411 service" if it encountered an MX record with an empty target hostname.
3412 The message is now "an MX or SRV record indicated no SMTP service".
3414 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3415 if the DNS of the sending site is misconfigured. This is quite a
3416 common situation. This change restores the behaviour of treating a
3417 helo verification defer as a failure.
3419 PH/51 If self=fail was set on a router, the bounce message did not include the
3420 actual error message.
3426 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3428 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3429 command itself ended in error, the underlying message about the transport
3430 filter timeout was being overwritten with the pipe command error. Now the
3431 underlying error message should be appended to the second error message.
3433 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3435 PH/02 Reduce the amount of output that "make" produces by default. Full output
3436 can still be requested.
3438 PH/03 The warning log line about a condition test deferring for a "warn" verb
3439 was being output only once per connection, rather than after each
3440 occurrence (because it was using the same function as for successful
3441 "warn" verbs). This seems wrong, so I have changed it.
3443 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3444 it should not have, which might have caused a crash in the right
3445 circumstances, but probably never did.
3447 PH/04 Installed a modified version of Tony Finch's patch to make submission
3448 mode fix the return path as well as the Sender: header line, and to
3449 add a /name= option so that you can make the user's friendly name appear
3452 TF/03 Added the control = fakedefer ACL modifier.
3454 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3455 Mark Lowes for thorough testing.
3457 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3459 TK/03 Merged latest SRS patch from Miles Wilton.
3461 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3462 with the definition in sysexits.h (which is #included earlier).
3463 Fortunately, Exim does not actually use EX_OK. The code used to try to
3464 preserve the sysexits.h value, by assumimg that macro definitions were
3465 scanned for macro replacements. I have been disabused of this notion,
3466 so now the code just undefines EX_OK before #including unistd.h.
3468 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3469 in the smtp transport. When a block could not be written in a single
3470 write() function, the timeout was being re-applied to each part-write.
3471 This seems wrong - if the receiver was accepting one byte at a time it
3472 would take for ever. The timeout is now adjusted when this happens. It
3473 doesn't have to be particularly precise.
3475 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3476 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3478 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3479 without a domain if the check to <postmaster@domain> fails.
3481 SC/01 Eximstats: added -xls and the ability to specify output files
3482 (patch written by Frank Heydlauf).
3484 SC/02 Eximstats: use FileHandles for outputing results.
3486 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3488 SC/04 Eximstats: fixed display of large numbers with -nvr option
3490 SC/05 Eximstats: fixed merging of reports with empty tables.
3492 SC/06 Eximstats: added the -include_original_destination flag
3494 SC/07 Eximstats: removed tabs and trailing whitespace.
3496 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3498 TK/06 MBOX spool code: Add real "From " MBOX separator line
3499 so the .eml file is really in mbox format (even though
3500 most programs do not really care). Patch from Alex Miller.
3502 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3503 The latter is generated from $received_to and is only set if the
3504 message has one envelope recipient. SA can use these headers,
3505 obviously out-of-the-box. Patch from Alex Miller.
3507 PH/08 The ${def test on a variable was returning false if the variable's
3508 value was "0", contrary to what the specification has always said!
3509 The result should be true unless the variable is empty.
3511 PH/09 The syntax error of a character other than { following "${if
3512 def:variable_name" (after optional whitespace) was not being diagnosed.
3513 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3514 accidental colon was present, for example, could give incorrect results.
3516 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3517 result is used (not including appendfile, where other changes are about
3520 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3521 This involved changing a lot of size variables from int to off_t. It
3522 should work with maildirs and everything.
3524 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3525 spamd dying while we are connected to it.
3527 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3530 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3531 function for BDB 4.3.
3533 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3535 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3536 the sources for PCRE have been re-arranged and more files are now
3539 PH/15 The code I had for printing potentially long long variables in PH/11
3540 above was not the best (it lost precision). The length of off_t variables
3541 is now inspected at build time, and an appropriate printing format (%ld
3542 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3543 to be "long long int" or "long int". This is needed for the internal
3544 formatting function string_vformat().
3546 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3547 the configuration file to be ":syslog", then the script "guesses" where
3548 the logs files are, rather than using the compiled in default. In our
3549 case the guess is not the same as the compiled default, so the script
3550 suddenly stopped working when I started to use syslog. The patch checks
3551 to see if log_file_path is "". If so, it attempts to read it from exim
3552 with no configuration file to get the compiled in version, before it
3553 falls back to the previous guessing code."
3555 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3556 implementing BATV in an Exim configuration. See NewStuff for the gory
3559 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3560 Makefile that are specific to HP-UX.
3562 PH/18 If the "use_postmaster" option was set for a recipient callout together
3563 with the "random" option, the postmaster address was used as the MAIL
3564 FROM address for the random test, but not for the subsequent recipient
3565 test. It is now used for both.
3567 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3568 patch removes a few documentation additions to RFC 3028, because the
3569 latest draft now contains them. It adds the new en;ascii-case comparator
3570 and a new error check for 8bit text in MIME parts. Comparator and
3571 require names are now matched exactly. I enabled the subaddress
3572 extension, but it is not well tested yet (read: it works for me)."
3574 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3575 rework some of the code of TK/09 above to avoid the hardwired use of
3576 "%lld" and "long long". Replaced the call to snprintf() with a call to
3579 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3580 records point to non-existent hosts", "retry timeout exceeded", and
3581 "retry time not reached for any host after a long failure period".
3583 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3584 experimental DomainKeys support:
3586 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3587 (2) On an error such as an illegally used "control", the wrong name for
3588 the control was given.
3590 These problems did NOT occur unless DomainKeys support was compiled.
3592 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3594 PH/24 Added ${if match_ip condition.
3596 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3597 databases so that it will be absolutely obvious if a crash occurs in the
3598 DB library. This is a regular occurrence (often caused by mis-matched
3601 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3602 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3603 picked up on a user's system that detects such things. There doesn't seem
3604 to be a gcc warning option for this - only an attribute that has to be
3605 put on the function's prototype. It seems that in Fedora Core 4 they have
3606 set this on a number of new functions. No doubt there will be more in due
3609 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3610 specify a transport. However, if an address that was verified by such a
3611 router was the subject of a callout, Exim crashed because it tried to
3612 read the rcpt_include_affixes from the non-existent transport. Now it
3613 just assumes that the setting of that option is false. This bug was
3614 introduced by 4.51/PH/31.
3616 PH/28 Changed -d+all to exclude +memory, because that information is very
3617 rarely of interest, but it makes the output a lot bigger. People tend to
3618 do -d+all out of habit.
3620 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3621 code in os-type was giving problems when libc.so lives in lib64, like on
3624 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3625 aren't the modern standard, and it seems that some systems' include files
3626 don't always have them. Exim was already checking for some of the newer
3627 ones like T_AAAA, and defining it itself. I've added checks for all the
3628 record types that Exim uses.
3630 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3631 not automatically generating a new one, as it is supposed to. This
3632 prevented TLS from working. If the file did exist, but contained invalid
3633 data, a new version was generated, as expected. It was only the case of a
3634 non-existent file that was broken.
3636 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3637 with a change in libdomainkeys > 0.64.
3639 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3640 from DNS. If the selector record carries the flag, it now has
3641 precedence over the domain-wide flag.
3643 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3645 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3646 the use of a port name that isn't defined in /etc/services) occurred, the
3647 message was deferred as in a normal delivery, and thus remained on the
3648 spool, instead of being failed because of the mua_wrapper setting. This
3649 is now fixed, and I tidied up some of the mua_wrapper messages at the
3652 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3653 the messages in a hash of arrays rather than using individual hashes.
3654 This is a bit cleaner and results in dramatic memory savings, albeit
3655 at a slight CPU cost.
3657 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3658 as requested by Marc Sherman.
3660 SC/10 Eximstats: added histograms for user specified patterns as requested
3663 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3665 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3666 fopen() in the content-scanning modules that did not already have it.
3672 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3673 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3675 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3677 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3679 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3680 to test Sieve filters that use "vacation".
3682 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3683 that changes the way the GnuTLS parameters are stored in the cache file.
3684 The new format can be generated externally. For backward compatibility,
3685 if the data in the cache doesn't make sense, Exim assumes it has read an
3686 old-format file, and it generates new data and writes a new file. This
3687 means that you can't go back to an older release without removing the
3690 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3691 work if there are any delivery delays because "one_time" forces the
3692 parent to be marked "delivered", so its unseen clone is never tried
3693 again. For this reason, Exim now forbids the simultaneous setting of
3696 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3697 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3698 entry if you want to know the details. The fix, however, neglected to
3699 consider the case when local delivery batching is involved. The test for
3700 "previously delivered" was not happening when checking to see if an
3701 address could be batched with a previous (undelivered) one; under
3702 certain circumstances this could lead to multiple deliveries to the same
3705 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3706 in its include files, and this causes problems building Exim.
3708 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3709 header_syntax) but Exim was just ignoring anything given after a slash.
3710 In particular, this caused confusion with an attempt to use "verify =
3711 reverse_host_lookup/defer_ok". An error is now given when options are
3712 supplied for verify items that do not have them. (Maybe reverse_host_
3713 lookup should have a defer_ok option, but that's a different point.)
3715 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3716 defined by RFC 821) to 2048, because there were problems with some AUTH
3717 commands, and RFC 1869 says the size should be increased for extended
3718 SMTP commands that take arguments.
3720 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3723 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3724 "unknown" error; now it says that the functionality isn't in the binary.
3726 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3727 an address' error message when a string expansion fails (syntax or
3728 whatever). Otherwise the password may appear in the log. Following change
3729 PH/42 below, there is no longer a chance of it appearing in a bounce
3732 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3734 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3735 of its addresses was ever tried. (Bugzilla bug #2.)
3737 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3738 the result incorrectly in the debug output. (It correctly added a newline
3739 to what was transported.)
3741 TF/01 Added $received_time.
3743 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3744 commented out examples of how to interface to a virus scanner and to
3745 SpamAssassin. Also added commented examples of av_scanner and
3746 spamd_address settings.
3748 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3749 and controls are allowed in which ACLs. There were a couple of minor
3750 errors. Some of the entries in the conditions table (which is a table of
3751 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3752 negation of where the condition IS allowed.
3754 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3756 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3757 header file does not have a version number, so I've had to invent a new
3758 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3759 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3760 radiusclient), but it was contributed by a Radius user.
3762 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3763 files or directories, for OpenSSL.
3765 PH/19 When an Exim process that is running as root has to create an Exim log
3766 file, it does so in a subprocess that runs as exim:exim so as to get the
3767 ownership right at creation (otherwise, other Exim processes might see
3768 the file with the wrong ownership). There was no test for failure of this
3769 fork() call, which would lead to the process getting stuck as it waited
3770 for a non-existent subprocess. Forks do occasionally fail when resources
3771 run out. I reviewed all the other calls to fork(); they all seem to check
3774 PH/20 When checking for unexpected SMTP input at connect time (before writing
3775 the banner), Exim was not dealing correctly with a non-positive return
3776 from the read() function. If the client had disconnected by this time,
3777 the result was a log entry for a synchronization error with an empty
3778 string after "input=" when read() returned zero. If read() returned -1
3779 (an event I could not check), uninitialized data bytes were printed.
3780 There were reports of junk text (parts of files, etc) appearing after
3783 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3785 PH/22 Added support for macro redefinition, and (re)definition in between
3786 driver and ACL definitions.
3788 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3789 forgetting to use the resulting value; it was using the unexpanded value.
3791 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3792 hadn't been configured. The fix is from Juergen Kreileder, who
3793 understands it better than I do:
3795 "Here's what I see happening with three configured cyrus_sasl
3796 authenticators configured (plain, login, cram-md5):
3798 On startup auth_cyrus_sasl_init() gets called for each of these.
3799 This means three calls to sasl_listmech() without a specified mech_list.
3800 => SASL tests which mechs of all available mechs actually work
3801 => three warnings about OTP not working
3802 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3804 With the patch, sasl_listmech() also gets called three times. But now
3805 SASL's mech_list option is set to the server_mech specified in the the
3806 authenticator. Or in other words, the answer from sasl_listmech()
3807 gets limited to just the mech you're testing for (which is different
3809 => the return list contains just 'plain' or 'login', 'cram-md5' or
3810 nothing depending on the value of ob->server_mech.
3812 I've just tested the patch: Authentication still works fine,
3813 unavailable mechs specified in the exim configuration are still
3814 caught, and the auth.log warnings about OTP are gone."
3816 PH/25 When debugging is enabled, the contents of the command line are added
3817 to the debugging output, even when log_selector=+arguments is not
3820 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3821 answer is "GNU", and only if the return is "GNU/something" is the answer
3824 PH/27 $acl_verify_message is now set immediately after the failure of a
3825 verification in an ACL, and so is available in subsequent modifiers. In
3826 particular, the message can be preserved by coding like this:
3828 warn !verify = sender
3829 set acl_m0 = $acl_verify_message
3831 Previously, $acl_verify_message was set only while expanding "message"
3832 and "log_message" when a very denied access.
3834 PH/28 Modified OS/os.c-Linux with
3836 -#ifndef OS_LOAD_AVERAGE
3837 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3839 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3840 nomenclature these days.)
3842 PH/29 Installed patch from the Sieve maintainer that adds the options
3843 sieve_useraddress and sieve_subaddress to the redirect router.
3845 PH/30 In these circumstances:
3846 . Two addresses routed to the same list of hosts;
3847 . First host does not offer TLS;
3848 . First host accepts first address;
3849 . First host gives temporary error to second address;
3850 . Second host offers TLS and a TLS session is established;
3851 . Second host accepts second address.
3852 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3853 and peerdn, if requested) that were in fact used only for the second
3856 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3857 attention to any local part prefix or suffix that was matched by the
3858 router that accepted the address. It now behaves in the same way as it
3859 does for delivery: the affixes are removed from the local part unless
3860 rcpt_include_affixes is set on the transport.
3862 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3863 timeout during the DATA phase of an incoming message.
3865 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3866 applied a patch sanctioned by the Sieve maintainer.
3868 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3869 the uid or gid is negative. A case of a negative gid caused this to be
3870 noticed. The fix allows for either to be negative.
3872 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3873 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3876 PH/36 The change PH/12 above was broken. Fixed it.
3878 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3879 the grounds that routing the same address twice would always produce the
3880 same answer. This might have been true once, but it is certainly no
3881 longer true now. Routing a child address may depend on the previous
3882 routing that produced that child. Some complicated redirection strategies
3883 went wrong when messages had multiple recipients, and made Exim's
3884 behaviour dependent on the order in which the addresses were given.
3886 I have moved the duplicate checking until after the routing is complete.
3887 Exim scans the addresses that are assigned to local and remote
3888 transports, and removes any duplicates. This means that more work will be
3889 done, as duplicates will always all be routed, but duplicates are
3890 presumably rare, so I don't expect this is of any significance.
3892 For deliveries to pipes, files, and autoreplies, the duplicate checking
3893 still happens during the routing process, since they are not going to be
3896 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3897 It corrects a timeout issue with spamd. This is Ian's comment: "The
3898 background is that sometimes spamd either never reads data from a
3899 connection it has accepted, or it never writes response data. The exiscan
3900 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3901 blindly assumes that writes won't block so it may never time out."
3903 PH/39 Allow G after quota size as well as K and M.
3905 PH/40 The value set for $authenticated_id in an authenticator may not contain
3906 binary zeroes or newlines because the value is written to log lines and
3907 to spool files. There was no check on this. Now the value is run through
3908 the string_printing() function so that such characters are converted to
3909 printable escape sequences.
3911 PH/41 $message_linecount is a new variable that contains the total number of
3912 lines in the message. Compare $body_linecount, which is the count for the
3915 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3916 bounce and delay warning messages, except in certain special cases, which
3919 (a) An SMTP error message from a remote host;
3920 (b) A message specified in a :fail: redirection;
3921 (c) A message specified in a "fail" command in a system filter;
3922 (d) A message specified in a FAIL return from the queryprogram router;
3923 (e) A message specified by the cannot_route_message router option.
3925 In these cases only, Exim does include the error details in bounce and
3926 warning messages. There are also a few cases where bland messages such
3927 as "unrouteable address" or "local delivery error" are given.
3929 PH/43 $value is now also set for the "else" part of a ${run expansion.
3931 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3932 being worked on, but at least Exim now implements the latest version to
3935 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3936 process to complete was treated as a delivery failure, a timeout while
3937 writing the message to the pipe was logged, but erroneously treated as a
3938 successful delivery. Such timeouts include transport filter timeouts. For
3939 consistency with the overall process timeout, these timeouts are now
3940 treated as errors, giving rise to delivery failures by default. However,
3941 there is now a new Boolean option for the pipe transport called
3942 timeout_defer, which, if set TRUE, converts the failures into defers for
3943 both kinds of timeout. A transport filter timeout is now identified in
3946 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3947 systems where "make" and "gmake" are different, calling "gmake" at top
3948 level broke things. I've arranged for the value of $(MAKE) to be passed
3949 from the Makefile to this script so that it can call the same version of
3953 A note about Exim versions 4.44 and 4.50
3954 ----------------------------------------
3956 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3957 changes of various kinds. As a consequence, a big documentation update was
3958 needed. This delayed the release for rather longer than seemed good, especially
3959 in the light of a couple of (minor) security issues. Therefore, the changes
3960 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3961 release. So 4.44 and 4.50 are in effect two different branches that both start
3964 I have left the 4.50 change log unchanged; it contains all the changes since
3965 4.43. The change log for 4.44 is below; many of its items are identical to
3966 those for 4.50. This seems to be the most sensible way to preserve the
3967 historical information.
3973 1. Minor wording change to the doc/README.SIEVE file.
3975 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3976 computation of the current number of files was incorrect.
3978 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3979 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3982 4. Give more explanation in the error message when the command for a transport
3983 filter fails to execute.
3985 5. There are several places where Exim runs a non-Exim command in a
3986 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3987 was being done only for the command run by the queryprogram router. It is
3988 now done for all such subprocesses. The other cases are: ${run, transport
3989 filters, and the commands run by the lmtp and pipe transports.
3991 6. Added CONFIGURE_GROUP build-time option.
3993 7. Some older OS have a limit of 256 on the maximum number of file
3994 descriptors. Exim was using setrlimit() to set 1000 as a large value
3995 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3996 systems. I've change it so that if it can't get 1000, it tries for 256.
3998 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3999 was an oversight, and furthermore, ever since the addition of extra
4000 controls (e.g. 4.43/32), the checks on when to allow different forms of
4001 "control" were broken. There should now be diagnostics for all cases when a
4002 control that does not make sense is encountered.
4004 9. Added the /retain_sender option to "control=submission".
4006 10. $recipients is now available in the predata ACL (oversight).
4008 11. Tidy the search cache before the fork to do a delivery from a message
4009 received from the command line. Otherwise the child will trigger a lookup
4010 failure and thereby defer the delivery if it tries to use (for example) a
4011 cached ldap connection that the parent has called unbind on.
4013 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4014 of $address_data from the recipient verification was clobbered by the
4015 sender verification.
4017 13. The value of address_data from a sender verification is now available in
4018 $sender_address_data in subsequent conditions in the ACL statement.
4020 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4022 15. Added a new option "connect=<time>" to callout options, to set a different
4025 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4026 was its contents. (It was OK if the option was not defined at all.)
4028 17. A "Completed" log line is now written for messages that are removed from
4029 the spool by the -Mrm option.
4031 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4032 information about exactly what failed.
4034 19. Added -dd to debug only the daemon process.
4036 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4037 handles timeouts, both on the server side and network timeouts. Renamed the
4038 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4040 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4041 It is now set to "smtps".
4043 22. $host_address is now set to the target address during the checking of
4044 ignore_target_hosts.
4046 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4047 being passed; this would have caused $sender_host_name to have been used if
4048 matching the list had actually called for a host name (not very likely,
4049 since this list is usually IP addresses). A host name is now passed as
4052 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4053 code that specifies a non-restarting handler (typically sigaction() in
4054 modern systems) in an attempt to fix a rare and obscure crash bug.
4056 25. Narrowed the window for a race in the daemon that could cause it to ignore
4057 SIGCHLD signals. This is not a major problem, because they are used only to
4058 wake it up if nothing else does.
4060 26. A malformed maildirsize file could cause Exim to calculate negative values
4061 for the mailbox size or file count. Odd effects could occur as a result.
4062 The maildirsize information is now recalculated if the size or filecount
4065 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4066 support for a long time. Removed HAVE_SYS_VFS_H.
4068 28. Installed the latest version of exipick from John Jetmore.
4070 29. In an address list, if the pattern was not a regular expression, an empty
4071 subject address (from a bounce message) matched only if the pattern was an
4072 empty string. Non-empty patterns were not even tested. This was the wrong
4073 because it is perfectly reasonable to use an empty address as part of a
4074 database query. An empty address is now tested by patterns that are
4075 lookups. However, all the other forms of pattern expect the subject to
4076 contain a local part and a domain, and therefore, for them, an empty
4077 address still always fails if the pattern is not itself empty.
4079 30. Exim went into a mad DNS loop when attempting to do a callout where the
4080 host was specified on an smtp transport, and looking it up yielded more
4081 than one IP address.
4083 31. Re-factored the code for checking spool and log partition space into a
4084 function that finds that data and another that does the check. The former
4085 is then used to implement four new variables: $spool_space, $log_space,
4086 $spool_inodes, and $log_inodes.
4088 32. The RFC2047 encoding function was originally intended for short strings
4089 such as real names; it was not keeping to the 75-character limit for
4090 encoded words that the RFC imposes. It now respects the limit, and
4091 generates multiple encoded words if necessary. To be on the safe side, I
4092 have increased the buffer size for the ${rfc2047: expansion operator from
4095 33. It is now permitted to omit both strings after an "if" condition; if the
4096 condition is true, the result is "true". As before, when the second string
4097 is omitted, a false condition yields an empty string. This makes it less
4098 cumbersome to write custom ACL and router conditions.
4100 34. Failure to deliver a bounce message always caused it to be frozen, even if
4101 there was an errors_to setting on the router. The errors_to setting is now
4104 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4105 canonical form (fully expanded) before being placed in
4106 $sender_host_address.
4108 36. The table in the code that translates DNS record types into text (T_A to
4109 "A" for instance) was missing entries for NS and CNAME. It is just possible
4110 that this could have caused confusion if both these types were looked up
4111 for the same domain, because the text type is used as part of Exim's
4112 per-process caching. But the chance of anyone hitting this buglet seems
4115 37. The dnsdb lookup has been extended in a number of ways.
4117 (1) There is a new type, "zns", which walks up the domain tree until it
4118 finds some nameserver records. It should be used with care.
4120 (2) There is a new type, "mxh", which is like "mx" except that it returns
4121 just the host names, not the priorities.
4123 (3) It is now possible to give a list of domains (or IP addresses) to be
4124 looked up. The behaviour when one of the lookups defers can be
4125 controlled by a keyword.
4127 (4) It is now possible to specify the separator character for use when
4128 multiple records are returned.
4130 38. The dnslists ACL condition has been extended: it is now possible to supply
4131 a list of IP addresses and/or domains to be looked up in a particular DNS
4134 39. Added log_selector=+queue_time_overall.
4136 40. When running the queue in the test harness, wait just a tad after forking a
4137 delivery process, to get repeatability of debugging output.
4139 41. Include certificate and key file names in error message when GnuTLS fails
4140 to set them up, because the GnuTLS error message doesn't include the name
4141 of the failing file when there is a problem reading it.
4143 42. Allow both -bf and -bF in the same test run.
4145 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4147 44. The "Exiscan patch" is now merged into the mainline Exim source.
4149 45. Sometimes the final signoff response after QUIT could fail to get
4150 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4151 < 0 before doing a fflush(). This bug looks as though it goes back to the
4152 introduction of TLS in release 3.20, but "sometimes" must have been rare
4153 because the tests only now provoked it.
4155 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4156 (this can affect the format of dates).
4158 47. exim_tidydb, when checking for the continued existence of a message for
4159 which it has found a message-specific retry record, was not finding
4160 messages that were in split spool directories. Consequently, it was
4161 deleting retry records that should have stayed in existence.
4163 48. Steve fixed some bugs in eximstats.
4165 49. The SPA authentication driver was not abandoning authentication and moving
4166 on to the next authenticator when an expansion was forced to fail,
4167 contradicting the general specification for all authenticators. Instead it
4168 was generating a temporary error. It now behaves as specified.
4170 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4171 (the order specifies the preference for clients). The order is now AES256,
4172 AES128, 3DES, ARCFOUR128.
4174 51. Small patch to Sieve code - explicitly set From: when generating an
4177 52. Exim crashed if a remote delivery caused a very long error message to be
4178 recorded - for instance if somebody sent an entire SpamAssassin report back
4179 as a large number of 550 error lines. This bug was coincidentally fixed by
4180 increasing the size of one of Exim's internal buffers (big_buffer) that
4181 happened as part of the Exiscan merge. However, to be on the safe side, I
4182 have made the code more robust (and fixed the comments that describe what
4185 53. Now that there can be additional text after "Completed" in log lines (if
4186 the queue_time_overall log selector is set), a one-byte patch to exigrep
4187 was needed to allow it to recognize "Completed" as not the last thing in
4190 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4191 patch that reportedly fixes this has been added. I am not expert enough to
4192 create a test for it. This is what the patch creator wrote:
4194 "I found a little strange behaviour of ldap code when working with
4195 Windows 2003 AD Domain, where users was placed in more than one
4196 Organization Units. When I tried to give exim partial DN, the exit code
4197 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4198 But simultaneously result of request was absolutely normal ldap result,
4199 so I produce this patch..."
4201 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4202 REFERENCE, so I have modified the code to exclude the patch when that macro
4205 55. Some experimental protocols are using DNS PTR records for new purposes. The
4206 keys for these records are domain names, not reversed IP addresses. The
4207 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4208 leaves it alone. Component reversal etc. now happens only for IP addresses.
4211 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4213 57. Double the size of the debug message buffer (to 2048) so that more of very
4214 long debug lines gets shown.
4216 58. The exicyclog utility now does better if the number of log files to keep
4217 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4219 59. Two changes related to the smtp_active_hostname option:
4221 (1) $smtp_active_hostname is now available as a variable.
4222 (2) The default for smtp_banner uses $smtp_active_hostname instead
4223 of $primary_hostname.
4225 60. The host_aton() function is supposed to be passed a string that is known
4226 to be a valid IP address. However, in the case of IPv6 addresses, it was
4227 not checking this. This is a hostage to fortune. Exim now panics and dies
4228 if the condition is not met. A case was found where this could be provoked
4229 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4230 components; fortuitously, this particular loophole had already been fixed
4231 by change 4.50/55 above.
4233 If there are any other similar loopholes, the new check in host_aton()
4234 itself should stop them being exploited. The report I received stated that
4235 data on the command line could provoke the exploit when Exim was running as
4236 exim, but did not say which command line option was involved. All I could
4237 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4238 running as the user.
4241 61. There was a buffer overflow vulnerability in the SPA authentication code
4242 (which came originally from the Samba project). I have added a test to the
4243 spa_base64_to_bits() function which I hope fixes it.
4246 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4247 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4249 63. The daemon start-up calls getloadavg() while still root for those OS that
4250 need the first call to be done as root, but it missed one case: when
4251 deliver_queue_load_max is set with deliver_drop_privilege. This is
4252 necessary for the benefit of the queue runner, because there is no re-exec
4253 when deliver_drop_privilege is set.
4255 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4256 This has been fixed.
4258 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4259 was in use, was not putting the data itself into the right store pool;
4260 consequently, it could be overwritten for a subsequent message in the same
4261 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4264 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4266 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4267 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4268 the function were treating the return as a boolean value, which happened to
4269 work because 0=false and not-0=true, but is not correct code.
4271 68. The host_aton() function was not handling scoped IPv6 addresses (those
4272 with, for example, "%eth0" on the end) correctly.
4274 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4275 negated items (that is, ~something) in unsigned ints. Some compilers
4276 apparently mutter when there is no cast.
4278 70. If an address verification called from an ACL failed, and did not produce a
4279 user-specific message (i.e. there was only a "system" message), nothing was
4280 put in $acl_verify_message. In this situation, it now puts the system
4283 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4284 session; change 4.31/43 added the unwanted input to the log line - except
4285 that it did not do this in the start of session case. It now does.
4287 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4288 This is wrong and can cause the other end to generate a synchronization
4289 error if it is another Exim or anything else that does the synchronization
4290 check. A QUIT command is no longer sent after a timeout.
4292 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4293 during host lookups.
4295 74. The defer_ok option of callout verification was not working if it was used
4296 when verifying addresses in header lines, that is, for this case:
4298 verify = header_sender/callout=defer_ok
4300 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4301 those file descriptors could be used for SMTP connections. If anything
4302 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4303 could be sent to the SMTP client, causing chaos. The daemon now opens
4304 stdin, stdout, and stderr to /dev/null when it puts itself into the
4307 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4308 log by default. The user can override this with suitable Perl magic.
4310 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4311 the log message when discard triggers, was not working for the DATA ACL or
4312 for the non-SMTP ACL.
4314 78. Error message wording change in sieve.c.
4316 79. If smtp_accept_max_per_host was set, the number of connections could be
4317 restricted to fewer than expected, because the daemon was trying to set up
4318 a new connection before checking whether the processes handling previous
4319 connections had finished. The check for completed processes is now done
4320 earlier. On busy systems, this bug wouldn't be noticed because something
4321 else would have woken the daemon, and it would have reaped the completed
4324 80. If a message was submitted locally by a user whose login name contained one
4325 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4326 It caused a spool format error. I have fixed the spool reading code. A
4327 related problem was that the "from" clause in the Received: line became
4328 illegal because of the space(s). It is now covered by ${quote_local_part.
4330 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4333 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4335 83. Patch from Sieve maintainer to fix unterminated string problem in
4336 "vacation" handling.
4338 84. Some minor changes to the Linux configuration files to help with other
4339 OS variants using glibc.
4341 85. One more patch for Sieve to update vacation handling to latest spec.
4344 ----------------------------------------------------
4345 See the note above about the 4.44 and 4.50 releases.
4346 ----------------------------------------------------
4352 1. Change 4.43/35 introduced a bug that caused file counts to be
4353 incorrectly computed when quota_filecount was set in an appendfile
4356 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4357 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4360 3. Give more explanation in the error message when the command for a transport
4361 filter fails to execute.
4363 4. There are several places where Exim runs a non-Exim command in a
4364 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4365 was being done only for the command run by the queryprogram router. It is
4366 now done for all such subprocesses. The other cases are: ${run, transport
4367 filters, and the commands run by the lmtp and pipe transports.
4369 5. Some older OS have a limit of 256 on the maximum number of file
4370 descriptors. Exim was using setrlimit() to set 1000 as a large value
4371 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4372 systems. I've change it so that if it can't get 1000, it tries for 256.
4374 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4375 was an oversight, and furthermore, ever since the addition of extra
4376 controls (e.g. 4.43/32), the checks on when to allow different forms of
4377 "control" were broken. There should now be diagnostics for all cases when a
4378 control that does not make sense is encountered.
4380 7. $recipients is now available in the predata ACL (oversight).
4382 8. Tidy the search cache before the fork to do a delivery from a message
4383 received from the command line. Otherwise the child will trigger a lookup
4384 failure and thereby defer the delivery if it tries to use (for example) a
4385 cached ldap connection that the parent has called unbind on.
4387 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4388 of $address_data from the recipient verification was clobbered by the
4389 sender verification.
4391 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4392 was its contents. (It was OK if the option was not defined at all.)
4394 11. A "Completed" log line is now written for messages that are removed from
4395 the spool by the -Mrm option.
4397 12. $host_address is now set to the target address during the checking of
4398 ignore_target_hosts.
4400 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4401 being passed; this would have caused $sender_host_name to have been used if
4402 matching the list had actually called for a host name (not very likely,
4403 since this list is usually IP addresses). A host name is now passed as
4406 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4407 code that specifies a non-restarting handler (typically sigaction() in
4408 modern systems) in an attempt to fix a rare and obscure crash bug.
4410 15. Narrowed the window for a race in the daemon that could cause it to ignore
4411 SIGCHLD signals. This is not a major problem, because they are used only to
4412 wake it up if nothing else does.
4414 16. A malformed maildirsize file could cause Exim to calculate negative values
4415 for the mailbox size or file count. Odd effects could occur as a result.
4416 The maildirsize information is now recalculated if the size or filecount
4419 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4420 support for a long time. Removed HAVE_SYS_VFS_H.
4422 18. Updated exipick to current release from John Jetmore.
4424 19. Allow an empty sender to be matched against a lookup in an address list.
4425 Previously the only cases considered were a regular expression, or an
4428 20. Exim went into a mad DNS lookup loop when doing a callout where the
4429 host was specified on the transport, if the DNS lookup yielded more than
4432 21. The RFC2047 encoding function was originally intended for short strings
4433 such as real names; it was not keeping to the 75-character limit for
4434 encoded words that the RFC imposes. It now respects the limit, and
4435 generates multiple encoded words if necessary. To be on the safe side, I
4436 have increased the buffer size for the ${rfc2047: expansion operator from
4439 22. Failure to deliver a bounce message always caused it to be frozen, even if
4440 there was an errors_to setting on the router. The errors_to setting is now
4443 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4444 canonical form (fully expanded) before being placed in
4445 $sender_host_address.
4447 24. Updated eximstats to version 1.33
4449 25. Include certificate and key file names in error message when GnuTLS fails
4450 to set them up, because the GnuTLS error message doesn't include the name
4451 of the failing file when there is a problem reading it.
4453 26. Expand error message when OpenSSL has problems setting up cert/key files.
4456 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4457 (this can affect the format of dates).
4459 28. exim_tidydb, when checking for the continued existence of a message for
4460 which it has found a message-specific retry record, was not finding
4461 messages that were in split spool directories. Consequently, it was
4462 deleting retry records that should have stayed in existence.
4464 29. eximstats updated to version 1.35
4465 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4466 1.35 - bugfix such that pie charts by volume are generated correctly
4468 30. The SPA authentication driver was not abandoning authentication and moving
4469 on to the next authenticator when an expansion was forced to fail,
4470 contradicting the general specification for all authenticators. Instead it
4471 was generating a temporary error. It now behaves as specified.
4473 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4474 (the order specifies the preference for clients). The order is now AES256,
4475 AES128, 3DES, ARCFOUR128.
4477 31. Small patch to Sieve code - explicitly set From: when generating an
4480 32. Exim crashed if a remote delivery caused a very long error message to be
4481 recorded - for instance if somebody sent an entire SpamAssassin report back
4482 as a large number of 550 error lines. This bug was coincidentally fixed by
4483 increasing the size of one of Exim's internal buffers (big_buffer) that
4484 happened as part of the Exiscan merge. However, to be on the safe side, I
4485 have made the code more robust (and fixed the comments that describe what
4488 33. Some experimental protocols are using DNS PTR records for new purposes. The
4489 keys for these records are domain names, not reversed IP addresses. The
4490 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4491 leaves it alone. Component reversal etc. now happens only for IP addresses.
4494 34. The host_aton() function is supposed to be passed a string that is known
4495 to be a valid IP address. However, in the case of IPv6 addresses, it was
4496 not checking this. This is a hostage to fortune. Exim now panics and dies
4497 if the condition is not met. A case was found where this could be provoked
4498 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4499 components; fortuitously, this particular loophole had already been fixed
4500 by change 4.50/55 or 4.44/33 above.
4502 If there are any other similar loopholes, the new check in host_aton()
4503 itself should stop them being exploited. The report I received stated that
4504 data on the command line could provoke the exploit when Exim was running as
4505 exim, but did not say which command line option was involved. All I could
4506 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4507 running as the user.
4510 35. There was a buffer overflow vulnerability in the SPA authentication code
4511 (which came originally from the Samba project). I have added a test to the
4512 spa_base64_to_bits() function which I hope fixes it.
4515 36. The daemon start-up calls getloadavg() while still root for those OS that
4516 need the first call to be done as root, but it missed one case: when
4517 deliver_queue_load_max is set with deliver_drop_privilege. This is
4518 necessary for the benefit of the queue runner, because there is no re-exec
4519 when deliver_drop_privilege is set.
4521 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4522 was in use, was not putting the data itself into the right store pool;
4523 consequently, it could be overwritten for a subsequent message in the same
4524 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4527 38. Sometimes the final signoff response after QUIT could fail to get
4528 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4529 < 0 before doing a fflush(). This bug looks as though it goes back to the
4530 introduction of TLS in release 3.20, but "sometimes" must have been rare
4531 because the tests only now provoked it.
4537 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4538 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4539 Now it returns an integer. A number of calls were still expecting a T/F
4540 return. Fortuitously, in all cases, the tests worked in OK situations,
4541 which is the norm. However, things would have gone wrong on any write
4542 failures on the smtp file descriptor. This function is used when sending
4543 messages over SMTP and also when doing verify callouts.
4545 2. When Exim is called to do synchronous delivery of a locally submitted
4546 message (the -odf or -odi options), it no longer closes stderr before doing
4549 3. Implemented the mua_wrapper option.
4551 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4553 5. Implemented the functions header_remove(), header_testname(),
4554 header_add_at_position(), and receive_remove_recipient(), and exported them
4557 6. If an ACL "warn" statement specified the addition of headers, Exim already
4558 inserted X-ACL-Warn: at the start if there was no header name. However, it
4559 was not making this test for the second and subsequent header lines if
4560 there were newlines in the string. This meant that an invalid header could
4561 be inserted if Exim was badly configured.
4563 7. Allow an ACL "warn" statement to add header lines at the start or after all
4564 the Received: headers, as well as at the end.
4566 8. Added the rcpt_4xx retry error code.
4568 9. Added postmaster_mailfrom=xxx to callout verification option.
4570 10. Added mailfrom=xxxx to the callout verification option, for verify=
4573 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4574 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4576 12. Inserted some casts to stop certain compilers warning when using pointer
4577 differences as field lengths or precisions in printf-type calls (mostly
4578 affecting debugging statements).
4580 13. Added optional readline() support for -be (dynamically loaded).
4582 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4583 same clock tick as a message's arrival, so that its received time was the
4584 same as the "first fail" time on the retry record, and that message
4585 remained on the queue past the ultimate address timeout, every queue runner
4586 would try a delivery (because it was past the ultimate address timeout) but
4587 after another failure, the ultimate address timeout, which should have then
4588 bounced the address, did not kick in. This was a "< instead of <=" error;
4589 in most cases the first failure would have been in the next clock tick
4590 after the received time, and all would be well.
4592 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4593 being recognized when the domain list was tested by the match_domain
4594 condition in an expansion string.
4596 16. Added the ${str2b64: operator.
4598 17. Exim was always calling setrlimit() to set a large limit for the number of
4599 processes, without checking whether the existing limit was already
4600 adequate. (It did check for the limit on file descriptors.) Furthermore,
4601 errors from getrlimit() and setrlimit() were being ignored. Now they are
4602 logged to the main and panic logs, but Exim does carry on, to try to do its
4603 job under whatever limits there are.
4605 18. Imported PCRE 5.0.
4607 19. Trivial typo in log message " temporarily refused connection" (the leading
4610 20. If the log selector return_path_on_delivery was set and an address was
4611 redirected to /dev/null, the delivery process crashed because it assumed
4612 that a return path would always be set for a "successful" delivery. In this
4613 case, the whole delivery is bypassed as an optimization, and therefore no
4616 21. Internal re-arrangement: the function for sending a challenge and reading
4617 a response while authentication was assuming a zero-terminated challenge
4618 string. It's now changed to take a pointer and a length, to allow for
4619 binary data in such strings.
4621 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4623 23. Exim was not respecting finduser_retries when seeking the login of the
4624 uid under which it was called; it was always trying 10 times. (The default
4625 setting of finduser_retries is zero.) Also, it was sleeping after the final
4626 failure, which is pointless.
4628 24. Implemented tls_on_connect_ports.
4630 25. Implemented acl_smtp_predata.
4632 26. If the domain in control=submission is set empty, Exim assumes that the
4633 authenticated id is a complete email address when it generates From: or
4634 Sender: header lines.
4636 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4637 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4638 chown and chgrp in /bin and hostname in /usr/bin.
4640 28. Exim was keeping the "process log" file open after each use, just as it
4641 does for the main log. This opens the possibility of it remaining open for
4642 long periods when the USR1 signal hits a daemon. Occasional processlog
4643 errors were reported, that could have been caused by this. Anyway, it seems
4644 much more sensible not to leave this file open at all, so that is what now
4647 29. The long-running daemon process does not normally write to the log once it
4648 has entered its main loop, and it closes the log before doing so. This is
4649 so that log files can straightforwardly be renamed and moved. However,
4650 there are a couple of unusual error situations where the daemon does write
4651 log entries, and I had neglected to close the log afterwards.
4653 30. The text of an SMTP error response that was received during a remote
4654 delivery was being truncated at 512 bytes. This is too short for some of
4655 the long messages that one sometimes sees. I've increased the limit to
4658 31. It is now possible to make retry rules that apply only when a message has a
4659 specific sender, in particular, an empty sender.
4661 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4662 it possible to be selective about when SMTP synchronization is enforced.
4664 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4666 32. Implemented hosts_connection_nolog.
4668 33. Added an ACL for QUIT.
4670 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4673 35. Added mailbox_size and mailbox_filecount to appendfile.
4675 36. Added control = no_multiline_responses to ACLs.
4677 37. There was a bug in the logic of the code that waits for the clock to tick
4678 in the case where the clock went backwards by a substantial amount such
4679 that the microsecond fraction of "now" was more than the microsecond
4680 fraction of "then" (but the whole seconds number was less).
4682 38. Added support for the libradius Radius client library this is found on
4683 FreeBSD (previously only the radiusclient library was supported).
4689 1. When certain lookups returned multiple values in the form name=value, the
4690 quoting of the values was not always being done properly. Specifically:
4691 (a) If the value started with a double quote, but contained no whitespace,
4693 (b) If the value contained whitespace other than a space character (i.e.
4694 tabs or newlines or carriage returns) it was not quoted.
4695 This fix has been applied to the mysql and pgsql lookups by writing a
4696 separate quoting function and calling it from the lookup code. The fix
4697 should probably also be applied to nisplus, ibase and oracle lookups, but
4698 since I cannot test any of those, I have not disturbed their existing code.
4700 2. A hit in the callout cache for a specific address caused a log line with no
4701 reason for rejecting RCPT. Now it says "Previous (cached) callout
4702 verification failure".
4704 3. There was an off-by-one bug in the queryprogram router. An over-long
4705 return line was truncated at 256 instead of 255 characters, thereby
4706 overflowing its buffer with the terminating zero. As well as fixing this, I
4707 have increased the buffer size to 1024 (and made a note to document this).
4709 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4710 when Exim is waiting for an SMTP response from a remote server, Exim
4711 restarts its select() call on the socket, thereby resetting its timeout.
4712 This is not a problem when such interrupts are rare. Somebody set up a cron
4713 job to run exiwhat every 2 minutes, which is less than the normal select()
4714 timeout (5 or 10 minutes). This meant that the select() timeout never
4715 kicked in because it was always reset. I have fixed this by comparing the
4716 time when an interrupt arrives with the time at the start of the first call
4717 to select(). If more time than the timeout has elapsed, the interrupt is
4718 treated as a timeout.
4720 5. Some internal re-factoring in preparation for the addition of Sieve
4721 extensions (by MH). In particular, the "personal" test is moved to a
4722 separate function, and given an option for scanning Cc: and Bcc: (which is
4723 not set for Exim filters).
4725 6. When Exim created an email address using the login of the caller as the
4726 local part (e.g. when creating a From: or Sender: header line), it was not
4727 quoting the local part when it contained special characters such as @.
4729 7. Installed new OpenBSD configuration files.
4731 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4732 try to make them clearer.
4734 9. Callout options, other than the timeout value, were being ignored when
4735 verifying sender addresses in header lines. For example, when using
4737 verify = header_sender/callout=no_cache
4739 the cache was (incorrectly) being used.
4741 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4742 only the Cygwin environment.
4744 11. When return_path_on_delivery was set as a log selector, if different remote
4745 addresses in the same message used different return paths and parallel
4746 remote delivery occurred, the wrong values would sometimes be logged.
4747 (Whenever a remote delivery process finished, the return path value from
4748 the most recently started remote delivery process was logged.)
4750 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4751 lines when AUTH and/or TLS are in use. This is the "received protocol"
4752 field. Exim used to use "asmtp" for authenticated SMTP, without any
4753 indication (in the protocol name) for TLS use. Now it follows the RFC and
4754 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4755 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4756 names appear in log lines as well as in Received: header lines.
4758 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4759 extensions, and comparison tests, and to fix some bugs.
4761 14. Changes to the "personal" filter test:
4763 (1) The test was buggy in that it was just doing the equivalent of
4764 "contains" tests on header lines. For example, if a user's address was
4765 anne@some.where, the "personal" test would incorrectly be true for
4767 To: susanne@some.where
4769 This test is now done by extracting each address from the header in turn,
4770 and checking the entire address. Other tests that are part of "personal"
4771 are now done using regular expressions (for example, to check local parts
4772 of addresses in From: header lines).
4774 (2) The list of non-personal local parts in From: addresses has been
4775 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4776 taken from the Sieve specification recommendations.
4778 (3) If the message contains any header line starting with "List-" it is
4779 treated as non-personal.
4781 (4) The test for "circular" in the Subject: header line has been removed
4782 because it now seems ill-conceived.
4784 15. Minor typos in src/EDITME comments corrected.
4786 16. Installed latest exipick from John Jetmore.
4788 17. If headers_add on a router specified a text string that was too long for
4789 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4790 of string_sprintf() is now avoided.
4792 18. $message_body_size was not set (it was always zero) when running the DATA
4793 ACL and the local_scan() function.
4795 19. For the "mail" command in an Exim filter, no default was being set for
4796 the once_repeat time, causing a random time value to be used if "once" was
4797 specified. (If the value happened to be <= 0, no repeat happened.) The
4798 default is now 0s, meaning "never repeat". The "vacation" command was OK
4799 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4800 (I found it when inspecting the code).
4802 20. There is now an overall timeout for performing a callout verification. It
4803 defaults to 4 times the callout timeout, which applies to individual SMTP
4804 commands during the callout. The overall timeout applies when there is more
4805 than one host that can be tried. The timeout is checked before trying the
4806 next host. This prevents very long delays if there are a large number of
4807 hosts and all are timing out (e.g. when the network connections are timing
4808 out). The value of the overall timeout can be changed by specifying an
4809 additional sub-option for "callout", called "maxwait". For example:
4811 verify = sender/callout=5s,maxwait=20s
4813 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4814 to the end before writing, but this should make it even safer).
4816 22. Exim was forgetting that it had advertised PIPELINING for the second and
4817 subsequent messages on an SMTP connection. It was also not resetting its
4818 memory on STARTTLS and an internal HELO.
4820 23. When Exim logs an SMTP synchronization error within a session, it now
4821 records whether PIPELINING has been advertised or not.
4823 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4824 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4825 rather than long int.
4827 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4829 26. Added the never_mail option to autoreply.
4835 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4836 crash if the getsockname() call failed; this can happen if a connection is
4837 closed very soon after it is established. The problem was simply in the
4838 order in which certain operations were done, causing Exim to try to write
4839 to the SMTP stream before it had set up the file descriptor. The bug has
4840 been fixed by making things happen in the correct order.
4846 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4847 before the connection was closed, thus losing the rejection response.
4849 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4850 some early Solaris releases, but causes trouble in current releases where
4851 socklen_t is defined.
4853 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4856 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4859 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4860 that releasing the top of it at the end releases what was used for sub-
4861 expansions (unless the block got too big). However, discard this block if
4862 the first thing is a variable or header, so that we can use its block when
4863 it is dynamic (useful for very large $message_headers, for example).
4865 6. Lookups now cache *every* query, not just the most recent. A new, separate
4866 store pool is used for this. It can be recovered when all lookup caches are
4867 flushed. Lookups now release memory at the end of their result strings.
4868 This has involved some general refactoring of the lookup sources.
4870 7. Some code has been added to the store_xxx() functions to reduce the amount
4871 of flapping under certain conditions.
4873 8. log_incoming_interface used to affect only the <= reception log lines. Now
4874 it causes the local interface and port to be added to several more SMTP log
4875 lines, for example "SMTP connection from", and rejection lines.
4877 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4879 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4881 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4882 could be overwritten at the end of the current message (or the start of a
4883 new message if it was set in a HELO ACL). The value is now preserved for
4884 the duration of the SMTP connection.
4886 12. If a transport had a headers_rewrite setting, and a matching header line
4887 contained an unqualified address, that address was qualified, even if it
4888 did not match any rewriting rules. The underlying bug was that the values
4889 of the flags that permit the existence of unqualified sender and recipient
4890 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4891 non-local messages, and by -bnq for local messages) were not being
4892 preserved with the message after it was received.
4894 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4895 "next input=" as part of the text comprising the host identity instead of
4896 the correct text. The code was using the same buffer for two different
4897 strings. However, depending on which order the printing function evaluated
4898 its arguments, the bug did not always show up. Under Linux, for example, my
4899 test suite worked just fine.
4901 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4902 doesn't work with some older versions of Perl. It has been changed to "my",
4903 which in any case is probably the better facility to use.
4905 15. A really picky compiler found some instances of statements for creating
4906 error messages that either had too many or two few arguments for the format
4909 16. The size of the buffer for calls to the DNS resolver has been increased
4910 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4911 for addresses that have a lot of PTR records. This alleviates a problem; it
4912 does not fully solve it.
4914 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4915 buffer now truncates the list and logs the incident, which is the same
4916 action as happens when Exim is looking up a host name and its aliases.
4917 Previously in this situation something unpredictable would happen;
4918 sometimes it was "internal error: store_reset failed".
4920 18. If a server dropped the connection unexpectedly when an Exim client was
4921 using GnuTLS and trying to read a response, the client delivery process
4922 crashed while trying to generate an error log message.
4924 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4925 string, for example:
4927 warn message = H1: something\nH2: something
4929 the text was added as a single header line from Exim's point of view
4930 though it ended up OK in the delivered message. However, searching for the
4931 second and subsequent header lines using $h_h2: did not work. This has been
4932 fixed. Similarly, if a system filter added multiple headers in this way,
4933 the routers could not see them.
4935 20. Expanded the error message when iplsearch is called with an invalid key to
4936 suggest using net-iplsearch in a host list.
4938 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4939 ACLs are no longer actually imposed (and a message to that effect is
4942 22. If a "gecos" field in a passwd entry contained escaped characters, in
4943 particular, if it contained a \" sequence, Exim got it wrong when building
4944 a From: or a Sender: header from that name. A second bug also caused
4945 incorrect handling when an unquoted " was present following a character
4946 that needed quoting.
4948 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4949 was not being matched caselessly.
4951 24. Arranged for all hyphens in the exim.8 source to be escaped with
4954 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4955 properly. Recipient callout cache records were still being keyed to include
4956 the sender, even when use_sender was set false. This led to far more
4957 callouts that were necessary. The sender is no longer included in the key
4958 when use_sender is false.
4960 26. Added "control = submission" modifier to ACLs.
4962 27. Added the ${base62d: operator to decode base 62 numbers.
4964 28. dnsdb lookups can now access SRV records.
4966 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4967 the configuration file.
4969 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4970 (-v) mode. This makes the output for a verbose queue run more intelligible.
4972 31. Added a use_postmaster feature to recipient callouts.
4974 32. Added the $body_zerocount variable, containing the number of binary zero
4975 bytes in the message body.
4977 33. The time of last modification of the "new" subdirectory is now used as the
4978 "mailbox time last read" when there is a quota error for a maildir
4981 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4983 35. Added +ignore_unknown as a special item in host lists.
4985 36. Code for decoding IPv6 addresses in host lists is now included, even if
4986 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4987 address was recognized as an IP address, but was then not correctly decoded
4988 into binary, causing unexpected and incorrect effects when compared with
4995 1. Very minor rewording of debugging text in manualroute to say "list of
4996 hosts" instead of "hostlist".
4998 2. If verify=header_syntax was set, and a header line with an unqualified
4999 address (no domain) and a large number of spaces between the end of the
5000 name and the colon was received, the reception process suffered a buffer
5001 overflow, and (when I tested it) crashed. This was caused by some obsolete
5002 code that should have been removed. The fix is to remove it!
5004 3. When running in the test harness, delay a bit after writing a bounce
5005 message to get a bit more predictability in the log output.
5007 4. Added a call to search_tidyup() just before forking a reception process. In
5008 theory, someone could use a lookup in the expansion of smtp_accept_max_
5009 per_host which, without the tidyup, could leave open a database connection.
5011 5. Added the variables $recipient_data and $sender_data which get set from a
5012 lookup success in an ACL "recipients" or "senders" condition, or a router
5013 "senders" option, similar to $domain_data and $local_part_data.
5015 6. Moved the writing of debug_print from before to after the "senders" test
5018 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5019 problems for message scanning, either using a data ACL, or using
5020 local_scan() because the Received: header was not generated till after they
5021 were called (in order to set the time as the time of reception completion).
5022 I have revised the way this works. The header is now generated after the
5023 body is received, but before the ACL or local_scan() are called. After they
5024 are run, the timestamp in the header is updated.
5030 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5031 before starting a queue runner without re-exec. This happened only when
5032 deliver_drop_privilege was set or when the Exim user was set to root. The
5033 effect of the bug was that timeouts during subsequent deliveries caused
5034 crashes instead of being properly handled. The handler is now left at its
5035 default (and expected) setting.
5037 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5038 message, again when deliver_drop_privilege is set or Exim is run as root.
5039 The bug described in (1) was not present in this case, but the tidying up
5040 of the other signals was missing. I have made the two cases consistent.
5042 3. The ignore_target_hosts setting on a manualroute router was being ignored
5043 for hosts that were looked up using the /MX notation.
5045 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5048 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5049 operated on the sender address. After changing the $sender_address to <>
5050 for the sender address verify, Exim was re-instated it as the original
5051 (before rewriting) address, but remembering that it had rewritten it, so it
5052 wasn't rewriting it again. This bug also had the effect of breaking the
5053 sender address verification caching when the sender address was rewritten.
5055 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5056 This has been changed so that if the ip literal address matches
5057 ignore_target_hosts, the router declines.
5059 7. Added expansion conditions match_domain, match_address, and match_local_
5060 part (NOT match_host).
5062 8. The placeholder for the Received: header didn't have a length field set.
5064 9. Added code to Exim itself and to exim_lock to test for a specific race
5065 condition that could lead to file corruption when using MBX delivery. The
5066 issue is with the lockfile that is created in /tmp. If this file is removed
5067 after a process has opened it but before that process has acquired a lock,
5068 there is the potential for a second process to recreate the file and also
5069 acquire a lock. This could lead to two Exim processes writing to the file
5070 at the same time. The added code performs the same test as UW imapd; it
5071 checks after acquiring the lock that its file descriptor still refers to
5072 the same named file.
5074 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5075 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5078 11. Added the smtp_active_hostname option. If used, this will typically be made
5079 to depend on the incoming interface address. Because $interface_address is
5080 not set up until the daemon has forked a reception process, error responses
5081 that can happen earlier (such as "too many connections") no longer contain
5084 12. If an expansion in a condition on a "warn" statement fails because a lookup
5085 defers, the "warn" statement is abandoned, and the next ACL statement is
5086 processed. Previously this caused the whole ACL to be aborted.
5088 13. Added the iplsearch lookup type.
5090 14. Added ident_timeout as a log selector.
5092 15. Added tls_certificate_verified as a log selector.
5094 16. Added a global option tls_require_ciphers (compare the smtp transport
5095 option of the same name). This controls incoming TLS connections.
5097 17. I finally figured out how to make tls_require_ciphers do a similar thing
5098 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5099 before starting the TLS session.
5101 18. Tabs are now shown as \t in -bP output.
5103 19. If the log selector return_path_on_delivery was set, Exim crashed when
5104 bouncing a message because it had too many Received: header lines.
5106 20. If two routers both had headers_remove settings, and the first one included
5107 a superfluous trailing colon, the final name in the first list and the
5108 first name in the second list were incorrectly joined into one item (with a
5109 colon in the middle).
5115 1. Added -C and -D options to the exinext utility, mainly to make it easier
5116 to include in the automated testing, but these could be helpful when
5117 multiple configurations are in use.
5119 2. The exinext utility was not formatting the output nicely when there was
5120 an alternate port involved in the retry record key, nor when there was a
5121 message id as well (for retries that were specific to a specific message
5122 and a specific host). It was also confused by IPv6 addresses, because of
5123 the additional colons they contain. I have fixed the IPv4 problem, and
5124 patched it up to do a reasonable job for IPv6.
5126 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5127 delivery, the log line now contains "pipelined" if PIPELINING was used.
5129 4. An SMTP transport process used to panic and die if the bind() call to set
5130 an explicit outgoing interface failed. This has been changed; it is now
5131 treated in the same way as a connect() failure.
5133 5. A reference to $sender_host_name in the part of a conditional expansion
5134 that was being skipped was still causing a DNS lookup. This no longer
5137 6. The def: expansion condition was not recognizing references to header lines
5138 that used bh_ and bheader_.
5140 7. Added the _cache feature to named lists.
5142 8. The code for checking quota_filecount in the appendfile transport was
5143 allowing one more file than it should have been.
5145 9. For compatibility with Sendmail, the command line option
5153 and sets the incoming protocol and host name (for trusted callers). The
5154 host name and its colon can be omitted when only the protocol is to be set.
5155 Note the Exim already has two private options, -pd and -ps, that refer to
5156 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5157 "s", but I don't think that's a major issue.
5159 10. A number of refactoring changes to the code, none of which should affect
5162 (a) The number of logging options was getting close to filling up the
5163 32-bit word that was used as a bit map. I have split them into two classes:
5164 those that are passed in the argument to log_write(), and those that are
5165 only ever tested independently outside of that function. These are now in
5166 separate 32-bit words, so there is plenty of room for expansion again.
5167 There is no change in the user interface or the logging behaviour.
5169 (b) When building, for example, log lines, the code previously used a
5170 macro that called string_cat() twice, in order to add two strings. This is
5171 not really sufficiently general. Furthermore, there was one instance where
5172 it was actually wrong because one of the argument was used twice, and in
5173 one call a function was used. (As it happened, calling the function twice
5174 did not affect the overall behaviour.) The macro has been replaced by a
5175 function that can join an arbitrary number of extra strings onto a growing
5178 (c) The code for expansion conditions now uses a table and a binary chop
5179 instead of a serial search (which was left over from when there were very
5180 few conditions). Also, it now recognizes conditions like "pam" even when
5181 the relevant support is not compiled in: a suitably worded error message is
5182 given if an attempt is made to use such a condition.
5184 11. Added ${time_interval:xxxxx}.
5186 12. A bug was causing one of the ddress fields not to be passed back correctly
5187 from remote delivery subprocesses. The field in question was not being
5188 subsequently used, so this caused to problems in practice.
5190 13. Added new log selectors queue_time and deliver_time.
5192 14. Might have fixed a bug in maildirsizefile handling that threw up
5193 "unexpected character" debug warnings, and recalculated the data
5194 unnecessarily. In any case, I expanded the warning message to give more
5197 15. Added the message "Restricted characters in address" to the statements in
5198 the default ACL that block characters like @ and % in local parts.
5200 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5201 Three changes have been made:
5203 (a) There was a serious bug; a negative response to MAIL caused the whole
5204 recipient domain to be cached as invalid, thereby blocking all messages
5205 to all local parts at the same domain, from all senders. This bug has
5206 been fixed. The domain is no longer cached after a negative response to
5207 MAIL if the sender used is not empty.
5209 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5212 (c) A new callout option, "use_sender" has been added for people who want
5213 the modified behaviour.
5219 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5222 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5223 indeed breaks things for older releases.
5225 3. Added additional logging to the case where there is a problem reading data
5226 from a filter that is running in a subprocess using a pipe, in order to
5227 try to track down a specific problem.
5229 4. Testing facility fudge: when running in the test harness and attempting
5230 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5231 getting "No route to host". Convert this to a timeout.
5233 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5236 6. Some OS don't have socklen_t but use size_t instead. This affects the
5237 fifth argument of getsockopt() amongst other things. This is now
5238 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5239 can be set for individual OS. I have set it for SunOS5, OSF1, and
5240 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5241 some earlier ones do not.
5243 7. Change 4.30/15 was not doing the test caselessly.
5245 8. The standard form for an IPv6 address literal was being rejected by address
5246 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5247 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5248 this, as well as the form without the "IPv6" on the front (but only when
5249 address literals are enabled, of course).
5251 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5253 10. Exim crashed if a message with an empty sender address specified by -f
5254 encountered a router with an errors_to setting. This could be provoked only
5255 by a command such as
5259 where an empty string was supplied; "<>" did not hit this bug.
5261 11. Installed PCRE release 4.5.
5263 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5264 remained set. It is now erased.
5266 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5267 times from message ids (which are base 36 rather than the normal 62).
5269 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5270 were being counted as actual protocol errors, and logged if the log
5271 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5272 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5273 and DATA following a set of rejected RCPTs do not count as protocol errors.
5274 In other words, Exim assumes they were pipelined, though this may not
5275 actually be the case. Of course, in all cases the client gets an
5276 appropriate error code.
5278 15. If a lookup fails in an ACL condition, a message about the failure may
5279 be available; it is used if testing the ACL cannot continue, because most
5280 such messages specify what the cause of the deferral is. However, some
5281 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5282 that caused an old message to be retained and used if a later statement
5283 caused a defer, replacing the real cause of the deferral.
5285 16. If an IP address had so many PTR records that the DNS lookup buffer
5286 was not large enough to hold them, Exim could crash while trying to process
5287 the truncated data. It now detects and logs this case.
5289 17. Further to 4.21/58, another change has been made: if (and only if) the
5290 first line of a message (the first header line) ends with CRLF, a bare LF
5291 in a subsequent header line has a space inserted after it, so as not to
5292 terminate the header.
5294 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5295 unnecessarily, used atoi() instead of strtol(), and didn't check the
5296 termination when getting file sizes from file names by regex.
5298 19. Completely re-implemented the support for maildirsize files, in the light
5299 of a number of problems with the previous contributed implementation
5300 (4.30/29). In particular:
5302 . If the quota is zero, the maildirsize file is maintained, but no quota is
5305 . If the maildir directory does not exist, it is created before any attempt
5306 to write a maildirsize file.
5308 . The quota value in the file is just a cache; if the quota is changed in
5309 the transport, the new value overrides.
5311 . A regular expression is available for excluding directories from the
5314 20. The autoreply transport checks the characters in options that define the
5315 message's headers; it allows continued headers, but it was checking with
5316 isspace() after an embedded newline instead of explicitly looking for a
5319 21. If all the "regular" hosts to which an address was routed had passed their
5320 expiry times, and had not reached their retry times, the address was
5321 bounced, even if fallback hosts were defined. Now Exim should go on to try
5324 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5325 equivalent code in the SMTP transport. Some hosts send humungous responses
5326 to HELO/EHLO, more than 1024 it seems.
5328 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5329 gives compiler warnings in some environments. I've now done it "properly",
5332 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5333 (because of problems with the built-in one) was declared to return uschar *
5334 instead of char *, causing compiler failure.
5336 25. Fixed a file descriptor leak when processing alias/forward files.
5338 26. Fixed a minor format string issue in dbfn.c.
5340 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5342 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5343 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5346 29. When a sender address is verified, it is cached, to save repeating the test
5347 when there is more than one recipient in a message. However, when the
5348 verification involves a callout, it is possible for different callout
5349 options to be set for different recipients. It is too complicated to keep
5350 track of this in the cache, so now Exim always runs a verification when a
5351 callout is required, relying on the callout cache for the optimization.
5352 The overhead is duplication of the address routing, but this should not be
5355 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5356 to be verified with callout=postmaster, and the main callout worked but the
5357 postmaster check failed, the verification correctly failed. However, if a
5358 subsequent RCPT command asked for sender verification *without* the
5359 postmaster check, incorrect caching caused this verification also to fail,
5362 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5363 it was not caching the DNS options (qualify_single, search_parents) that
5364 were used when the lookup failed. A subsequent lookup with different
5365 options therefore always gave the same answer, though there were cases
5366 where it should not have. (Example: a "domains = !$mx_any" option on a
5367 dnslookup router: the "domains" option is always processed without any
5368 widening, but the router might have qualify_single set.) Now Exim uses the
5369 cached value only when the same options are set.
5371 32. Added John Jetmore's "exipick" utility to the distribution.
5373 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5374 than a timeout (e.g. a certificate is required, and is not provided), an
5375 Exim server now closes the connection immediately. Previously it waited for
5376 the client to close - but if the client is SSL, it seems that they each
5377 wait for each other, leading to a delay before one of them times out.
5379 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5380 maintained 0.8.x compatibility because I don't think many are using it, and
5381 it is clearly obsolete.
5383 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5386 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5387 client certificate was expired. A simple patch fixes this, though I don't
5388 understand the full logic of why the verify callback is called multiple
5391 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5392 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5393 which causes problems with some clients (such as the Certicom SSL Plus
5394 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5395 disables the coutermeasure allowing Eudora to connect."
5397 38. Exim was not checking that a write() to a log file succeeded. This could
5398 lead to Bad Things if a log got too big, in particular if it hit a file
5399 size limit. Exim now panics and dies if it cannot write to a log file, just
5400 as it does if it cannot open a log file.
5402 39. Modified OS/Makefile-Linux so that it now contains
5404 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5406 The two -D definitions ensure that Exim is compiled with large file
5407 support, which makes it possible to handle log files that are bigger than
5410 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5411 instance) a domain was checked against a named list that involved a lookup,
5412 causing $domain_data to be set, then another domain was checked against the
5413 same list, then the first domain was re-checked, the value of $domain_data
5414 after the final check could be wrong. In particular, if the second check
5415 failed, it could be set empty. This bug probably also applied to
5418 41. The strip_trailing_dot option was not being applied to the address given
5419 with the -f command-line option.
5421 42. The code for reading a message's header from the spool was incrementing
5422 $received_count, but never initializing it. This meant that the value was
5423 incorrect (doubled) while delivering a message in the same process in which
5424 it was received. In the most common configuration of Exim, this never
5425 happens - a fresh exec is done - but it can happen when
5426 deliver_drop_privilege is set.
5428 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5429 it now includes up to 150 characters of the unexpected data in the log
5432 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5433 and building data strings. The size of both of these buffers was 10 000
5434 bytes - far larger than anybody would *ever* want, thought I. Needless to
5435 say, somebody hit the limit. I have increased the maximum line length to
5436 20 000 and the maximum data length of concatenated lines to 100 000. I have
5437 also fixed two bugs, because there was no checking on these buffers. Tsk,
5438 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5439 buffer is too small.
5441 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5442 lsearch lookups. Now it does.
5444 46. When parsing a route_list item in a manualroute router, a fixed-length
5445 buffer was used for the list of hosts. I made this 1024 bytes long,
5446 thinking that nobody would ever have a list of hosts that long. Wrong.
5447 Somebody had a whole pile of complicated expansion conditions, and the
5448 string was silently truncated, leading to an expansion error. It turns out
5449 that it is easier to change to an unlimited length (owing to other changes
5450 that have happened since this code was originally written) than to build
5451 structure for giving a limitation error. The length of the item that
5452 expands into the list of hosts is now unlimited.
5454 47. The lsearch lookup could not handle data where the length of text line was
5455 more than 4095 characters. Such lines were truncated, leading to shortened
5456 data being returned. It should now handle lines of any length.
5458 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5459 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5462 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5463 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5464 no longer happens in comment lines. A list of replacements is now placed
5465 at the head of all of the source files, except those whose only change is
5466 to replace PERL_COMMAND in the very first #! line.
5468 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5469 messages on the queue, with a bottom-up merge sort, using code contributed
5470 by Michael Haardt. This should make operations like -bp somewhat faster on
5471 large queues. It won't affect queue runners, except when queue_run_in_order
5474 51. Installed eximstats 1.31 in the distribution.
5476 52. Added support for SRV lookups to the dnslookup router.
5478 53. If an ACL referred to $message_body or $message_body_end, the value was not
5479 reset for any messages that followed in the same SMTP session.
5481 54. The store-handling optimization for building very long strings was not
5482 differentiating between the different store pools. I don't think this
5483 actually made any difference in practice, but I've tidied it.
5485 55. While running the routers to verify a sender address, $sender_address
5486 was still set to the sender address. This is wrong, because when routing to
5487 send a bounce to the sender, it would be empty. Therefore, I have changed
5488 it so that, while verifying a sender address, $sender_address is set to <>.
5489 (There is no change to what happens when verifying a recipient address.)
5491 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5492 target A or AAAA records (if not already returned) without resetting the
5493 qualify_single or search_parents options of the DNS resolver. These are
5494 inappropriate in this case because the targets of MX and SRV records must
5495 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5496 target that, when qualified, matched something in the local domain. These
5497 two options are now turned off when doing these lookups.
5499 57. It seems that at least some releases of Reiserfs (which does not have the
5500 concept of a fixed number of inodes) returns zero and not -1 for the
5501 number of available inodes. This interacted badly with check_spool_inodes,
5502 which assumed that -1 was the "no such thing" setting. What I have done is
5503 to check that the total number of inodes is greater than zero before doing
5504 the test of how many are available.
5506 58. When a "warn" ACL statement has a log_message modifier, the message is
5507 remembered, and not repeated. This is to avoid a lot of repetition when a
5508 message has many recipients that cause the same warning to be written.
5509 Howewer, Exim was preserving the list of already written lines for an
5510 entire SMTP session, which doesn't seem right. The memory is now reset if a
5511 new message is started.
5513 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5514 debugging output unless log_rewrite was also set.
5516 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5517 of (int)(handle) when we know that handle contains (void *)(-1).
5519 61. The Exim daemon panic-logs an error return when it closes the incoming
5520 connection. However "connection reset by peer" seems to be common, and
5521 isn't really an error worthy of noting specially, so that particular error
5524 62. When Exim is trying to find all the local interfaces, it used to panic and
5525 die if the ioctl to get the interface flags failed. However, it seems that
5526 on at least one OS (Solaris 9) it is possible to have an interface that is
5527 included in the list of interfaces, but for which you get a failure error
5528 for this call. This happens when the interface is not "plumbed" into a
5529 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5530 failure of the "get flags" call assumes that the interface is down.
5532 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5533 makes life easier for people who are doing arithmetic on fields extracted
5534 from dates, where you often get leading zeros that should not be
5535 interpreted as octal.
5537 64. Added qualify_domain to the redirect router, to override the global
5540 65. If a pathologically long header line contained very many addresses (the
5541 report of this problem mentioned 10 000) and each of them was rewritten,
5542 Exim could use up a very large amount of memory. (It kept on making new
5543 copies of the header line as it rewrote, and never released the old ones.)
5544 At the expense of a bit more processing, the header rewriting function has
5545 been changed so that it no longer eats memory in this way.
5547 66. The generation of the Received: header has been moved from the time that a
5548 message starts to be received, to the time that it finishes. The timestamp
5549 in the Received: header should now be very close to that of the <= log
5550 line. There are two side-effects of this change:
5552 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5553 logged header lines no longer include the local Received: line, because
5554 it has not yet been created. The same applies to a copy of the message
5555 that is returned to a non-SMTP sender when a message is rejected.
5557 (b) When a filter file is tested using -bf, no additional Received: header
5558 is added to the test message. After some thought, I decided that this
5561 This change does not affect the value of $received_for. It is still set
5562 after address rewriting, but before local_scan() is called.
5564 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5566 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5567 gave an unhelpful panic error message, and a defer error. I have managed to
5568 change this behaviour so that it now rejects any supplied certificate,
5569 which seems right, as the list of acceptable certificates is empty.
5571 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5572 gave an unhelpful defer error. I have not managed to make this reject any
5573 supplied certificates, but the error message it gives is "no certificate
5574 supplied", which is not helpful.
5576 70. exigrep's output now also includes lines that are not associated with any
5577 message, but which match the given pattern. Implemented by a patch from
5578 Martin Sluka, which also tidied up the Perl a bit.
5580 71. Recipient callout verification, like sender verification, was using <> in
5581 the MAIL FROM command. This isn't really the right thing, since the actual
5582 sender may affect whether the remote host accepts the recipient or not. I
5583 have changed it to use the actual sender in the callout; this means that
5584 the cache record is now keyed on a recipient/sender pair, not just the
5585 recipient address. There doesn't seem to be a real danger of callout loops,
5586 since a callout by the remote host to check the sender would use <>.
5587 [SEE ABOVE: changed after hitting problems.]
5589 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5590 temporary errors. However, in the case of such a code being given after
5591 the end of a data transmission (i.e. after ".") Exim was failing to write
5592 a retry record for the message. (Yes, there was some broken host that was
5593 actually sending 8xx at this point.)
5595 73. An unknown lookup type in a host list could cause Exim to panic-die when
5596 the list was checked. (An example that provoked this was putting <; in the
5597 middle of a list instead of at the start.) If this happened during a DATA
5598 ACL check, a -D file could be left lying around. This kind of configuration
5599 error no longer causes Exim to die; instead it causes a defer errror. The
5600 incident is still logged to the main and panic logs.
5602 74. Buglet left over from Exim 3 conversion. The message "too many messages
5603 in one connection" was written to the rejectlog but not the mainlog, except
5604 when address rewriting (yes!) was being logged.
5606 75. Added write_rejectlog option.
5608 76. When a system filter was run not as root (that is, when system_filter_user
5609 was set), the values of the $n variables were not being returned to the
5610 main process; thus, they were not subsequently available in the $sn
5613 77. Added +return_path_on_delivery log selector.
5615 78. A connection timeout was being treated differently from recipients deferred
5616 when testing hosts_max_try with a message that was older than the host's
5617 retry timeout. (The host should not be counted, thus allowing all hosts to
5618 be tried at least once before bouncing.) This may have been the cause of an
5619 occasionally reported bug whereby a message would remain on the queue
5620 longer than the retry timeout, but would be bounced if a delivery was
5621 forced. I say "may" because I never totally pinned down the problem;
5622 setting up timeout/retry tests is difficult. See also the next item.
5624 79. The ultimate address timeout was not being applied to errors that involved
5625 a combination of host plus message (for example, a timeout on a MAIL
5626 command). When an address resolved to a number of possible hosts, and they
5627 were not all tried for each delivery (e.g. because of hosts_max_try), a
5628 message could remain on the queue longer than the retry timeout.
5630 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5633 81. Fixed an obscure SMTP outgoing bug which required at least the following
5634 conditions: (a) there was another message waiting for the same server;
5635 (b) the server returned 5xx to all RCPT commands in the first message so
5636 that the message was not completed; (c) the server dropped the connection
5637 or gave a negative response to the RSET that Exim sends to abort the
5638 transaction. The observed case was a dropped connection after DATA that had
5639 been sent in pipelining mode. That is, the server had advertised PIPELINING
5640 but was not implementing it correctly. The effect of the bug was incorrect
5641 behaviour, such as trying another host, and this could lead to a crash.
5647 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5648 and daemon.c were passed as pointers to ints; they should have been
5649 pointers to socklen_t variables (which are typically unsigned ints).
5651 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5654 3. Fixed a really odd bug that affected only the testing scheme; patching a
5655 certain fixed string in the binary changed the value of another string that
5656 happened to be identical to the end of the original first string.
5658 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5659 name", it returns that address as the IP address. On some operating
5660 systems (e.g. Solaris), it also passes back the IP address string as the
5661 "host name". However, on others (e.g. Linux), it passes back an empty
5662 string. Exim wasn't checking for this, and was changing the host name to an
5663 empty string, assuming it had been canonicized.
5665 5. Although rare, it is permitted to have more than one PTR record for a given
5666 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5667 all the names associated with an address, because they do in Solaris.
5668 However, it seems that they do not in Linux for data that comes from the
5669 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5670 I found this out when I moved to a new Linux workstation and tried to run
5671 the Exim test suite.
5673 To get round this problem I have changed the code so that it now does its
5674 own call to the DNS to look up PTR records when searching for a host name.
5675 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5676 addresses that are only in /etc/hosts are still found.
5678 This behaviour is, however, controlled by an option called host_lookup_
5679 order, which defaults to "bydns:byaddr". If people want to use the other
5680 order, or indeed, just use one or the other means of lookup, they can
5681 specify it in this variable.
5683 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5684 some operating systems, this comes back from gethostbyaddr() as an empty
5685 string, and this is what Exim used to test for. However, it seems that in
5686 other systems, "." is yielded. Exim now tests for this case too.
5688 7. The values of check_spool_space and check_log_space are now held internally
5689 as a number of kilobytes instead of an absolute number of bytes. If a
5690 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5691 kilobyte. This means that much larger values can be stored.
5693 8. Exim monitor: an attempt to get the action menu when not actually pointing
5694 at a message produces an empty menu entitled "No message selected". This
5695 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5696 no entries in it ("Shell widget menu has zero width and/or height"). So I
5697 have added a single, blank menu entry in this case.
5699 9. Added ${quote_local_part.
5701 10. MIME decoding is now applied to the contents of Subject: header lines when
5704 11. Now that a reference to $sender_host_address automatically causes a reverse
5705 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5706 host lookup before query-style lookups in lists that might use this
5707 variable. This has therefore been abolished, and the "net-" prefix is no
5708 longer necessary for query-style lookups.
5710 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5711 have been a typo for LFLAGS, so it has been changed.
5713 13. The install script calls Exim with "-C /dev/null" in order to find the
5714 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5715 to be output. Howeve, since Exim outputs its version number before the
5716 error, it didn't break the script. It just looked ugly. I fixed this by
5717 always allowing "-C /dev/null" if the caller is root.
5719 14. Ignore overlarge ACL variable number when reading spool file - insurance
5720 against a later release with more variables having written the file.
5722 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5723 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5724 this, as well as the form without the "IPv6" on the front.
5726 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5727 OS/Makefile-Darwin file.
5729 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5730 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5732 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5733 variable. However, some OS use ints for inodes. I've added cast to long int
5734 to get rid of the compiler warning.
5736 19. I had forgotten to lock out "/../" in configuration file names when
5737 ALT_CONFIG_PREFIX was set.
5739 20. Routers used for verification do not need to specify transports. However,
5740 if such a router generated a host list, and callout was configured, Exim
5741 crashed, because it could not find a port number from the (non-existent)
5742 transport. It now assumes port 25 in this circumstance.
5744 21. Added the -t option to exigrep.
5746 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5747 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5748 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5749 them is set, LOOKUP_LSEARCH is forced.
5751 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5753 24. Added sender and host information to the "rejected by local_scan()" log
5754 line; previously there was no indication of these.
5756 25. Added .include_if_exists.
5758 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5759 Linux. It turns out that not all file systems support this. Apparently some
5760 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5761 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5762 not supported on the file descriptor, is now ignored when Exim is trying to
5763 sync a directory. This applies only to Linux.
5765 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5767 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5768 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5771 29. Added support for maildirsize files from supplied patch (modified a bit).
5773 30. The use of :fail: followed by an empty string could lead Exim to respond to
5774 sender verification failures with (e.g.):
5776 550 Verification failed for <xxx>
5777 550 Sender verify failed
5779 where the first response line was missing the '-' that indicates it is not
5780 the final line of the response.
5782 31. The loop for finding the name of the user that called Exim had a hardwired
5783 limit of 10; it now uses the value of finduser_retries, which is used for
5784 all other user lookups.
5786 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5789 33. Exim was neglecting to zero errno before one call of strtol() when
5790 expanding a string and expecting an integer value. On some systems this
5791 resulted in spurious "integer overflow" errors. Also, it was casting the
5792 result into an int without checking.
5794 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5795 did not work. The code looks as if it has *never* worked, though it appears
5796 to have been documented since at least releast 1.62. I have made it work.
5798 35. The "timeout_DNS" error in retry rules, also documented since at least
5799 1.62, also never worked. As it isn't clear exactly what this means, and
5800 clearly it isn't a major issue, I have abolished the feature by treating it
5801 as "timeout", and writing a warning to the main and panic logs.
5803 36. The display of retry rules for -brt wasn't always showing the error code
5806 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5807 timeout_connect_A, timeout_connect_MX.
5809 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5810 to the empty sender.
5812 39. The daemon was not analysing the content of -oX till after it had closed
5813 stderr and disconnected from the controlling terminal. This meant that any
5814 syntax errors were only noted on the panic log, and the return code from
5815 the command was 0. By re-arranging the code a little, I've made the
5816 decoding happen first, so such errors now appear on stderr, and the return
5817 code is 1. However, the actual setting up of the sockets still happens in
5818 the disconnected process, so errors there are still only recorded on the
5821 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5822 connections (as happens on some IP stacks) was logged at start up time as
5823 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5824 it from "IPv6 and IPv4", which means that two separate sockets are being
5827 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5828 says whether AF_INET or AF_INET6 was passed as an argument.
5830 42. Exiwhat output was messed up when time zones were included in log
5833 43. Exiwhat now gives more information about the daemon's listening ports,
5834 and whether -tls-on-connect was used.
5836 44. The "port" option of the smtp transport is now expanded.
5838 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5839 silently ignored. Now an error message is written to the main and panic
5842 46. There's a new ACL modifier called "logwrite" which writes to a log file
5843 as soon as it is encountered.
5845 47. Added $local_user_uid and $local_user_gid at routing time.
5847 48. Exim crashed when trying to verify a sender address that was being
5850 49. Exim was recognizing only a space character after ".include". It now also
5851 recognizes a tab character.
5853 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5854 extracting the relevant information from the specification. The man page no
5855 longer contains scrambled data for the -d option, and I've added a section
5856 at the front about calling Exim under different names.
5858 51. Added "extra_headers" argument to the "mail" command in filter files.
5860 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5863 53. Installed eximstats 1.29.
5865 54. Added transport_filter_timeout as a generic transport option.
5867 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5868 Cc: header lines. This was required by RFC 822, but it not required by RFC
5871 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5872 incoming messages that did not have them. Now it does so only if the
5873 message originates locally, that is, if there is no associated remote host
5874 address. When Resent- header lines are present, this applies to the Resent-
5875 lines rather than the non-Resent- lines.
5877 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5878 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5880 58. Messages for configuration errors now include the name of the main
5881 configuration file - useful now that there may be more than one file in a
5882 list (.included file names were always shown).
5884 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5885 for those rare installations that do not start the daemon as root or run it
5886 setuid root. I've cut out the call to initgroups() if the daemon is not
5889 60. The Exim user and group can now be bound into the binary as text strings
5890 that are looked up at the start of Exim's processing.
5892 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5894 62. Added $mailstore_basename variable.
5896 63. Installed patch to sieve.c from Michael Haardt.
5898 64. When Exim failed to open the panic log after failing to open the main log,
5899 the original message it was trying to log was written to stderr and debug
5900 output, but if they were not available (the usual case in production), it
5901 was lost. Now it is written to syslog before the two lines that record the
5902 failures to open the logs.
5904 65. Users' Exim filters run in subprocesses under the user's uid. It is
5905 possible for a "deliver" command or an alias in a "personal" command to
5906 provoke an address rewrite. If logging of address rewriting is configured,
5907 this fails because the process is not running as root or exim. There may be
5908 a better way of dealing with this, but for the moment (because 4.30 needs
5909 to be released), I have disabled address rewrite logging when running a
5910 filter in a non-root, non-exim process.
5916 1. The buildconfig auxiliary program wasn't quoting the value set for
5917 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5918 not defined. This bug was masked in 4.22 by the effect that was fixed in
5921 2. Some messages that were rejected after a message id was allocated were
5922 shown as "incomplete" by exigrep. It no longer does this for messages that
5923 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5925 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5926 have allow_domain_literals set, the ID did not get logged in the <= line.
5927 Domain literals are now always recognized in Message-ID: header lines.
5929 4. The first argument for a ${extract expansion item is the key name or field
5930 number. Leading and trailing spaces in this item were not being ignored,
5931 causing some misleading effects.
5933 5. When deliver_drop_privilege was set, single queue runner processes started
5934 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5935 same command in the process it spins off) were not dropping privilege.
5937 6. When the daemon running as "exim" started a queue runner, it always
5938 re-executed Exim in the spun-off process. This is a waste of effort when
5939 deliver_drop_privilege is set. The new process now just calls the
5940 queue-runner function directly.
5946 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5949 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5950 ignored. Though the use of -oP was forcing the writing of a pid file, it
5951 was always written to the default place.
5953 3. If the message "no IP address found for host xxxx" is generated during
5954 incoming verification, it is now followed by identification of the incoming
5955 connection (so you can more easily find what provoked it).
5957 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5959 5. Added some features to "harden" Exim a bit more against certain attacks:
5961 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5962 be put in Local/Makefile. This is like the never_users runtime option,
5963 but it cannot be overridden. The default setting is "root".
5965 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5966 prefix string with which any file named in a -C command line option
5969 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5970 is retained for -C and -D only if the caller of Exim is root. Without
5971 it, the exim user may also use -C and -D and retain privilege.
5973 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5974 command line option is disabled.
5976 6. Macro names set by the -D option must start with an upper case letter, just
5977 like macro names defined in the configuration file.
5979 7. Added "dereference=" facility to LDAP.
5981 8. Two instances of the typo "uknown" in the source files are fixed.
5983 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5984 the Configure-Makefile script screwed up while processing it.
5986 10. Incorporated PCRE 4.4.
5988 11. The SMTP synchronization check was not operating right at the start of an
5989 SMTP session. For example, it could not catch a HELO sent before the client
5990 waited for the greeting. There is now a check for outstanding input at the
5991 point when the greeting is written. Because of the duplex, asynchronous
5992 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5993 way, but not yet received, when the check is performed.
5995 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5996 on TCP/IP sockets, because this apparently causes some broken clients to
5999 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6000 unchanged) from the Cygwin maintainer.
6002 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6003 was supported instead of testing for mbx. Effectively a typo.
6005 15. The spa authenticator server code was not checking that the input it
6006 received was valid base64.
6008 16. The debug output line for the "set" modifier in ACLs was not showing the
6009 name of the variable that was being set.
6011 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6013 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6014 lookup on its own. Something else was needed to trigger the lookup. For
6015 example, a match in host_lookup or the need for a host name in a host list.
6016 Now, if $sender_host_name is referenced and the host name has not yet been
6017 looked up, a lookup is performed. If the lookup fails, the variable remains
6018 empty, and $host_lookup_failed is set to "1".
6020 19. Added "eqi" as a case-independent comparison operator.
6022 20. The saslauthd authentication condition could segfault if neither service
6023 nor realm was specified.
6025 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6026 error message that was logged was misleading, and incoming SMTP
6027 connections were dropped. The message is now more accurate, and temporary
6028 errors are given to SMTP connections.
6030 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6031 (except RSET) with a 421 error, until QUIT is received. However, it was
6032 failing to send a response to QUIT.
6034 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6035 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6036 ACL is now run after the helo_try_verify_hosts code.
6038 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6039 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6040 case-independent, so other case variants are also recognized). Apparently
6041 some systems use these upper case variants.
6043 25. If more than two messages were waiting for the same host, and a transport
6044 filter was specified for the transport, Exim sent two messages over the
6045 same TCP/IP connection, and then failed with "socket operation on non-
6046 socket" when it tried to send the third.
6048 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6050 27. The extern definition of crypt16() in expand.c was not being excluded when
6051 the OS had its own crypt16() function.
6053 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6054 as a preferred synonym for return_size_limit, both as an option and as an
6057 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6059 30. Changed the default configuration ACL to relax the local part checking rule
6060 for addresses that are not in any local domains. For these addresses,
6061 slashes and pipe symbols are allowed within local parts, but the sequence
6062 /../ is explicitly forbidden.
6064 31. SPA server authentication was not clearing the challenge buffer before
6067 32. log_message in a "warn" ACL statement was writing to the reject log as
6068 well as to the main log, which contradicts the documentation and doesn't
6069 seem right (because no rejection is happening). So I have stopped it.
6071 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6072 However, I am unable to do any testing of this.
6074 34. Fixed an infelicity in the appendfile transport. When checking directories
6075 for a mailbox, to see if any needed to be created, it was accidentally
6076 using path names with one or more superfluous leading slashes; tracing
6077 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6079 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6080 contents are added to the log line that is written for every discarded
6081 recipient. (Previously a log_message setting was ignored.)
6083 36. The ${quote: operator now quotes the string if it is empty.
6085 37. The install script runs exim in order to find its version number. If for
6086 some reason other than non-existence or emptiness, which it checks, it
6087 could not run './exim', it was installing it with an empty version number,
6088 i.e. as "exim-". This error state is now caught, and the installation is
6091 38. An argument was missing from the function that creates an error message
6092 when Exim fails to connect to the socket for saslauthd authentication.
6093 This could cause Exim to crash, or give a corrupted message.
6095 39. Added isip, isip4, and isip6 to ${if conditions.
6097 40. The ACL variables $acl_xx are now saved with the message, and can be
6098 accessed later in routers, transports, and filters.
6100 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6101 strings in the file are not string-expanded.
6103 42. If a MAIL command specified a SIZE value that was too large to fit into an
6104 int variable, the check against message_size_limit failed. Such values are
6105 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6106 day this will have to be increased, but I don't think I want to be around
6107 when emails are that large.
6114 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6115 iconv() is not standard in FreeBSD.
6117 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6118 IPv6 enabled. The observed symptom was a segmentation fault on return from
6119 the function os_common_find_running_interfaces() in src/os.c.
6121 3. In the check_special_case() function in daemon.c I had used "errno" as an
6122 argument name, which causes warnings on some systems. This was basically a
6123 typo, since it was named "eno" in the comments!
6125 4. The code that waits for the clock to tick (at a resolution of some fraction
6126 of a second) so as to ensure message-id uniqueness was always waiting for
6127 at least one whole tick, when it could have waited for less. [This is
6128 almost certainly not relevant at current processor speeds, where it is
6129 unlikely to ever wait at all. But we try to future-proof.]
6131 5. The function that sleeps for a time interval that includes fractions of a
6132 second contained a race. It did not block SIGALRM between setting the
6133 timer, and suspending (a couple of lines later). If the interval was short
6134 and the sigsuspend() was delayed until after it had expired, the suspension
6135 never ended. On busy systems this could lead to processes getting stuck for
6138 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6139 process, before it forks any delivery processes. The open lookup caching
6140 mechanism meant that the open file or database connection was passed into
6141 the delivery process. The problem was that delivery processes always tidy
6142 up cached lookup data. This could cause a problem for the next delivery
6143 process started by the queue runner, because the external queue runner
6144 process does not know about the closure. So the next delivery process
6145 still has data in the lookup cache. In the case of a file lookup, there was
6146 no problem because closing a file descriptor in a subprocess doesn't affect
6147 the parent. However, if the lookup was caching a connection to a database,
6148 the connection was closed, and the second delivery process was likely to
6149 see errors such as "PGSQL: query failed: server closed the connection
6150 unexpectedly". The problem has been fixed by closing all cached lookups
6151 in a queue runner before running a delivery process.
6153 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6154 seem to have the "const" qualifier which it has on other OS. I've
6157 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6158 *of the same type* (client or server) with the same public name that an
6159 error should be diagnosed.
6161 9. When Exim looked up a host name for an IP address, but failed to find the
6162 original IP address when looking up the host name (a safety check), it
6163 output the message "<ip address> does not match any IP for NULL", which was
6164 confusing, to say the least. The bug was that the host name should have
6165 appeared instead of "NULL".
6167 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6168 user that is built into the binary, and the -C or -D options is used, root
6169 privilege is dropped before the configuration file is read. In addition,
6170 logging is switched to stderr instead of the normal log files. If the
6171 configuration then re-defines the Exim user, the unprivileged environment
6172 is probably not what is expected, so Exim logs a panic warning message (but
6175 However, if deliver_drop_privilege is set, the unprivileged state may well
6176 be exactly what is intended, so the warning has been cut out in that case,
6177 and Exim is allowed to try to write to its normal log files.
6183 1. smtp_return_error_details was not giving details for temporary sender
6184 or receiver verification errors.
6186 2. Diagnose a configuration error if two authenticators have the same public
6189 3. Exim used not to create the message log file for a message until the first
6190 delivery attempt. This could be confusing when incoming messages were held
6191 for policy or load reasons. The message log file is now created at the time
6192 the message is received, and an initial "Received" line is written to it.
6194 4. The automatically generated man page for command line options had a minor
6195 bug that caused no ill effects; however, a more serious problem was that
6196 the procedure for building the man page automatically didn't always
6197 operate. Consequently, release 4.20 contains an out-of-date version. This
6198 shouldn't happen again.
6200 5. When building Exim with embedded Perl support, the script that builds the
6201 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6202 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6204 6. The freeze_tell option was not being used for messages that were frozen on
6205 arrival, either by an ACL or by local_scan().
6207 7. Added the smtp_incomplete_transaction log selector.
6209 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6210 was accepting AUTH without a new EHLO.
6212 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6213 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6216 10. Logging of TCP/IP connections (when configured) now happens in the main
6217 daemon process instead of the child process, so that the TCP/IP connection
6218 count is more accurate (but it can never be perfect).
6220 11. The use of "drop" in a nested ACL was not being handled correctly in the
6221 outer ACL. Now, if condition failure induced by the nested "drop" causes
6222 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6223 or "require"), the connection is dropped.
6225 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6226 that yield "discard" can now be used with an "accept" or a "discard" verb,
6227 but an error is generated for any others (because I can't see a useful way
6228 to define what should happen).
6230 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6231 are now processed in the same way as lines in the Exim configuration file.
6232 In particular, continuation lines are supported.
6234 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6236 15. Added -ti meaning -t -i.
6238 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6239 domains, and warn by logging if others are found.
6241 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6242 returned from GIFCONF when Exim is trying to find the list of interfaces on
6243 a host. The code in os.c has been modified to copy each ifreq to an aligned
6244 structure in all cases.
6246 Also, in some cases, the returned ifreq's were being copied to a 'struct
6247 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6248 means the last couple of bytes of an IPv6 address could be chopped if the
6249 ifreq contained only a normal sockaddr (14 bytes storage).
6251 18. Named domain lists were not supported in the hosts_treat_as_local option.
6252 An entry such as +xxxx was not recognized, and was treated as a literal
6255 19. Ensure that header lines added by a DATA ACL are included in the reject log
6256 if the ACL subsequently rejects the message.
6258 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6259 MD5 (which is deprecated).
6261 21. When testing a filter file using -bf, Exim was writing a message when it
6262 took the sender from a "From " line in the message, but it was not doing so
6263 when it took $return_path from a Return-Path: header line. It now does.
6265 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6266 with a valid header line field name (a series of printing characters
6267 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6269 23. Changed "disc" in the source to "disk" to conform to the documentation and
6270 the book and for uniformity.
6272 24. Ignore Sendmail's -Ooption=value command line item.
6274 25. When execve() failed while trying to run a command in a pipe transport,
6275 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6276 could be confused with a return value of 69 from the command itself. This
6277 has been changed to 127, the value the shell returns if it is asked to run
6278 a non-existent command. The wording for the related log line suggests a
6279 non-existent command as the problem.
6281 26. If received_header_text expands to an empty string, do not add a Received:
6282 header line to the message. (Well, it adds a token one on the spool, but
6283 marks it "old" so that it doesn't get used or transmitted.)
6285 27. Installed eximstats 1.28 (addition of -nt option).
6287 28. There was no check for failure on the call to getsockname() in the daemon
6288 code. This can fail if there is a shortage of resources on the system, with
6289 ENOMEM, for example. A temporary error is now given on failure.
6291 29. Contrary to the C standard, it seems that in some environments, the
6292 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6293 program. Exim now does this explicitly; it affects the formatting of
6294 timestamps using strftime().
6296 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6297 complaints. I've now initialized all the variables, to avoid this.
6299 32. Header lines added by a system filter were not being "seen" during
6300 transport-time rewrites.
6302 33. The info_callback() function passed to OpenSSL is set up with type void
6303 (*)(SSL *, int, int), as described somewhere. However, when calling the
6304 function (actually a macro) that sets it up, the type void(*)() is
6305 expected. I've put in a cast to prevent warnings from picky compilers.
6307 34. If a DNS black list lookup found a CNAME record, but there were no A
6308 records associated with the domain it pointed at, Exim crashed.
6310 35. If a DNS black list lookup returned more than one A record, Exim ignored
6311 all but the first. It now scans all returned addresses if a particular IP
6312 value is being sought. In this situation, the contents of the
6313 $dnslist_value variable are a list of all the addresses, separated by a
6316 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6317 to accept a host name and all its aliases if the forward lookup for any of
6318 them yielded the IP address of the incoming connection. Now it accepts only
6319 those names whose forward lookup yields the correct IP address. Any other
6320 names are discarded. This closes a loophole whereby a rogue DNS
6321 administrator could create reverse DNS records to break through a
6322 wildcarded host restriction in an ACL.
6324 37. If a user filter or a system filter that ran in a subprocess used any of
6325 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6326 the wrong values were passed to the pipe command ($thisaddress had the
6327 value of $0, $0 had the value of $1, etc). This bug was introduced by
6328 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6330 38. Improved the line breaking for long SMTP error messages from ACLs.
6331 Previously, if there was no break point between 40 and 75 characters, Exim
6332 left the rest of the message alone. Two changes have been made: (a) I've
6333 reduced the minimum length to 35 characters; (b) if it can't find a break
6334 point between 35 and 75 characters, it looks ahead and uses the first one
6335 that it finds. This may give the occasional overlong line, but at least the
6336 remaining text gets split now.
6338 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6339 file descriptors might be low, and that setting 1000 would always raise it.
6340 It turns out that in some environments, the limit is already over 1000 and
6341 that lowering it causes trouble. So now Exim takes care not to decrease it.
6343 40. When delivering a message, the value of $return_path is set to $sender_
6344 address at the start of routing (routers may change the value). By an
6345 oversight, this default was not being set up when an address was tested by
6346 -bt or -bv, which affected the outcome if any router or filter referred to
6349 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6350 log only when "message" or "log_message" are set. However, if one of the
6351 conditions was an address verification, or a call to a nested ACL, the
6352 messages generated by the underlying test were being passed through. This
6353 no longer happens. The underlying message is available in $acl_verify_
6354 message for both "message" and "log_message" expansions, so it can be
6355 passed through if needed.
6357 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6358 new expansion $bh_ to give the encoded byte string without charset
6359 translation. Translation happens only if iconv() is available; HAVE_ICONV
6360 indicates this at build time. HEADERS_CHARSET gives the charset to
6361 translate to; headers_charset can change it in the configuration, and
6362 "headers charset" can change it in an individual filter file.
6364 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6365 that creates RFC 2047 encoded "words" labels them as that charset instead
6366 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6367 expansion operator; (ii) when Exim creates a From: line for a local
6368 message; (iii) when a header line is rewritten to include a "phrase" part.
6370 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6371 buggy, causing it to skip the first lines of messages whose message ID
6372 ended in 'D'. This would not have bitten before Exim release 4.14, because
6373 message IDs were unlikely to end in 'D' before then. The effect was to have
6374 incorrect size information for certain domains.
6376 45. #include "config.h" was missing at the start of the crypt16.c module. This
6377 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6380 46. If there was a timeout during a "random" callout check, Exim treated it as
6381 a failure of the random address, and carried on sending RSET and the real
6382 address. If the delay was just some slowness somewhere, the response to the
6383 original RCPT would be taken as a response to RSET and so on, causing
6384 mayhem of various kinds.
6386 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6387 when I implemented it. It didn't allow for the fact that some option values
6388 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6389 the right test for positive values.
6391 48. Domain names in DNS records are case-independent. Exim always looks them up
6392 in lower case. Some resolvers return domain names in exactly the case they
6393 appear in the zone file, that is, they may contain uppercase letters. Not
6394 all resolvers do this - some return always lower case. Exim was treating a
6395 change of case by a resolver as a change of domain, similar to a widening
6396 of a domain abbreviation. This triggered its re-routing code and so it was
6397 trying to route what was effectively the same domain again. This normally
6398 caused routing to fail (because the router wouldn't handle the domain
6399 twice). Now Exim checks for this case specially, and just changes the
6400 casing of the domain that it ultimately uses when it transmits the message
6403 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6406 50. If a filter generated a file delivery with a non-absolute name (possible if
6407 no home directory exists for the router), the forbid_file option was not
6410 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6411 the existing equality matching.
6413 52. Exim was using ints instead of ino_t variables in some places where it was
6414 dealing with inode numbers.
6416 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6417 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6418 called TMPDIR, and if it finds it is different, it changes its value.
6420 54. The smtp_printf() function is now made available to local_scan() so
6421 additional output lines can be written before returning. There is also an
6422 smtp_fflush() function to enable the detection of a dropped connection.
6423 The variables smtp_input and smtp_batched_input are exported to
6426 55. Changed the default runtime configuration: the message "Unknown user"
6427 has been removed from the ACL, and instead placed on the localuser router,
6428 using the cannot_route_message feature. This means that any verification
6429 failures that generate their own messages won't get overridden. Similarly,
6430 the "Unrouteable address" message that was in the ACL for unverifiable
6431 relay addresses has also been removed.
6433 56. Added hosts_avoid_esmtp to the smtp transport.
6435 57. The exicyclog script was not checking for the esoteric option
6436 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6437 will work only if exicyclog is run under the appropriate euid.
6439 58. Following a discussion on the list, the rules by which Exim recognises line
6440 endings on incoming messages have been changed. The -dropcr and drop_cr
6441 options are now no-ops, retained only for backwards compatibility. The
6442 following line terminators are recognized: LF CRLF CR. However, special
6443 processing applies to CR:
6445 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6446 nor a local message in the state where . is a terminator.
6448 (ii) If a bare CR is encountered in a header line, an extra space is added
6449 after the line terminator so as not to end the header. The reasoning
6450 behind this is that bare CRs in header lines are most likely either
6451 to be mistakes, or people trying to play silly games.
6453 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6454 was being incorrectly given as 18 bytes larger than it should have been.
6455 This is a VOB (very old bug).
6457 60. This may never have affected anything current, but just in case it has:
6458 When the local host is found other than at the start of a list of hosts,
6459 the local host, those with the same MX, and any that follow, are discarded.
6460 When the list in question was part of a longer list of hosts, the following
6461 hosts (not currently being processed) were also being discarded. This no
6462 longer happens. I'm not sure if this situation could ever has previously
6465 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6468 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6469 header. This is something that is recommended in a new Internet Draft, and
6470 is something that is documented as being done by Sendmail. There are two
6471 possible values. For messages generated by the autoreply transport, Exim
6474 Auto-Submitted: auto-replied
6476 whereas for all other generated messages (e.g. bounces) it adds
6478 Auto-Submitted: auto-generated
6480 63. The "personal" condition in filters now includes a test for the
6481 Auto-Submitted: header. If it contains the string "auto-" the message it
6482 not considered personal.
6484 64. Added rcpt_include_affixes as a generic transport option.
6486 65. Added queue_only_override (default true).
6488 66. Added the syslog_duplication option.
6490 67. If what should have been the first header line of a message consisted of
6491 a space followed by a colon, Exim was mis-interpreting it as a header line.
6492 It isn't of course - it is syntactically invalid and should therefore be
6493 treated as the start of the message body. The misbehaviour could have
6494 caused a number of strange effects, including loss of data in subsequent
6495 header lines, and spool format errors.
6497 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6498 client host had authenticated. This control can now be exercised by an ACL
6499 for more flexibility.
6501 69. By default, callouts do not happen when testing with -bh. There is now a
6502 variant, -bhc, which does actually run the callout code, including
6503 consulting and updating the callout cache.
6505 70. Added support for saslauthd authentication, courtesy of Alexander
6508 71. If statvfs() failed on the spool or log directories while checking their
6509 size for availability, Exim confusingly gave the error "space shortage".
6510 Furthermore, in debugging mode it crashed with a floating point exception.
6511 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6512 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6513 serious problem, Exim now writes to the main and panic logs when this
6514 happens, with details of the failure. It then refuses to accept the
6515 incoming message, giving the message "spool directory problem" or "log
6516 directory problem" with a 421 code for SMTP messages.
6518 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6519 0, 1, and 2 exist, because some OS complain for execs without them (see
6520 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6521 descriptors. However, the code omitted to check that the open succeeded,
6522 causing mysterious errors if for some reason the permissions on /dev/null
6523 got screwed. Now Exim writes a message to the main and panic logs, and
6524 bombs out if it can't open /dev/null.
6526 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6527 interact so that it is all more flexible. It is supposed to remain
6528 backwards compatible. Also added extra_local_interfaces.
6530 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6531 to bomb out with an assertion failure - to the client this appears as a
6532 connection drop. This problem occurs in the part of the code that was taken
6533 from the Samba project. Fortunately, the assertion is in a very simple
6534 function, so I have fixed this by reproducing the function inline in the
6535 one place where it is called, and arranging for authentication to fail
6536 instead of killing the process with assert().
6538 75. The SPA client code was not working when the server requested OEM rather
6539 than Unicode encoding.
6541 76. Added code to make require_files with a specific uid setting more usable in
6542 the case where statting the file as root fails - usually a non-root-mounted
6543 NFS file system. When this happens and the failure is EACCES, Exim now
6544 forks a subprocess and does the per-uid checking as the relevant uid.
6546 77. Added process_log_path.
6548 78. If log_file_path was not explicitly set, a setting of check_log_space or
6549 check_log_inodes was ignored.
6551 79. If a space check for the spool or log partitions fails, the incident is now
6552 logged. Of course, in the latter case the data may get lost...
6554 80. Added the %p formatting code to string_format() so that it can be used to
6555 print addresses in debug_print(). Adjusted all the address printing in the
6556 debugging in store.c to use %p rather than %d.
6558 81. There was a concern that a line of code in smtp_in.c could overflow a
6559 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6560 initially expressed, the concern was not well-founded, because trailing
6561 spaces are removed early. However, if the trailing spaces were followed by
6562 a NULL, they did not get removed, so the overflow was possible. Two fixes
6565 (a) I re-wrote the offending code in a cleaner fashion.
6566 (b) If an incoming SMTP command contains a NULL character, it is rejected
6569 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6570 runs initgroups(), so that if the Exim user is in any additional groups,
6571 they will be used during message reception.
6577 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob