1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
12 than -2003 did; needs libidn2 in addition to libidn.
14 JH/02 The path option on a pipe transport is now expanded before use.
16 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
17 Patch provided by "Björn", documentation fix added too.
19 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
20 missing a wire-to-host endian conversion.
22 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
23 close after a BDAT command line could be taken as a following command,
24 giving a synch failure. Fix by only checking for synch immediately
25 before acknowledging the chunk.
27 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
28 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
29 macro. Patches provided by Josh Soref.
31 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
32 Previously we did not; the RFC seems ambiguous and VRFY is not listed
33 by IANA as a service extension. However, John Klensin suggests that we
36 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
37 the dkim code may be unix-mode line endings rather than smtp wire-format
38 CRLF, so prepend a CR to any bare LF.
40 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
41 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
43 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
44 the first were themselves being wrongly included in the feed into dkim
45 processing; with most chunk sizes in use this resulted in an incorrect
46 body hash calculated value.
48 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
49 DKIM signature block, for verification. Although advised against by
50 standards it is specifically not ruled illegal.
52 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
54 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
55 missing a body hash (the bh= tag).
57 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
58 It seems that HAProxy sends the Proxy Protocol information in clear and
59 only then does a TLS startup, so do the same.
61 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
62 TCP connections (such as for Spamd) unless the daemon successfully set
63 Fast Open mode on its listening sockets. This fixes breakage seen on
64 too-old kernels or those not configured for Fast Open, at the cost of
65 requiring both directions being enabled for TFO, and TFO never being used
66 by non-daemon-related Exim processes.
68 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
69 endings, at least on the first header line. Try to canonify any that get
70 past that check, despite the cost.
72 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
73 now limited to an arbitrary five deep, while parsing addresses with the
74 strip_excess_angle_brackets option enabled.
76 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
77 instead leave the unprompted TLS handshake in socket buffer for the
78 TLS library to consume.
80 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
82 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
84 JH/16 Drop variables when they go out of scope. Memory management drops a whole
85 region in one operation, for speed, and this leaves assigned pointers
86 dangling. Add checks run only under the testsuite which checks all
87 variables at a store-reset and panics on a dangling pointer; add code
88 explicitly nulling out all the variables discovered. Fixes one known
89 bug: a transport crash, where a dangling pointer for $sending_ip_address
90 originally assigned in a verify callout, is re-used.
92 PP/06 Drop '.' from @INC in various Perl scripts.
94 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
96 PP/08 Reduce a number of compilation warnings under clang; building with
97 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
98 should be warning-free.
100 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
102 HS/01 Fix portability problems introduced by PP/08 for platforms where
103 realloc(NULL) is not equivalent to malloc() [SunOS et al].
105 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
106 chunk. This allows us to accept broken chunked messages. We need a more
107 general solution here.
109 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
110 already-broken messages in the queue.
112 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
114 JH/19 Fix reference counting bug in routing-generated-address tracking.
120 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
121 supports it and a size is available (ie. the sending peer gave us one).
123 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
124 years of being deprecated). The replacements are the ACLs
125 acl_smtp_mime and acl_not_smtp_mime.
127 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
128 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
129 or even an in-clear connection were permitted. Now, if the host lookup
130 was dnssec and dane was requested then the host is only used if the
131 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
132 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
133 if one fails this test.
134 This means that a poorly-configured remote DNS will make it incommunicado;
135 but it protects against a DNS-interception attack on it.
137 JH/04 Bug 1810: make continued-use of an open smtp transport connection
138 non-noisy when a race steals the message being considered.
140 JH/05 If main configuration option tls_certificate is unset, generate a
141 self-signed certificate for inbound TLS connections.
143 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
144 in rewrites and routers.
146 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
147 and logged a warning sing 4.83; now they are a configuration file error.
149 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
150 (lacking @domain). Apply the same qualification processing as RCPT.
152 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
154 JH/10 Support ${sha256:} applied to a string (as well as the previous
157 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
158 a cutthrough deliver is pending, as we always want to make a connection.
159 This also avoids re-routing the message when later placing the cutthrough
160 connection after a verify cache hit.
161 Do not update it with the verify result either.
163 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
164 when routing results in more than one destination address.
166 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
167 signing (which inhibits the cutthrough capability). Previously only
168 the presence of an option was tested; now an expansion evaluating as
169 empty is permissible (obviously it should depend only on data available
170 when the cutthrough connection is made).
172 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
173 the relevant preceding SMTP command did not note the pipelining mode.
175 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
176 Previously they were not counted.
178 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
179 as one having no matching records. Previously we deferred the message
180 that needed the lookup.
182 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
183 distinguished as "(=".
185 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
186 for missing MX records. Previously it only worked for missing A records.
188 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
190 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
191 after the data-go-ahead and data-ack. Patch from Jason Betts.
193 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
194 even for a "none" policy. Patch from Tony Meyer.
196 JH/22 Fix continued use of a connection for further deliveries. If a port was
197 specified by a router, it must also match for the delivery to be
200 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
201 When one of the recipients of a message was unsuitable for the connection
202 (has no matching addresses), we lost track of needing to mark it
203 deferred. As a result mail would be lost.
205 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
207 JH/25 Decoding ACL controls is now done using a binary search; the source code
208 takes up less space and should be simpler to maintain. Merge the ACL
209 condition decode tables also, with similar effect.
211 JH/26 Fix problem with one_time used on a redirect router which returned the
212 parent address unchanged. A retry would see the parent address marked as
213 delivered, so not attempt the (identical) child. As a result mail would
216 JH/27 Fix a possible security hole, wherein a process operating with the Exim
217 UID can gain a root shell. Credit to http://www.halfdog.net/ for
218 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
221 JH/28 Enable {spool,log} filesystem space and inode checks as default.
222 Main config options check_{log,spool}_{inodes,space} are now
223 100 inodes, 10MB unless set otherwise in the configuration.
225 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
226 Previously it only applied to the main-section connection policy
229 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
231 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
232 by me. Added RFC7919 DH primes as an alternative.
234 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
235 are not in the system include path.
237 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
238 GnuTLS, when a session startup failed (eg because the client disconnected)
239 Exim did stdio operations after fclose. This was exposed by a recent
240 change which nulled out the file handle after the fclose.
242 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
243 signed directly by the cert-signing cert, rather than an intermediate
244 OCSP-signing cert. This is the model used by LetsEncrypt.
246 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
248 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
249 an incoming connection.
251 HS/02 Bug 1802: Do not half-close the connection after sending a request
254 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
255 fallback to "prime256v1".
257 JH/34 SECURITY: Use proper copy of DATA command in error message.
258 Could leak key material. Remotely exploitable. CVE-2016-9963.
264 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
265 and 3.4.4 - once the server is enabled to respond to an OCSP request
266 it does even when not requested, resulting in a stapling non-aware
267 client dropping the TLS connection.
269 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
270 support variable-length bit vectors. No functional change.
272 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
273 The I= interface field on outgoing lines is now after the H= remote
274 host field, same as incoming lines. There is a separate
275 outgoing_interface log selector which allows you to disable the
278 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
279 If not running log_selector +smtp_connection the mainlog would be held
280 open indefinitely after a "too many connections" event, including to a
281 deleted file after a log rotate. Leave the per net connection logging
282 leaving it open for efficiency as that will be quickly detected by the
283 check on the next write.
285 HS/01 Bug 1671: Fix post transport crash.
286 Processing the wait-<transport> messages could crash the delivery
287 process if the message IDs didn't exist for some reason. When
288 using 'split_spool_directory=yes' the construction of the spool
289 file name failed already, exposing the same netto behaviour.
291 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
292 mime_regex ACL conditions.
294 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
295 to DSN fail messages (bounces): remote IP, remote greeting, remote response
296 to HELO, local diagnostic string.
298 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
299 log line to debug. Even when configured with a tls authenticator many
300 client connections are expected to not authenticate in this way, so
301 an authenticate fail is not an error.
303 HS/02 Add the Exim version string to the process info. This way exiwhat
304 gives some more detail about the running daemon.
306 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
307 matter for fast-change records such as DNSBLs.
309 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
310 retry record, even if constant. There may be multiple transports with
311 different interface settings and the retry behaviour needs to be kept
314 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
316 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
318 JH/10 Bug 840: fix log_defer_output option of pipe transport
320 JH/11 Bug 830: use same host for all RCPTS of a message, even under
321 hosts_randomize. This matters a lot when combined with mua_wrapper.
323 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
324 ${quote_pgsql:<string>} operator.
326 JH/13 Bug 1708: avoid misaligned access in cached lookup.
328 JH/14 Change header file name for freeradius-client. Relevant if compiling
329 with Radius support; from the Gentoo tree and checked under Fedora.
331 JH/15 Bug 1712: Introduce $prdr_requested flag variable
333 JH/16 Bug 1714: Permit an empty string as expansion result for transport
334 option transport_filter, meaning no filtering.
336 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
338 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
339 defaults to "*" (all hosts). The variable is now available when not built
340 with TLS, default unset, mainly to enable keeping the testsuite sane.
341 If a server certificate is not supplied (via tls_certificate) an error is
342 logged, and clients will find TLS connections fail on startup. Presumably
343 they will retry in-clear.
344 Packagers of Exim are strongly encouraged to create a server certificate
345 at installation time.
347 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
348 with the $config_file variable.
350 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
351 in transport context, after the attempt, and per-recipient. The latter type
352 is per host attempted. The event data is the error message, and the errno
353 information encodes the lookup type (A vs. MX) used for the (first) host,
354 and the trailing two digits of the smtp 4xx response.
356 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
357 to write to mainlog (or rejectlog, paniclog) in the window between file
358 creation and permissions/ownership being changed. Particularly affects
359 installations where exicyclog is run as root, rather than exim user;
360 result is that the running daemon panics and dies.
362 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
364 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
365 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
366 "pri" and "weight". Note that the previous implicit priority given by the
367 list order is no longer honoured.
369 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
372 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
373 by defining SUPPORT_SOCKS.
375 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
376 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
377 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
378 variables are renamed to proxy_{local,external}_{address,port}.
380 JH/27 Move Internationalisation support from Experimental to mainline, enabled
381 for a build by defining SUPPORT_I18N
383 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
384 of the query string, and make ${quote_redis:} do that quoting.
386 JH/29 Move Events support from Experimental to mainline, enabled by default
387 and removable for a build by defining DISABLE_EVENT.
389 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
391 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
392 cached by the daemon.
394 JH/32 Move Redis support from Experimental to mainline, enabled for a build
395 by defining LOOKUP_REDIS. The libhiredis library is required.
397 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
398 keys are given for lookup.
400 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
401 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
402 only supported when built with TLS support. The PolarSSL SHA routines
403 are still used when the TLS library is too old for convenient support.
405 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
406 openssl_options), for security. OpenSSL forces this from version 1.1.0
407 server-side so match that on older versions.
409 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
410 allocation for $value could be released as the expansion processing
411 concluded, but leaving the global pointer active for it.
413 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
414 and to use the domains and local_parts ACL conditions.
416 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
417 incorrectly not doubled on cutthrough transmission, hence seen as a
418 body-termination at the receiving system - resulting in truncated mails.
419 Commonly the sender saw a TCP-level error, and retransmitted the message
420 via the normal store-and-forward channel. This could result in duplicates
421 received - but deduplicating mailstores were liable to retain only the
422 initial truncated version.
424 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
426 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
428 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
429 we're in there, support oversigning also; bug 1309.
431 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
433 HS/04 Add support for keep_environment and add_environment options.
435 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
436 either intentional arithmetic overflow during PRNG, or testing config-
439 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
440 delivery resulted in actual delivery. Cancel cutthrough before DATA
443 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
444 rejects a recipient: pass the reject to the originator.
446 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
447 Many were false-positives and ignorable, but it's worth fixing the
450 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
451 for the new environment-manipulation done at startup. Move the routines
452 from being local to tls.c to being global via the os.c file.
454 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
455 an extract embedded as result-arg for a map, the first arg for extract
456 is unavailable so we cannot tell if this is a numbered or keyed
457 extraction. Accept either.
463 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
466 JH/02 The smtp transport option "multi_domain" is now expanded.
468 JH/03 The smtp transport now requests PRDR by default, if the server offers
471 JH/04 Certificate name checking on server certificates, when exim is a client,
472 is now done by default. The transport option tls_verify_cert_hostnames
473 can be used to disable this per-host. The build option
474 EXPERIMENTAL_CERTNAMES is withdrawn.
476 JH/05 The value of the tls_verify_certificates smtp transport and main options
477 default to the word "system" to access the system default CA bundle.
478 For GnuTLS, only version 3.0.20 or later.
480 JH/06 Verification of the server certificate for a TLS connection is now tried
481 (but not required) by default. The verification status is now logged by
482 default, for both outbound TLS and client-certificate supplying inbound
485 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
488 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
489 Status Notification (bounce) messages are now MIME format per RFC 3464.
490 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
491 under the control of the dsn_advertise_hosts option, and routers may
492 have a dsn_lasthop option.
494 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
495 default, modifiable by a malware= option. The list separator for
496 the options can now be changed in the usual way. Bug 68.
498 JH/10 The smtp_receive_timeout main option is now expanded before use.
500 JH/11 The incoming_interface log option now also enables logging of the
501 local interface on delivery outgoing connections.
503 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
504 if the interface and destination host and port all match.
506 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
509 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
510 Patch from Andrew Lewis.
512 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
513 now supports optional time-restrictions, weighting, and priority
514 modifiers per server. Patch originally by <rommer@active.by>.
516 JH/16 The spamd_address main option now supports a mixed list of local
517 and remote servers. Remote servers can be IPv6 addresses, and
518 specify a port-range.
520 JH/17 Bug 68: The spamd_address main option now supports an optional
521 timeout value per server.
523 JH/18 Bug 1581: Router and transport options headers_add/remove can
524 now have the list separator specified.
526 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
529 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
532 JH/21 Support for the A6 type of dns record is withdrawn.
534 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
535 rather than the verbs used.
537 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
538 from 255 to 1024 chars.
540 JH/24 Verification callouts now attempt to use TLS by default.
542 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
543 are generic router options now. The defaults didn't change.
545 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
546 Original patch from Alexander Shikoff, worked over by JH.
548 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
549 files if ZCAT_COMMAND is not executable.
551 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
553 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
555 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
556 Normally benign, it bites when the pair was led to by a CNAME;
557 modern usage is to not canonicalize the domain to a CNAME target
558 (and we were inconsistent anyway for A-only vs AAAA+A).
560 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
562 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
563 when evaluating $sender_host_dnssec.
565 JH/31 Check the HELO verification lookup for DNSSEC, adding new
566 $sender_helo_dnssec variable.
568 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
570 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
572 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
574 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
575 documented as working, but never had. Support all but $spam_report.
577 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
578 added for tls authenticator.
580 HS/03 Add perl_taintmode main config option
586 TL/01 When running the test suite, the README says that variables such as
587 no_msglog_check are global and can be placed anywhere in a specific
588 test's script, however it was observed that placement needed to be near
589 the beginning for it to behave that way. Changed the runtest perl
590 script to read through the entire script once to detect and set these
591 variables, reset to the beginning of the script, and then run through
592 the script parsing/test process like normal.
594 TL/02 The BSD's have an arc4random API. One of the functions to induce
595 adding randomness was arc4random_stir(), but it has been removed in
596 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
597 function when detected.
599 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
600 cause callback expansion.
602 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
603 syntax errors in an expansion can be treated as a string instead of
604 logging or causing an error, due to the internal use of bool_lax
605 instead of bool when processing it.
607 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
608 server certificates when making smtp deliveries.
610 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
612 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
614 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
616 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
617 Merged patch from Sebastian Wiedenroth.
619 JH/05 Fix results-pipe from transport process. Several recipients, combined
620 with certificate use, exposed issues where response data items split
621 over buffer boundaries were not parsed properly. This eventually
622 resulted in duplicates being sent. This issue only became common enough
623 to notice due to the introduction of connection certificate information,
624 the item size being so much larger. Found and fixed by Wolfgang Breyha.
626 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
627 size buffer was used, resulting in syntax errors when an expansion
630 JH/07 Add support for directories of certificates when compiled with a GnuTLS
631 version 3.3.6 or later.
633 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
634 is EXPERIMENTAL_EVENT, the main-configuration and transport options
635 both become "event_action", the variables become $event_name, $event_data
636 and $event_defer_errno. There is a new variable $verify_mode, usable in
637 routers, transports and related events. The tls:cert event is now also
638 raised for inbound connections, if the main configuration event_action
641 TL/06 In test suite, disable OCSP for old versions of openssl which contained
642 early OCSP support, but no stapling (appears to be less than 1.0.0).
644 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
645 server certificate names available under the smtp transport option
646 "tls_verify_cert_hostname" now do not permit multi-component wildcard
649 JH/10 Time-related extraction expansions from certificates now use the main
650 option "timezone" setting for output formatting, and are consistent
651 between OpenSSL and GnuTLS compilations. Bug 1541.
653 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
654 encoded parameter in the incoming message. Bug 1558.
656 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
657 include certificate info, eximon was claiming there were spoolfile
660 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
662 JH/14 Log delivery-related information more consistently, using the sequence
663 "H=<name> [<ip>]" wherever possible.
665 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
666 are problematic for Debian distribution, omit them from the release
669 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
671 JH/16 Fix string representation of time values on 64bit time_t architectures.
674 JH/17 Fix a null-indirection in certextract expansions when a nondefault
675 output list separator was used.
680 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
681 checkers that were complaining about end of non-void function with no
684 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
685 This was a regression introduced in 4.83 by another bugfix.
687 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
689 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
690 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
696 TF/01 Correctly close the server side of TLS when forking for delivery.
698 When a message was received over SMTP with TLS, Exim failed to clear up
699 the incoming connection properly after forking off the child process to
700 deliver the message. In some situations the subsequent outgoing
701 delivery connection happened to have the same fd number as the incoming
702 connection previously had. Exim would try to use TLS and fail, logging
703 a "Bad file descriptor" error.
705 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
706 utilities have not been installed.
708 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
709 temporary space as the ACL may create new global variables.
711 TL/01 LDAP support uses per connection or global context settings, depending
712 upon the detected version of the libraries at build time.
714 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
715 to extract and use the src ip:port in logging and expansions as if it
716 were a direct connection from the outside internet. PPv2 support was
717 updated based on HAProxy spec change in May 2014.
719 JH/02 Add ${listextract {number}{list}{success}{fail}}.
721 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
722 Properly escape header and check for NULL return.
724 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
727 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
729 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
730 characters in header names, implemented as a verify condition.
731 Contributed by Michael Fischer v. Mollard.
733 TL/05 Rename SPF condition results err_perm and err_temp to standardized
734 results permerror and temperror. Previous values are deprecated but
735 still accepted. In a future release, err_perm and err_temp will be
736 completely removed, which will be a backward incompatibility if the
737 ACL tests for either of these two old results. Patch contributed by
738 user bes-internal on the mailing list.
740 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
742 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
743 selectors, in both main and reject logs.
745 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
748 JH/07 Add malware type "sock" for talking to simple daemon.
750 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
752 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
753 routers/transports under cutthrough routing.
755 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
756 numbers. Touch up "bool" conditional to keep the same definition.
758 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
760 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
762 JH/12 Expand items in router/transport headers_add or headers_remove lists
763 individually rather than the list as a whole. Bug 1452.
765 Required for reasonable handling of multiple headers_ options when
766 they may be empty; requires that headers_remove items with embedded
767 colons must have them doubled (or the list-separator changed).
769 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
770 view the policy declared in the DMARC record. Currently, $dmarc_status
771 is a combined value of both the record presence and the result of the
774 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
776 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
777 dnslookup router and the smtp transport (applying to the forward
780 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
781 of ldap servers used for a specific lookup. Patch provided by Heiko
784 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
785 New variable $lookup_dnssec_authenticated for observability.
787 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
788 Patch submitted by Lars Timman.
790 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
792 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
793 Requires trusted mode and valid format message id, aborts otherwise.
794 Patch contributed by Heiko Schlichting.
796 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
797 certextract with support for various fields. Bug 1358.
799 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
800 is requested by default, modifiable by smtp transport option
803 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
804 operate on certificate variables to give certificate fingerprints
805 Also new ${sha256:cert_variable}.
807 JH/23 The PRDR feature is moved from being Experimental into the mainline.
809 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
810 Christian Aistleitner.
812 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
814 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
815 file. Patch from Wolfgang Breyha.
817 JH/25 Expand the coverage of the delivery $host and $host_address to
818 client authenticators run in verify callout. Bug 1476.
820 JH/26 Port service names are now accepted for tls_on_connect_ports, to
821 align with daemon_smtp_ports. Bug 72.
823 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
824 support and error reporting did not work properly.
826 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
827 and is readable. Patch from Andrew Colin Kissa.
829 TL/14 Enhance documentation of ${run expansion and how it parses the
830 commandline after expansion, particularly in the case when an
831 unquoted variable expansion results in an empty value.
833 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
835 PP/02 Fix internal collision of T_APL on systems which support RFC3123
836 by renaming away from it. Addresses GH issue 15, reported by
839 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
841 TL/15 SECURITY: prevent double expansion in math comparison functions
842 (can expand unsanitized data). Not remotely exploitable.
849 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
851 PP/02 Make -n do something, by making it not do something.
852 When combined with -bP, the name of an option is not output.
854 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
857 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
858 $sender_host_name and config options to manage this, and basic check
861 PP/05 DSCP support for outbound connections and control modifier for inbound.
863 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
864 (Only plugin which currently uses this is kerberos4, which nobody should
865 be using, but we should make it available and other future plugins might
866 conceivably use it, even though it would break NAT; stuff *should* be
867 using channel bindings instead).
869 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
870 name; added for Sendmail compatibility; requires admin caller.
871 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
872 just ignore it); requires trusted caller.
873 Also parse but ignore: -Ac -Am -X<logfile>
876 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
878 TL/02 Add +smtp_confirmation as a default logging option.
880 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
881 Patch by Magnus Holmgren from 2007-02-20.
883 TL/04 Bugzilla 1281 - Spec typo.
884 Bugzilla 1283 - Spec typo.
885 Bugzilla 1290 - Spec grammar fixes.
887 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
889 TL/06 Add Experimental DMARC support using libopendmarc libraries.
891 TL/07 Fix an out of order global option causing a segfault. Reported to dev
892 mailing list by by Dmitry Isaikin.
894 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
896 JH/02 Support "G" suffix to numbers in ${if comparisons.
898 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
900 NM/01 Bugzilla 1197 - Spec typo
901 Bugzilla 1196 - Spec examples corrections
903 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
905 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
906 gnutls_enable_pkcs11, but renamed to more accurately indicate its
909 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
910 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
912 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
913 "acl {{name}{arg}...}", and optional args on acl condition
916 JH/05 Permit multiple router/transport headers_add/remove lines.
918 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
920 JH/07 Avoid using a waiting database for a single-message-only transport.
921 Performance patch from Paul Fisher. Bugzilla 1262.
923 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
926 JH/09 Add $headers_added variable, with content from use of ACL modifier
927 add_header (but not yet added to the message). Bugzilla 199.
929 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
930 Pulled from Bugzilla 817 by Wolfgang Breyha.
932 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
934 (nb: this is the same fix as in Exim 4.80.1)
936 JH/11 Add A= logging on delivery lines, and a client_set_id option on
939 JH/12 Add optional authenticated_sender logging to A= and a log_selector
942 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
944 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
945 advertise SMTP AUTH mechanism to us, instead of a generic
946 protocol violation error. Also, make Exim more robust to bad
947 data from the Dovecot auth socket.
949 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
951 When a queue runner is handling a message, Exim first routes the
952 recipient addresses, during which it prunes them based on the retry
953 hints database. After that it attempts to deliver the message to
954 any remaining recipients. It then updates the hints database using
957 So if a recipient address works intermittently, it can get repeatedly
958 deferred at routing time. The retry hints record remains fresh so the
959 address never reaches the final cutoff time.
961 This is a fairly common occurrence when a user is bumping up against
962 their storage quota. Exim had some logic in its local delivery code
963 to deal with this. However it did not apply to per-recipient defers
964 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
966 This change adds a proper retry rule check during routing so that the
967 final cutoff time is checked against the message's age. We only do
968 this check if there is an address retry record and there is not a
969 domain retry record; this implies that previous attempts to handle
970 the address had the retry_use_local_parts option turned on. We use
971 this as an approximation for the destination being like a local
972 delivery, as in LMTP.
974 I suspect this new check makes the old local delivery cutoff check
975 redundant, but I have not verified this so I left the code in place.
977 TF/02 Correct gecos expansion when From: is a prefix of the username.
979 Test 0254 submits a message to Exim with the header
983 When I ran the test suite under the user fanf2, Exim expanded
984 the header to contain my full name, whereas it should have added
985 a Resent-Sender: header. It erroneously treats any prefix of the
986 username as equal to the username.
988 This change corrects that bug.
990 GF/01 DCC debug and logging tidyup
991 Error conditions log to paniclog rather than rejectlog.
992 Debug lines prefixed by "DCC: " to remove any ambiguity.
994 TF/03 Avoid unnecessary rebuilds of lookup-related code.
996 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
997 Bug spotted by Jeremy Harris; was flawed since initial commit.
998 Would have resulted in OCSP responses post-SNI triggering an Exim
999 NULL dereference and crash.
1001 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1003 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1004 Bug detection, analysis and fix by Samuel Thibault.
1005 Bugzilla 1331, Debian bug #698092.
1007 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1009 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1010 Server implementation by Todd Lyons, client by JH.
1011 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1012 config variable "prdr_enable" controls whether the server
1013 advertises the facility. If the client requests PRDR a new
1014 acl_data_smtp_prdr ACL is called once for each recipient, after
1015 the body content is received and before the acl_smtp_data ACL.
1016 The client is controlled by both of: a hosts_try_prdr option
1017 on the smtp transport, and the server advertisement.
1018 Default client logging of deliveries and rejections involving
1019 PRDR are flagged with the string "PRDR".
1021 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1022 fclose(). Diagnosis by Todd Lyons.
1024 PP/17 Update configure.default to handle IPv6 localhost better.
1025 Patch by Alain Williams (plus minor tweaks).
1028 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1029 This is now consistent with GnuTLS, and is now documented: the
1030 previous undocumented portable approach to treating the option as
1031 unset was to force an expansion failure. That still works, and
1032 an empty string is now equivalent.
1034 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1035 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1036 not performing validation itself.
1038 PP/20 Added force_command boolean option to pipe transport.
1039 Patch from Nick Koston, of cPanel Inc.
1041 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1044 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1046 PP/21 Fix eximon continuous updating with timestamped log-files.
1047 Broken in a format-string cleanup in 4.80, missed when I repaired the
1048 other false fix of the same issue.
1049 Report and fix from Heiko Schlichting.
1052 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1053 Report from Prashanth Katuri.
1055 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1056 It's SecureTransport, so affects any MacOS clients which use the
1057 system-integrated TLS libraries, including email clients.
1059 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1060 using a MIME ACL for non-SMTP local injection.
1061 Report and assistance in diagnosis by Warren Baker.
1063 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1065 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1067 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1068 last id that failed so it may be referenced in subsequent ACL's.
1070 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1073 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1074 ldap library initialization, allowing self-signed CA's to be
1075 used. Also properly sets require_cert option later in code by
1076 using NULL (global ldap config) instead of ldap handle (per
1077 session). Bug diagnosis and testing by alxgomz.
1079 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1080 the src/util/ subdirectory.
1082 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1083 renamed to Transport Post Delivery Action by Jeremy Harris, as
1086 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1087 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1088 redis_servers = needs to be configured which will be used by the redis
1089 lookup. Patch from Warren Baker, of The Packet Hub.
1091 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1093 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1094 hostname or reverse DNS when processing a host list. Used suggestions
1095 from multiple comments on this bug.
1097 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1099 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1100 Missed a few lines, added it to make the runtest require no keyboard
1103 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1104 contains upper case chars. Make router use caseful_local_part.
1106 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1107 support when GnuTLS has been built with p11-kit.
1113 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1115 This, or similar/improved, will also be change PP/11 of 4.82.
1121 PP/01 Handle short writes when writing local log-files.
1122 In practice, only affects FreeBSD (8 onwards).
1123 Bugzilla 1053, with thanks to Dmitry Isaikin.
1125 NM/01 Bugzilla 949 - Documentation tweak
1127 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1130 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1132 PP/02 Implemented gsasl authenticator.
1134 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1136 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1137 `pkg-config foo` for cflags/libs.
1139 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1140 with rest of GSASL and with heimdal_gssapi.
1142 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1143 `pkg-config foo` for cflags/libs for the TLS implementation.
1145 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1146 properties get this fed in as external SSF. A number of robustness
1147 and debugging improvements to the cyrus_sasl authenticator.
1149 PP/08 cyrus_sasl server now expands the server_realm option.
1151 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1152 Patch by Jeremy Harris.
1154 PP/10 Added dbmjz lookup type.
1156 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1158 PP/12 MAIL args handles TAB as well as SP, for better interop with
1159 non-compliant senders.
1160 Analysis and variant patch by Todd Lyons.
1162 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1163 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1164 Patch from Dirk Mueller <dmueller@suse.com>
1166 PP/13 tls_peerdn now print-escaped for spool files.
1167 Observed some $tls_peerdn in wild which contained \n, which resulted
1168 in spool file corruption.
1170 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1171 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1172 or write after TLS renegotiation, which otherwise led to messages
1175 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1176 as a tracking header (ie: a signed header comes before the signature).
1177 Patch from Wolfgang Breyha.
1179 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1180 comma-sep list; embedded commas doubled.
1182 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1184 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1186 Report and patch from Dmitry Banschikov.
1188 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1189 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1190 failures. We appear to now support TLS1.1+ with Exim.
1192 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1193 lets Exim select keys and certificates based upon TLS SNI from client.
1194 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1195 before an outbound SMTP session. New log_selector, +tls_sni.
1197 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1198 NULL dereference. Report and patch from Alun Jones.
1200 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1201 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1202 Not seeing resolver debug output on NetBSD, but suspect this is a
1203 resolver implementation change.
1205 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1206 Left warnings. Added "eximon gdb" invocation mode.
1208 PP/21 Defaulting "accept_8bitmime" to true, not false.
1210 PP/22 Added -bw for inetd wait mode support.
1212 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1213 locate the relevant includes and libraries. Made this the default.
1215 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1216 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1218 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1219 This may cause build issues on older platforms.
1221 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1222 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1223 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1224 Added SNI support via GnuTLS too.
1225 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1227 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1229 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1230 Applied second patch from Janne, implementing suggestion to default
1231 multiple-strings-in-record handling to match SPF spec.
1233 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1235 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1236 read-only, out of scope).
1237 Patch from Wolfgang Breyha, report from Stuart Northfield.
1239 PP/29 Fix three issues highlighted by clang analyser static analysis.
1240 Only crash-plausible issue would require the Cambridge-specific
1241 iplookup router and a misconfiguration.
1242 Report from Marcin Mirosław.
1244 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1246 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1247 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1248 As part of this, removing so much warning spew let me fix some minor
1249 real issues in debug logging.
1251 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1252 assignment on my part. Fixed.
1254 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1255 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1256 Janne Snabb (who went above and beyond: thank you).
1258 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1259 string otherwise requires a connection and a bunch more work and it's
1260 relatively easy to get wrong. Should also expose TLS library linkage
1263 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1264 64-bit ${eval} (JH/03).
1266 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1267 GNU libc to support some of the 64-bit stuff, should not lead to
1268 conflicts. Defined before os.h is pulled in, so if a given platform
1269 needs to override this, it can.
1271 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1272 protection layer was required, which is not implemented.
1273 Bugzilla 1254, patch from Wolfgang Breyha.
1275 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1276 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1277 tls_dhparam take prime identifiers. Also unbreak combination of
1278 OpenSSL+DH_params+TLSSNI.
1280 PP/39 Disable SSLv2 by default in OpenSSL support.
1286 PP/01 Solaris build fix for Oracle's LDAP libraries.
1287 Bugzilla 1109, patch from Stephen Usher.
1289 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1291 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1294 TF/02 Fix a couple more cases where we did not log the error message
1295 when unlink() failed. See also change 4.74-TF/03.
1297 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1298 lock up or crash if it happened to be inside a call to libc when it
1299 got a SIGUSR1 from exiwhat.
1301 The SIGUSR1 handler appends the current process status to the process
1302 log which is later printed by exiwhat. It used to use the general
1303 purpose logging code to do this, but several functions it calls are
1304 not safe for signals.
1306 The new output code in the SIGUSR1 handler is specific to the process
1307 log, and simple enough that it's easy to inspect for signal safety.
1308 Removing some special cases also simplifies the general logging code.
1309 Removing the spurious timestamps from the process log simplifies
1312 TF/04 Improved ratelimit ACL condition.
1314 The /noupdate option has been deprecated in favour of /readonly which
1315 has clearer semantics. The /leaky, /strict, and /readonly update modes
1316 are mutually exclusive. The update mode is no longer included in the
1317 database key; it just determines when the database is updated. (This
1318 means that when you upgrade Exim will forget old rate measurements.)
1320 Exim now checks that the per_* options are used with an update mode that
1321 makes sense for the current ACL. For example, when Exim is processing a
1322 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1323 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1324 must specify per_mail/readonly. If you omit the update mode it defaults to
1325 /leaky where that makes sense (as before) or /readonly where required.
1327 The /noupdate option is now undocumented but still supported for
1328 backwards compatibility. It is equivalent to /readonly except that in
1329 ACLs where /readonly is required you may specify /leaky/noupdate or
1330 /strict/noupdate which are treated the same as /readonly.
1332 A useful new feature is the /count= option. This is a generalization
1333 of the per_byte option, so that you can measure the throughput of other
1334 aggregate values. For example, the per_byte option is now equivalent
1335 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1337 The per_rcpt option has been generalized using the /count= mechanism
1338 (though it's more complicated than the per_byte equivalence). When it is
1339 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1340 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1341 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1342 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1343 Note that using per_rcpt with a non-readonly update mode in more than
1344 one ACL will cause the recipients to be double-counted. (The per_mail
1345 and per_byte options don't have this problem.)
1347 The handling of very low rates has changed slightly. If the computed rate
1348 is less than the event's count (usually one) then this event is the first
1349 after a long gap. In this case the rate is set to the same as this event's
1350 count, so that the first message of a spam run is counted properly.
1352 The major new feature is a mechanism for counting the rate of unique
1353 events. The new per_addr option counts the number of different
1354 recipients that someone has sent messages to in the last time period. It
1355 behaves like per_rcpt if all the recipient addresses are different, but
1356 duplicate recipient addresses do not increase the measured rate. Like
1357 the /count= option this is a general mechanism, so the per_addr option
1358 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1359 example, measure the rate that a client uses different sender addresses
1360 with the options per_mail/unique=$sender_address. There are further
1361 details in the main documentation.
1363 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1365 TF/06 Removed a few PCRE remnants.
1367 TF/07 Automatically extract Exim's version number from tags in the git
1368 repository when doing development or release builds.
1370 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1371 Bugzilla 879. Patch from Paul Fisher.
1373 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1374 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1377 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1379 PP/05 Variable $av_failed, true if the AV scanner deferred.
1380 Bugzilla 1078. Patch from John Horne.
1382 PP/06 Stop make process more reliably on build failure.
1383 Bugzilla 1087. Patch from Heiko Schlittermann.
1385 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1386 Bugzilla 1089. Patch from Heiko Schlittermann.
1388 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1389 Bugzilla 1131. Patch from Holger Weiß.
1391 PP/09 Handle IPv6 addresses with SPF.
1392 Bugzilla 860. Patch from Wolfgang Breyha.
1394 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1396 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1399 PP/11 match_* no longer expand right-hand-side by default.
1400 New compile-time build option, EXPAND_LISTMATCH_RHS.
1401 New expansion conditions, "inlist", "inlisti".
1403 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1405 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1407 PP/14 fix log_write() format string regression from TF/03.
1408 Bugzilla 1152. Patch from Dmitry Isaikin.
1414 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1416 PP/02 Harmonised TLS library version reporting; only show if debugging.
1417 Layout now matches that introduced for other libraries in 4.74 PP/03.
1419 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1421 PP/04 New "dns_use_edns0" global option.
1423 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1426 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1427 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1429 TK/01 Updated PolarSSL code to 0.14.2.
1430 Bugzilla 1097. Patch from Andreas Metzler.
1432 PP/07 Catch divide-by-zero in ${eval:...}.
1433 Fixes bugzilla 1102.
1435 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1438 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1439 format-string attack -- SECURITY: remote arbitrary code execution.
1441 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1442 time unintentionally subject to list matching rules, letting the header
1443 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1444 arbitrary string expansion). This allowed for information disclosure.
1446 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1447 INT_MIN/-1 -- value coerced to INT_MAX.
1453 NM/01 Workaround for PCRE version dependency in version reporting
1456 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1457 This fixes portability to compilers other than gcc, notably
1458 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1460 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1461 makefiles for portability to HP-UX and POSIX correctness.
1463 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1464 Also via indented variable definition in the Makefile.
1465 (Debugging by Oliver Heesakkers).
1467 PP/02 Restore caching of spamd results with expanded spamd_address.
1468 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1470 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1471 Improves build reliability. Fix from: Frank Elsner
1473 NM/02 Fix wide character breakage in the rfc2047 coding
1474 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1476 NM/03 Allow underscore in dnslist lookups
1477 Fixes bug 1026. Patch from Graeme Fowler
1479 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1480 Code patches from Adam Ciarcinski of NetBSD.
1482 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1485 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1486 is logged, to avoid truncation. Patch from John Horne.
1488 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1489 Patch from Jakob Hirsch.
1491 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1492 SQL string expansion failure details.
1493 Patch from Andrey Oktyabrski.
1495 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1496 Patch from Simon Arlott.
1498 PP/09 New lookups functionality failed to compile on old gcc which rejects
1499 extern declarations in function scope.
1500 Patch from Oliver Fleischmann
1502 PP/10 Use sig_atomic_t for flags set from signal handlers.
1503 Check getgroups() return and improve debugging.
1504 Fixed developed for diagnosis in bug 927 (which turned out to be
1507 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1508 Patch from Mark Zealey.
1510 PP/12 Bugzilla 1056: Improved spamd server selection.
1511 Patch from Mark Zealey.
1513 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1514 Based on patch from Heiko Schlittermann.
1516 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1517 Patch from Uwe Doering, sign-off by Michael Haardt.
1519 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1520 variable declaration deep within a block. Bug and patch from
1523 PP/15 lookups-Makefile IRIX compatibility coercion.
1525 PP/16 Make DISABLE_DKIM build knob functional.
1527 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1528 Patch by Simon Arlott
1530 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1531 variable argument macros. Our copy now differs from upstream.
1537 TF/01 Failure to get a lock on a hints database can have serious
1538 consequences so log it to the panic log.
1540 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1541 controlled using the smtp_confirmation log selector.
1543 TF/03 Include the error message when we fail to unlink a spool file.
1545 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1546 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1547 for maintaining out-of-tree patches for some time.
1549 PP/01 Bugzilla 139: Documentation and portability issues.
1550 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1551 Handle per-OS dynamic-module compilation flags.
1553 PP/02 Let /dev/null have normal permissions.
1554 The 4.73 fixes were a little too stringent and complained about the
1555 permissions on /dev/null. Exempt it from some checks.
1556 Reported by Andreas M. Kirchwitz.
1558 PP/03 Report version information for many libraries, including
1559 Exim version information for dynamically loaded libraries. Created
1560 version.h, now support a version extension string for distributors
1561 who patch heavily. Dynamic module ABI change.
1563 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1564 privilege escalation vulnerability whereby the Exim run-time user
1565 can cause root to append content of the attacker's choosing to
1568 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1571 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1572 If dropping privileges for untrusted macros, we disabled normal logging
1573 on the basis that it would fail; for the Exim run-time user, this is not
1574 the case, and it resulted in successful deliveries going unlogged.
1575 Fixed. Reported by Andreas Metzler.
1581 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1582 only prepend for the Resent-* case. Fixes regression introduced in
1583 Exim 4.70 by NM/22 for Bugzilla 607.
1585 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1586 increasing numbers of administrators be bitten by this.
1588 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1590 PP/03 Bugzilla 994: added openssl_options main configuration option.
1592 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1594 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1596 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1598 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1599 without a peer certificate, leading to a segfault because of an
1600 assumption that peers always have certificates. Be a little more
1601 paranoid. Problem reported by Martin Tscholak.
1603 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1604 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1605 NB: ClamAV planning to remove STREAM in "middle of 2010".
1606 CL also introduces -bmalware, various -d+acl logging additions and
1607 more caution in buffer sizes.
1609 PP/09 Implemented reverse_ip expansion operator.
1611 PP/10 Bugzilla 937: provide a "debug" ACL control.
1613 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1615 PP/12 Bugzilla 973: Implement --version.
1617 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1619 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1621 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1623 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1624 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1625 ignore trailing whitespace.
1627 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1629 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1632 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1633 Notification from Dr Andrew Aitchison.
1635 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1636 ExtendedDetectionInfo response format.
1637 Notification from John Horne.
1639 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1642 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1643 XSL and documented dependency on system catalogs, with examples of how
1646 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1649 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1650 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1651 configuration file which is writeable by the Exim user or group.
1653 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1654 of configuration files to cover files specified with the -C option if
1655 they are going to be used with root privileges, not just the default
1658 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1659 option (effectively making it always true).
1661 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1662 files to be used while preserving root privileges.
1664 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1665 that rogue child processes cannot use them.
1667 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1668 run-time user, instead of root.
1670 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1671 Exim run-time user without dropping privileges.
1673 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1674 result string, instead of calling string_vformat() twice with the same
1677 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1678 for other users. Others should always drop root privileges if they use
1679 -C on the command line, even for a whitelisted configure file.
1681 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1683 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1689 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1690 $data_path, and $header_path variables; fixed documentation bugs and
1693 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1694 exipick to access non-standard spools, including the "frozen" queue
1697 NM/01 Bugzilla 965: Support mysql stored procedures.
1698 Patch from Alain Williams
1700 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1702 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1703 Patch from Andreas Metzler
1705 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1706 Patch from Kirill Miazine
1708 NM/05 Bugzilla 671: Added umask to procmail example.
1710 JJ/03 installed exipick 20100323.0, fixing doc bug
1712 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1713 directory. Notification and patch from Dan Rosenberg.
1715 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1717 TK/02 Improve log output when DKIM signing operation fails.
1719 MH/01 Treat the transport option dkim_domain as a colon separated
1720 list, not as a single string, and sign the message with each element,
1721 omitting multiple occurences of the same signer.
1723 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1724 Bugzilla 985, 986. Patch by Simon Arlott
1726 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1727 Patch by Simon Arlott
1729 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1730 MBX locking. Notification from Dan Rosenberg.
1736 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1738 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1740 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1742 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1744 NM/04 Bugzilla 915: Flag broken perl installation during build.
1750 TK/01 Added patch by Johannes Berg that expands the main option
1751 "spamd_address" if it starts with a dollar sign.
1753 TK/02 Write list of recipients to X-Envelope-Sender header when building
1754 the mbox-format spool file for content scanning (suggested by Jakob
1757 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1758 (http://www.dcc-servers.net/) support via dccifd. Activated by
1759 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1761 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1762 by Mark Daniel Reidel <mr@df.eu>.
1764 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1765 When building exim an external PCRE library is now needed -
1766 PCRE is a system library on the majority of modern systems.
1767 See entry on PCRE_LIBS in EDITME file.
1769 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1770 conversation. Added nologin parameter to request.
1771 Patch contributed by Kirill Miazine.
1773 TF/01 Do not log submission mode rewrites if they do not change the address.
1775 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1777 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1778 log files in place. Contributed by Roberto Lima.
1780 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1782 TF/03 Bugzilla 615: When checking the local_parts router precondition
1783 after a local_part_suffix or local_part_prefix option, Exim now
1784 does not use the address's named list lookup cache, since this
1785 contains cached lookups for the whole local part.
1787 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1788 Robert Millan. Documentation is in experimental-spec.txt.
1790 TF/04 Bugzilla 668: Fix parallel build (make -j).
1792 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1794 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1795 Patch provided by Jan Srzednicki.
1797 TF/05 Leading white space used to be stripped from $spam_report which
1798 wrecked the formatting. Now it is preserved.
1800 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1801 that they are available at delivery time.
1803 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1805 TF/08 TLS error reporting now respects the incoming_interface and
1806 incoming_port log selectors.
1808 TF/09 Produce a more useful error message if an SMTP transport's hosts
1809 setting expands to an empty string.
1811 NM/06 Bugzilla 744: EXPN did not work under TLS.
1812 Patch provided by Phil Pennock.
1814 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1815 Patch provided by Richard Godbee.
1817 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1818 acl_smtp_notquit, added index entry.
1820 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1821 Patch provided by Eugene Bujak.
1823 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1824 accept(). Patch provided by Maxim Dounin.
1826 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1827 Patch provided by Phil Pennock.
1829 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1831 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1832 Patch provided by Brad "anomie" Jorsch.
1834 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1835 Patch provided by Dean Brooks.
1837 TK/05 Add native DKIM support (does not depend on external libraries).
1839 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1840 Patch provided by Graeme Fowler.
1842 NM/16 Bugzilla 851: Documentation example syntax fix.
1844 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1846 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1849 NM/19 Bugzilla 745: TLS version reporting.
1850 Patch provided by Phil Pennock.
1852 NM/20 Bugzilla 167: bool: condition support.
1853 Patch provided by Phil Pennock.
1855 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1856 clients. Patch provided by Phil Pennock.
1858 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1859 Patch provided by Brad "anomie" Jorsch.
1861 NM/23 Bugzilla 687: Fix misparses in eximstats.
1862 Patch provided by Heiko Schlittermann.
1864 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1865 Patch provided by Heiko Schlittermann.
1867 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1868 plus update to original patch.
1870 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1872 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1873 Patch provided by David Brownlee.
1875 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1877 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1879 NM/30 Bugzilla 888: TLS documentation bugfixes.
1881 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1883 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1884 Unlike the original bugzilla I have changed all shell scripts in src tree.
1886 NM/33 Bugzilla 898: Transport filter timeout fix.
1887 Patch by Todd Rinaldo.
1889 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1890 Patch by Serge Demonchaux.
1892 NM/35 Bugzilla 39: Base64 decode bug fixes.
1893 Patch by Jakob Hirsch.
1895 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1897 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1899 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1901 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1907 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1908 ALT-N's libdkim that I have put here:
1909 http://duncanthrax.net/exim-experimental/
1911 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1912 were called 'true' and it seems that C99 defines that as a reserved
1913 keyword to be used with 'bool' variable types. That means you could
1914 not include C99-style headers which use bools without triggering
1915 build errors in sieve.c.
1917 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1918 as mailq or other aliases. Changed the --help handling significantly
1919 to do whats expected. exim_usage() emits usage/help information.
1921 SC/01 Added the -bylocaldomain option to eximstats.
1923 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1925 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1927 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1933 PH/01 Another patch from the Sieve maintainer.
1935 PH/02 When an IPv6 address is converted to a string for single-key lookup
1936 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1937 dots are used instead of colons so that keys in lsearch files need not
1938 contain colons. This was done some time before quoting was made available
1939 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1940 (notated using the quote facility) so as to distinguish them from IPv4
1941 keys. This meant that lookups for IP addresses in host lists did not work
1942 for iplsearch lookups.
1944 This has been fixed by arranging for IPv6 addresses to be expressed with
1945 colons if the lookup type is iplsearch. This is not incompatible, because
1946 previously such lookups could never work.
1948 The situation is now rather anomalous, since one *can* have colons in
1949 ordinary lsearch keys. However, making the change in all cases is
1950 incompatible and would probably break a number of configurations.
1952 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1955 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1956 conversion specification without a maximum field width, thereby enabling
1957 a rogue spamd server to cause a buffer overflow. While nobody in their
1958 right mind would setup Exim to query an untrusted spamd server, an
1959 attacker that gains access to a server running spamd could potentially
1960 exploit this vulnerability to run arbitrary code as the Exim user.
1962 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1963 $primary_hostname instead of what libspf2 thinks the hosts name is.
1965 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1966 a directory entry by the name of the lookup key. Previously, if a
1967 symlink pointed to a non-existing file or a file in a directory that
1968 Exim lacked permissions to read, a lookup for a key matching that
1969 symlink would fail. Now it is enough that a matching directory entry
1970 exists, symlink or not. (Bugzilla 503.)
1972 PH/03 The body_linecount and body_zerocount variables are now exported in the
1975 PH/04 Added the $dnslist_matched variable.
1977 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1978 This means they are set thereafter only if the connection becomes
1981 PH/06 Added the client_condition to authenticators so that some can be skipped
1982 by clients under certain conditions.
1984 PH/07 The error message for a badly-placed control=no_multiline_responses left
1985 "_responses" off the end of the name.
1987 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1989 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1990 (without spaces) instead of just copying the configuration text.
1992 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1994 PH/11 Added $max_received_linelength.
1996 PH/12 Added +ignore_defer and +include_defer to host lists.
1998 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1999 way in which PCRE > 7.0 is built.
2001 PH/14 Implemented queue_only_load_latch.
2003 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2004 MAIL command. The effect was to mangle the value on 64-bit systems.
2006 PH/16 Another patch from the Sieve maintainer.
2008 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2010 PH/18 If a system quota error occurred while trying to create the file for
2011 a maildir delivery, the message "Mailbox is full" was not appended to the
2012 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2013 only to a quota excession during the actual writing of the file.
2015 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2016 characters?) which causes problems in log lines. The DN values are now
2017 passed through string_printing() before being added to log lines.
2019 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2020 and InterBase are left for another time.)
2022 PH/21 Added message_body_newlines option.
2024 PH/22 Guard against possible overflow in moan_check_errorcopy().
2026 PH/23 POSIX allows open() to be a macro; guard against that.
2028 PH/24 If the recipient of an error message contained an @ in the local part
2029 (suitably quoted, of course), incorrect values were put in $domain and
2030 $local_part during the evaluation of errors_copy.
2036 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2037 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2040 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2041 issue a MAIL command.
2043 PH/02 In an ACL statement such as
2045 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2047 if a client was not listed at all, or was listed with a value other than
2048 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2049 the condition was not true (as it should be), so access was not denied.
2050 The bug was that the ! inversion was incorrectly passed on to the second
2051 item. This has been fixed.
2053 PH/03 Added additional dnslists conditions == and =& which are different from
2054 = and & when the dns lookup returns more than one IP address.
2056 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2057 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2059 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2060 FSYNC, which compiles an option called disable_fsync that allows for
2061 bypassing fsync(). The documentation is heavily laced with warnings.
2063 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2065 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2066 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2067 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2068 including adding "make clean"; (3) Added -fPIC when compiling the test
2069 dynamically loaded module, to get rid of a warning.
2071 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2072 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2073 = 0s. The bug is otherwise harmless.
2075 PH/07 There was a bug in the dovecot authenticator such that the value of
2076 $auth1 could be overwritten, and so not correctly preserved, after a
2077 successful authentication. This usually meant that the value preserved by
2078 the server_setid option was incorrect.
2080 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2082 PH/09 Installed PCRE release 7.0.
2084 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2085 run for batched SMTP input. It is now run at the start of every message
2086 in the batch. While fixing this I discovered that the process information
2087 (output by running exiwhat) was not always getting set for -bs and -bS
2088 input. This is fixed, and it now also says "batched" for BSMTP.
2090 PH/11 Added control=no_pipelining.
2092 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2093 patch, slightly modified), and move the expansion of helo_data till after
2094 the connection is made in the smtp transport (so it can use these
2097 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2099 PH/14 Added log_selector = +pid.
2101 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2103 PH/16 Add ${if forany and ${if forall.
2105 PH/17 Added dsn_from option to vary the From: line in DSNs.
2107 PH/18 Flush SMTP output before performing a callout, unless control =
2108 no_callout_flush is set.
2110 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2111 was true (the default) a successful delivery failed to delete the retry
2112 item, thus causing premature timeout of the address. The bug is now
2115 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2117 PH/21 Long custom messages for fakedefer and fakereject are now split up
2118 into multiline responses in the same way that messages for "deny" and
2119 other ACL rejections are.
2121 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2122 with slight modification.
2124 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2125 draft, changing the syntax and factoring some duplicate code".
2127 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2128 for deliveries of the second and subsequent messages over the same SMTP
2131 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2132 ${reduce, with only minor "tidies".
2134 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2136 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2137 expansion side effects.
2139 PH/27 When a message times out after an over-quota error from an Exim-imposed
2140 quota, the bounce message says "mailbox is full". This message was not
2141 being given when it was a system quota that was exceeded. It now should
2144 MH/03 Made $recipients available in local_scan(). local_scan() already has
2145 better access to the recipient list through recipients_list[], but
2146 $recipients can be useful in postmaster-provided expansion strings.
2148 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2149 in the case of a MAIL command with additional options following the
2150 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2151 were accidentally chopped off.
2153 PH/29 SMTP synchronization checks are implemented when a command is read -
2154 there is a check that no more input is waiting when there shouldn't be
2155 any. However, for some commands, a delay in an ACL can mean that it is
2156 some time before the response is written. In this time, more input might
2157 arrive, invalidly. So now there are extra checks after an ACL has run for
2158 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2159 pipelining has not been advertised.
2161 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2163 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2164 returned in the SMTP response when smtp_return_error_details was false.
2165 This has been fixed.
2167 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2168 instead of the C I/O that was originally supplied, because problems were
2169 reported on Solaris.
2171 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2172 Exim which did not show up earlier: it was assuming that a call to
2173 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2174 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2175 was a macro that became an assignment, so it seemed to work. This has
2176 changed to a proper function call with a void return, hence the compile
2177 error. Exim's code has been fixed.
2179 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2182 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2184 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2186 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2187 because it turns out to be too big or there is a timeout) there is no
2188 "Completed" line in the log. When some messages of this type were
2189 selected by exigrep, they were listed as "not completed". Others were
2190 picked up by some special patterns. I have improved the selection
2191 criteria to be more general.
2193 PH/38 The host_find_failed option in the manualroute router can now be set
2194 to "ignore", to completely ignore a host whose IP address cannot be
2195 found. If all hosts are ignored, the behaviour is controlled by the new
2196 host_all_ignored option.
2198 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2199 homing or because of multiple MX records with /mx) generated more than
2200 one IP address, and the following item turned out to be the local host,
2201 all the secondary addresses of the first item were incorrectly removed
2202 from the list, along with the local host and any following hosts (which
2203 is what is supposed to happen).
2205 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2206 whoever called Exim into the -H file. In the case of the daemon it was
2207 behaving confusingly. When first started, it used values for whoever
2208 started the daemon, but after a SIGHUP it used the Exim user (because it
2209 calls itself on a restart). I have changed the code so that it now always
2212 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2213 message are rejected with the same error (e.g. no authentication or bad
2214 sender address), and a DATA command is nevertheless sent (as can happen
2215 with PIPELINING or a stupid MUA), the error message that was given to the
2216 RCPT commands is included in the rejection of the DATA command. This is
2217 intended to be helpful for MUAs that show only the final error to their
2220 PH/42 Another patch from the Sieve maintainer.
2222 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2223 Eximstats - Fixed some broken HTML links and added missing column headers
2225 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2226 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2228 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2230 PH/43 Yet another patch from the Sieve maintainer.
2232 PH/44 I found a way to check for a TCP/IP connection going away before sending
2233 the response to the final '.' that terminates a message, but only in the
2234 case where the client has not sent further data following the '.'
2235 (unfortunately, this is allowed). However, in many cases there won't be
2236 any further data because there won't be any more messages to send. A call
2237 to select() can be used: if it shows that the input is "ready", there is
2238 either input waiting, or the socket has been closed. An attempt to read
2239 the next input character can distinguish the two cases. Previously, Exim
2240 would have sent an OK response which the client would never have see.
2241 This could lead to message repetition. This fix should cure that, at
2242 least in a lot of common cases.
2244 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2245 advertised in response to EHLO.
2251 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2252 fixed by 4.65/MH/01 (is this a record?) are fixed:
2254 (i) An empty string was always treated as zero by the numeric comparison
2255 operators. This behaviour has been restored.
2257 (ii) It is documented that the numeric comparison operators always treat
2258 their arguments as decimal numbers. This was broken in that numbers
2259 starting with 0 were being interpreted as octal.
2261 While fixing these problems I realized that there was another issue that
2262 hadn't been noticed. Values of message_size_limit (both the global option
2263 and the transport option) were treated as octal if they started with 0.
2264 The documentation was vague. These values are now always treated as
2265 decimal, and I will make that clear in the documentation.
2271 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2272 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2275 MH/01 Don't check that the operands of numeric comparison operators are
2276 integers when their expansion is in "skipping" mode (fixes bug
2277 introduced by 4.64-PH/07).
2279 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2280 child addresses, Exim now panics and dies. Previously, because the count
2281 is held in a short int, deliveries were likely to be lost. As such a
2282 large number of recipients for a single message is ridiculous
2283 (performance will be very, very poor), I have chosen to impose a limit
2284 rather than extend the field.
2290 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2291 leftover -K file (the existence of which was triggered by #402).
2292 While we were at it, introduced process PID as part of the -K
2293 filename. This should rule out race conditions when creating
2296 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2297 processing considerably. Previous code took too long for large mails,
2298 triggering a timeout which in turn triggers #401.
2300 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2301 in the DK code in transports.c. sendfile() is not really portable,
2302 hence the _LINUX specificness.
2304 TF/01 In the add_headers option to the mail command in an Exim filter,
2305 there was a bug that Exim would claim a syntax error in any
2306 header after the first one which had an odd number of characters
2309 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2310 callout verification, Exim cached a "reject" for the entire domain. This
2311 is correct for most verifications, but it is not correct for a recipient
2312 verification with use_sender or use_postmaster set, because in that case
2313 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2314 case of MAIL FROM:<> rejection from other early rejections (e.g.
2315 rejection of HELO). When verifying a recipient using a non-null MAIL
2316 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2317 Whatever the result of the callout, the value of the domain cache is
2318 left unchanged (for any other kind of callout, getting as far as trying
2319 RCPT means that the domain itself is ok).
2321 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2324 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2325 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2326 session, the daemon ignores ECONNECT errors and logs others; it now
2327 ignores EPIPE as well.
2329 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2330 (quoted-printable decoding).
2332 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2333 later the small subsequent patch to fix an introduced bug.
2335 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2337 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2339 PH/08 An error is now given if message_size_limit is specified negative.
2341 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2342 to be given (somewhat) arbitrary names.
2344 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2347 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2348 miscellaneous code fixes
2350 PH/10 Added the log_reject_target ACL modifier to specify where to log
2353 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2354 hostname. This is wrong, because it relates to the incoming message (and
2355 probably the interface on which it is arriving) and not to the outgoing
2356 callout (which could be using a different interface). This has been
2357 changed to use the value of the helo_data option from the smtp transport
2358 instead - this is what is used when a message is actually being sent. If
2359 there is no remote transport (possible with a router that sets up host
2360 addresses), $smtp_active_hostname is used.
2362 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2363 tweaks were necessary in order to get it to work (see also 21 below):
2364 (a) The code assumed that strncpy() returns a negative number on buffer
2365 overflow, which isn't the case. Replaced with Exim's string_format()
2367 (b) There were several signed/unsigned issues. I just did the minimum
2368 hacking in of casts. There is scope for a larger refactoring.
2369 (c) The code used strcasecmp() which is not a standard C function.
2370 Replaced with Exim's strcmpic() function.
2371 (d) The code set only $1; it now sets $auth1 as well.
2372 (e) A simple test gave the error "authentication client didn't specify
2373 service in request". It would seem that Dovecot has changed its
2374 interface. Fortunately there's a specification; I followed it and
2375 changed what the client sends and it appears to be working now.
2377 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2380 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2381 address A is aliased to B and C, where B exists and C does not. Without
2382 -v the output is "A verified" because verification stops after a
2383 successful redirection if more than one address is generated. However,
2384 with -v the child addresses are also verified. Exim was outputting "A
2385 failed to verify" and then showing the successful verification for C,
2386 with its parentage. It now outputs "B failed to verify", showing B's
2387 parentage before showing the successful verification of C.
2389 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2390 look up a TXT record in a specific list after matching in a combined
2393 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2394 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2395 they consult the DNS. I had assumed they would set it the way they
2396 wanted; and indeed my experiments on Linux seem to show that in some
2397 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2398 To be on the safe side, however, I have now made the interface to
2399 host_find_byname() similar to host_find_bydns(), with an argument
2400 containing the DNS resolver options. The host_find_byname() function now
2401 sets these options at its start, just as host_find_bydns() does. The smtp
2402 transport options dns_qualify_single and dns_search_parents are passed to
2403 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2404 of host_find_byname() use the default settings of RES_DEFNAMES
2405 (qualify_single) but not RES_DNSRCH (search_parents).
2407 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2408 spool_read_header() do less string testing, by means of a preliminary
2409 switch on the second character of optional "-foo" lines. (This is
2410 overdue, caused by the large number of possibilities that now exist.
2411 Originally there were few.) While I was there, I also converted the
2412 str(n)cmp tests so they don't re-test the leading "-" and the first
2413 character, in the hope this might squeeze out yet more improvement.
2415 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2416 flag allowing group syntax was set by the header_syntax check but not
2417 turned off, possible causing trouble later; (2) The flag was not being
2418 set at all for the header_verify test, causing "group"-style headers to
2419 be rejected. I have now set it in this case, and also caused header_
2420 verify to ignore an empty address taken from a group. While doing this, I
2421 came across some other cases where the code for allowing group syntax
2422 while scanning a header line wasn't quite right (mostly, not resetting
2423 the flag correctly in the right place). These bugs could have caused
2424 trouble for malformed header lines. I hope it is now all correct.
2426 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2427 with the "reply" argument non-NULL. The code, however (which originally
2428 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2429 but it didn't always do it. This confused somebody who was copying the
2430 code for some other use. I have removed all the tests.
2432 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2433 feature that was used to support insecure browsers during the U.S. crypto
2434 embargo. It requires special client support, and Exim is probably the
2435 only MTA that supported it -- and would never use it because real RSA is
2436 always available. This code has been removed, because it had the bad
2437 effect of slowing Exim down by computing (never used) parameters for the
2438 RSA_EXPORT functionality.
2440 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2441 authenticator to fail if there's a tab character in the incoming data
2442 (there should never be unless someone is messing about, as it's supposed
2443 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2444 if the connection is using TLS or if the remote IP is the same as the
2445 local IP, and the "valid-client-cert option" if a client certificate has
2448 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2449 authenticators. This can be used for authorization after authentication
2450 succeeds. (In the case of plaintext, it servers for both authentication
2453 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2454 if any retry times were supplied.
2456 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2457 connection, where there is no client IP address to check. In this
2458 situation, the verify now always succeeds.
2460 PH/25 Applied John Jetmore's -Mset patch.
2462 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2464 PH/27 In a string expansion for a processed (not raw) header when multiple
2465 headers of the same name were present, leading whitespace was being
2466 removed from all of them, but trailing whitespace was being removed only
2467 from the last one. Now trailing whitespace is removed from each header
2468 before concatenation. Completely empty headers in a concatenation (as
2469 before) are ignored.
2471 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2472 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2474 PH/29 [Removed. This was a change that I later backed out, and forgot to
2475 correct the ChangeLog entry (that I had efficiently created) before
2476 committing the later change.]
2478 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2479 address errors (4xx response to RCPT) over the same connection as other
2480 messages routed to the same hosts. Such deliveries are always "forced",
2481 so retry times are not inspected. This resulted in far too many retries
2482 for the affected addresses. The effect occurred only when there were more
2483 hosts than the hosts_max_try setting in the smtp transport when it had
2484 the 4xx errors. Those hosts that it had tried were not added to the list
2485 of hosts for which the message was waiting, so if all were tried, there
2486 was no problem. Two fixes have been applied:
2488 (i) If there are any address or message errors in an SMTP delivery, none
2489 of the hosts (tried or untried) are now added to the list of hosts
2490 for which the message is waiting, so the message should not be a
2491 candidate for sending over the same connection that was used for a
2492 successful delivery of some other message. This seems entirely
2493 reasonable: after all the message is NOT "waiting for some host".
2494 This is so "obvious" that I'm not sure why it wasn't done
2495 previously. Hope I haven't missed anything, but it can't do any
2496 harm, as the worst effect is to miss an optimization.
2498 (ii) If, despite (i), such a delivery is accidentally attempted, the
2499 routing retry time is respected, so at least it doesn't keep
2500 hammering the server.
2502 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2503 in ${readsocket because some servers need this prod.
2505 PH/32 Added some extra debug output when updating a wait-xxx database.
2507 PH/33 The hint "could be header name not terminated by colon", which has been
2508 given for certain expansion errors for a long time, was not being given
2509 for the ${if def:h_colon_omitted{... case.
2511 PH/34 The spec says: "With one important exception, whenever a domain list is
2512 being scanned, $domain contains the subject domain." There was at least
2513 one case where this was not true.
2515 PH/35 The error "getsockname() failed: connection reset by peer" was being
2516 written to the panic log as well as the main log, but it isn't really
2517 panic-worthy as it just means the connection died rather early on. I have
2518 removed the panic log writing for the ECONNRESET error when getsockname()
2521 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2522 runs only) independently of the message's sender address. This meant
2523 that, if the 4xx error was in fact related to the sender, a different
2524 message to the same recipient with a different sender could confuse
2525 things. In particular, this can happen when sending to a greylisting
2526 server, but other circumstances could also provoke similar problems.
2527 I have changed the default so that the retry time for these errors is now
2528 based a combination of the sender and recipient addresses. This change
2529 can be overridden by setting address_retry_include_sender=false in the
2532 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2533 remote server are returned as part of bounce messages. This was not
2534 happening for LMTP over a pipe (the lmtp transport), but now it is the
2535 same for both kinds of LMTP.
2537 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2538 in header lines that were in fact CNAMEs. This is no longer the case.
2540 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2541 and queue runs started by the daemon processed all messages. This has
2542 been fixed so that -R and -S can now usefully be given with -q<time>.
2544 PH/40 Import PCRE release 6.7 (fixes some bugs).
2546 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2548 PH/42 Give an error if -q is specified more than once.
2550 PH/43 Renamed the variables $interface_address and $interface_port as
2551 $received_ip_address and $received_port, to make it clear that these
2552 values apply to message reception, and not to the outgoing interface when
2553 a message is delivered. (The old names remain recognized, of course.)
2555 PH/44 There was no timeout on the connect() call when using a Unix domain
2556 socket in the ${readsocket expansion. There now is.
2558 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2559 be meaningful with "accept".
2561 SC/01 Eximstats V1.43
2562 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2564 SC/02 Eximstats V1.44
2565 Use a glob alias rather than an array ref in the generated
2566 parser. This improves both readability and performance.
2568 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2569 Collect SpamAssassin and rejection statistics.
2570 Don't display local sender or destination tables unless
2571 there is data to show.
2572 Added average volumes into the top table text output.
2574 SC/04 Eximstats V1.46
2575 Collect data on the number of addresses (recipients)
2576 as well as the number of messages.
2578 SC/05 Eximstats V1.47
2579 Added 'Message too big' to the list of mail rejection
2580 reasons (thanks to Marco Gaiarin).
2582 SC/06 Eximstats V1.48
2583 Mainlog lines which have GMT offsets and are too short to
2584 have a flag are now skipped.
2586 SC/07 Eximstats V1.49 (Alain Williams)
2587 Added the -emptyok flag.
2589 SC/08 Eximstats V1.50
2590 Fixes for obtaining the IP address from reject messages.
2592 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2593 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2594 whitespace changes from 4.64-PH/27
2596 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2599 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2600 are found, allow negative numbers in numeric criteria)
2602 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2604 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2605 to match changes made in 4.64-PH/43
2607 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2609 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2610 "domain" is actually an IP address, and force a failure. This locks out
2611 those revolvers/nameservers that support "A-for-A" lookups, in
2612 contravention of the specifications.
2614 PH/48 When a host name was looked up from an IP address, and the subsequent
2615 forward lookup of the name timed out, the host name was left in
2616 $sender_host_name, contrary to the specification.
2618 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2619 restricted to single-key lookups, Exim was not diagnosing an error if
2620 * or *@ was used with a query-style lookup.
2622 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2624 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2625 long ago, but noone interested enough thought of it. Let's just say that
2626 the "1.1" means that there are some new functions that weren't there at
2627 some point in the past.
2629 PH/51 Error processing for expansion failure of helo_data from an smtp
2630 transport during callout processing was broken.
2632 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2633 tested/used via the -bh/-bhc/-bs options.
2635 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2636 bug, fixed in subsequent PCRE releases).
2638 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2639 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2641 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2647 SC/01 Use a glob alias rather than an array ref in eximstats generated
2648 parser. This improves both readability and performance.
2650 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2651 Don't display local sender or destination tables in eximstats unless
2652 there is data to show.
2653 Added average volumes into the eximstats top table text output.
2655 SC/03 Collect data on the number of addresses (recipients) as well
2656 as the number of messages in eximstats.
2658 TF/01 Correct an error in the documentation for the redirect router. Exim
2659 does (usually) call initgroups() when daemonizing.
2661 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2662 with consistent privilege compared to when running as a daemon.
2664 TF/03 Note in the spec that $authenticated_id is not set for local
2665 submissions from trusted users.
2667 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2668 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2670 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2671 by adding some example configuration directives to the default
2672 configuration file. A little bit of work is required to uncomment the
2673 directives and define how usernames and passwords are checked, but
2674 there is now a framework to start from.
2676 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2677 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2678 without this. I don't know how relevant this is to other LDAP libraries.
2680 PH/02 Add the verb name to the "unknown ACL verb" error.
2682 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2684 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2686 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2687 directory not expanded when it should be if an expanded home directory
2688 was set for the address (which is overridden by the transport).
2690 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2693 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2694 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2695 because it is too late at that time, and has no effect.
2697 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2698 security issue with \' (bugzilla #107). I could not use the
2699 PQescapeStringConn() function, because it needs a PGconn value as one of
2702 PH/08 When testing addresses using -bt, indicate those final addresses that
2703 are duplicates that would not cause an additional delivery. At least one
2704 person was confused, thinking that -bt output corresponded to deliveries.
2705 (Suppressing duplicates isn't a good idea as you lose the information
2706 about possibly different redirections that led to the duplicates.)
2708 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2709 systems where poll() doesn't work, in particular OS X.
2711 PH/10 Added more information to debugging output for retry time not reached.
2713 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2714 operations in malware.c.
2716 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2719 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2720 syslog_duplication set false, log lines that would normally be written
2721 both the the main log and to the reject log were not written to syslog at
2724 PH/14 In the default configuration, change the use of "message" in ACL warn
2725 statements to "add_header".
2727 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2728 not followed by a command (e.g. "seen endif").
2730 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2731 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2734 PH/17 Added extra conditions to the default value of delay_warning_condition
2738 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2739 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2740 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2743 The Auto-Submitted: and various List- headers are standardised, whereas I
2744 don't think Precedence: ever was.
2746 PH/18 Refactored debugging code in route_finduser() to show more information,
2747 in particular, the error code if getpwnam() issues one.
2749 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2750 This is apparently needed in addition to the PH/07 change above to avoid
2751 any possible encoding problems.
2753 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2754 but not after initializing Perl.
2756 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2757 output them only if debugging. By default they are written stderr,
2758 apparently, which is not desirable.
2760 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2763 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2766 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2768 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2769 authenticated or an ident call has been made. Suppress the default
2770 values for $authenticated_id and $authenticated_sender (but permit -oMai
2771 and -oMas) when testing with -bh.
2773 PH/24 Re-jigged the order of the tests in the default configuration so that the
2774 tests for valid domains and recipients precede the DNS black list and CSA
2775 tests, on the grounds that those ones are more expensive.
2777 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2778 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2779 This bug exists in every version of Exim that I still have, right back to
2782 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2783 However, an attempt to turn on case-sensitivity in a regex key by
2784 including (?-i) didn't work because the subject string was already
2785 lowercased, and the effects were non-intuitive. It turns out that a
2786 one-line patch can be used to allow (?-i) to work as expected.
2792 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2793 other effects) broke the use of negated acl sub-conditions.
2795 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2798 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2799 "Deny" causes Exim to reject the incoming connection with a 554 error.
2800 Unfortunately, if there is a major crisis, such as a disk failure,
2801 tcp-wrappers gives "deny", whereas what one would like would be some
2802 kind of temporary error. A kludge has been added to help with this.
2803 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2804 554 error is used if errno is still zero or contains ENOENT (which occurs
2805 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2808 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2810 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2811 errors. Otherwise a message that provokes a temporary error (when other
2812 messages do not) can cause a whole host to time out.
2814 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2815 addresses were routed directly to files or pipes from a redirect router.
2816 File deliveries just didn't batch; pipe deliveries might have suffered
2819 PH/06 A failure to get a lock for a hints database would erroneously always say
2820 "Failed to get write lock", even when it was really a read lock.
2822 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2823 of 0600. This has been changed to use the value of the lockfile_mode
2824 option (which defaults to 0600).
2826 PH/08 Applied small patch from the Sieve maintainer.
2828 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2829 folder from quota calculations, a direct delivery into this folder messed
2830 up the contents of the maildirsize file. This was because the regex was
2831 used only to exclude .Trash (or whatever) when the size of the mailbox
2832 was calculated. There was no check that a delivery was happening into an
2833 excluded directory. This bug has been fixed by ignoring all quota
2834 processing for deliveries into excluded directories.
2836 PH/10 Added the maildirfolder_create_regex option to appendfile.
2842 PH/01 The code for finding all the local interface addresses on a FreeBSD
2843 system running IPv6 was broken. This may well have applied to all BSD
2844 systems, as well as to others that have similar system calls. The broken
2845 code found IPv4 interfaces correctly, but gave incorrect values for the
2846 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2847 that it would not match correctly against @[] and not recognize the IPv6
2850 PH/02 The ipliteral router was not recognizing addresses of the form user@
2851 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2853 PH/03 Added disable_ipv6.
2855 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2856 decoded form, because it is most often used to construct To: headers
2857 lines in autoreplies, and the decoded form may well be syntactically
2858 invalid. However, $reply_address has leading white space removed, and all
2859 newlines turned into spaces so that the autoreply transport does not
2862 PH/05 If group was specified without a user on a router, and no group or user
2863 was specified on a transport, the group from the router was ignored.
2865 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2866 for visible compile-time settings that can be used to change these
2867 numbers, for those that want even more. Backwards compatibility with old
2868 spool files has been maintained. However, going back to a previous Exim
2869 release will lost any variables that are in spool files.
2871 PH/07 Two small changes when running in the test harness: increase delay when
2872 passing a TCP/IP connection to a new process, in case the original
2873 process has to generate a bounce, and remove special handling of
2874 127.0.0.2 (sic), which is no longer necessary.
2876 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2877 be the same on different OS.
2879 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2882 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2883 whether --show-vars was specified or not
2885 JJ/02 exipick: Added support for new ACL variable spool format introduced
2888 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2889 syntactically invalid From: or Reply-to: line, and a filter used this to
2890 generate an autoreply, and therefore failed to obtain an address for the
2891 autoreply, Exim could try to deliver to a non-existent relative file
2892 name, causing unrelated and misleading errors. What now happens is that
2893 it logs this as a hard delivery error, but does not attempt to create a
2896 PH/11 The exinext utility has a -C option for testing purposes, but although
2897 the given file was scanned by exinext itself; it wasn't being passed on
2898 when Exim was called.
2900 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2901 an end-of-file indication when reading a command response.
2903 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2904 compiled. In many other places in Exim, IPv6 addresses are always
2905 recognized, so I have changed this. It also means that IPv4 domain
2906 literals of the form [IPV4:n.n.n.n] are now always recognized.
2908 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2909 used if the router is not running as root, for example, when verifying at
2910 ACL time, or when using -bh. The debugging output from this situation was
2911 non-existent - all you got was a failure to exec. I have made two
2914 (a) Failures to set uid/gid, the current directory, or a process leader
2915 in a subprocess such as that created by queryprogram now generate
2916 suitable debugging output when -d is set.
2918 (b) The queryprogram router detects when it is not running as root,
2919 outputs suitable debugging information if -d is set, and then runs
2920 the subprocess without attempting to change uid/gid.
2922 PH/15 Minor change to Makefile for building test_host (undocumented testing
2925 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2926 additional section of a DNS packet that returns MX or SRV records.
2927 Instead, it always explicitly searches for A/AAAA records. This avoids
2928 major problems that occur when a DNS server includes only records of one
2929 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2930 fixed another bug: if SRV records were looked up and the corresponding
2931 address records were *not* found in the additional section, the port
2932 values from the SRV records were lost.
2934 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2935 using the correct key (the original address) when searching the retry
2936 rules in order to find which one to use for generating the retry hint.
2938 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2939 adding the default one. Similarly, if it contains a Reply-To: header, the
2940 errors_reply_to option, if set, is not used.
2942 PH/19 When calculating a retry time, Exim used to measure the "time since
2943 failure" by looking at the "first failed" field in the retry record. Now
2944 it does not use this if it is later than than the arrival time of the
2945 message. Instead it uses the arrival time. This makes for better
2946 behaviour in cases where some deliveries succeed, thus re-setting the
2947 "first failed" field. An example is a quota failure for a huge message
2948 when small messages continue to be delivered. Without this change, the
2949 "time since failure" will always be short, possible causing more frequent
2950 delivery attempts for the huge message than are intended.
2951 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2953 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2954 $1, $2, $3) because the numerical variables can be reset during some
2955 expansion items (e.g. "match"), thereby losing the authentication data.
2957 PH/21 Make -bV show the size of off_t variables so that the test suite can
2958 decide whether to run tests for quotas > 2G.
2960 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2961 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2962 filecount value is greater than 2G or if a quota value is greater than 2G
2963 on a system where the size of off_t is not greater than 4, a panic error
2966 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2967 never match. The debug and -bh output now contains an explicit error
2968 message indicating a malformed IPv4 address or mask.
2970 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2971 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2972 PH/23 above applies.
2974 PH/25 Do not write to syslog when running in the test harness. The only
2975 occasion when this arises is a failure to open the main or panic logs
2976 (for which there is an explicit test).
2978 PH/26 Added the /no_tell option to "control=freeze".
2980 PH/27 If a host name lookup failed very early in a connection, for example, if
2981 the IP address matched host_lookup and the reverse lookup yielded a name
2982 that did not have a forward lookup, an error message of the form "no IP
2983 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2984 could be logged. Now it outputs the IP address instead of "NULL".
2986 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2987 allows the sender and the authenticated sender to be set when
2988 submitting a message from within Exim. Since child_open_exim() is
2989 documented for local_scan(), the new function should be too.
2991 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2992 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2993 results in an empty string is now treated as unset.
2995 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2997 PH/31 Added sender_verify_fail logging option.
2999 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3000 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3001 tidied the source and removed it altogether.
3003 PH/33 When a queue run was abandoned because the load average was too high, a
3004 log line was always written; now it is written only if the queue_run log
3005 selector is set. In addition, the log line for abandonment now contains
3006 information about the queue run such as the pid. This is always present
3007 in "start" and "stop" lines but was omitted from the "abandon" line.
3009 PH/34 Omit spaces between a header name and the colon in the error message that
3010 is given when verify = headers_syntax fails (if there are lots of them,
3011 the message gets confusing).
3013 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3014 names, as there are now some PTR records that contain slashes. This check
3015 is only to protect against broken name servers that fall over on strange
3016 characters, so the fact that it applies to all lookups doesn't matter.
3018 PH/36 Now that the new test suite is complete, we can remove some of the
3019 special code in Exim that was needed for the old test suite. For example,
3020 sorting DNS records because real resolvers return them in an arbitrary
3021 order. The new test suite's fake resolver always returns records in the
3024 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3025 bounces) except when queue_only is set, to avoid logging races between
3026 the different processes.
3028 PH/38 Panic-die if .include specifies a non-absolute path.
3030 PH/39 A tweak to the "H" retry rule from its user.
3032 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3033 a label. They prevented compilation on older perls.
3035 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3036 a warning to be raised on newish perls.
3038 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3039 on queue. Changes to match documented behaviour of showing count of
3040 messages matching specified criteria.
3042 PH/40 Changed the default ident timeout from 30s to 5s.
3044 PH/41 Added support for the use of login_cap features, on those BSD systems
3045 that have them, for controlling the resources used by pipe deliveries.
3047 PH/42 The content-scanning code uses fopen() to create files in which to put
3048 message data. Previously it was not paying any attention to the mode of
3049 the files. Exim runs with umask(0) because the rest of the code creates
3050 files with open(), and sets the required mode explicitly. Thus, these
3051 files were ending up world-writeable. This was not a big issue, because,
3052 being within the spool directory, they were not world-accessible. I have
3053 created a function called modefopen, which takes an additional mode
3054 argument. It sets umask(777), creates the file, chmods it to the required
3055 mode, then resets the umask. All the relevant calls to fopen() in the
3056 content scanning code have been changed to use this function.
3058 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3059 to 24 hours. This avoids potential overflow problems when processing G
3060 and H retry rules. I suspect nobody ever tinkers with this value.
3062 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3064 PH/45 When the plaintext authenticator is running as a client, the server's
3065 challenges are checked to ensure they are valid base64 strings. By
3066 default, the authentication attempt is cancelled if an invalid string is
3067 received. Setting client_ignore_invalid_base64 true ignores these errors.
3068 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3069 they are received. Thus, the responses can be made to depend on the
3070 challenges. If an invalid string is ignored, an empty string is placed in
3073 PH/46 Messages that are created by the autoreply transport now contains a
3074 References: header, in accordance with RFCs 2822 and 3834.
3076 PH/47 Added authenticated_sender_force to the smtp transport.
3078 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3080 PH/49 Installed latest patch from the Sieve maintainer.
3082 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3083 was also set, the appendfile transport was unnecessarily scanning a
3084 directory of message files (e.g. for maildir delivery) to find the count
3085 of files (along with the size), even though it did not need this
3086 information. It now does the scan only if it needs to find either the
3087 size of the count of files.
3089 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3091 PH/52 Two bugs concerned with error handling when the smtp transport is
3094 (i) Exim was not creating retry information for temporary errors given
3095 for individual recipients after the DATA command when the smtp transport
3096 was used in LMTP mode. This meant that they could be retried too
3097 frequently, and not timed out correctly.
3099 (ii) Exim was setting the flag that allows error details to be returned
3100 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3101 recipients that were returned after the DATA command.
3103 PH/53 This is related to PH/52, but is more general: for any failing address,
3104 when detailed error information was permitted to be returned to the
3105 sender, but the error was temporary, then after the final timeout, only
3106 "retry timeout exceeded" was returned. Now it returns the full error as
3107 well as "retry timeout exceeded".
3109 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3110 do this, and (what is worse) MTAs that accept it.
3112 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3113 will now be deprecated.
3115 PH/56 New os.c-cygwin from the Cygwin maintainer.
3117 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3118 formats (previously only available in exim formats via -bpr, -bpru,
3119 and -bpra. Now also available in native and exiqgrep formats)
3121 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3122 with very large, slow to parse queues
3124 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3126 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3128 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3129 responses to authentication challenges, though it was showing the
3130 challenges; (ii) I've removed the CR characters from the debug output for
3133 PH/58 Allow for the insertion of a newline as well as a space when a string
3134 is turned into more than one encoded-word during RFC 2047 encoding. The
3135 Sieve code now uses this.
3137 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3138 data_4xx, lost_connection, tls_required.
3140 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3141 message was being sent as an SMTP response.
3143 PH/61 Add -l and -k options to exicyclog.
3145 PH/62 When verifying, if an address was redirected to one new address, so that
3146 verification continued, and the new address failed or deferred after
3147 having set something in $address_data, the value of $address_data was not
3148 passed back to the ACL. This was different to the case when no
3149 redirection occurred. The value is now passed back in both cases.
3151 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3152 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3153 use login_cap.h, so on its own it isn't the distinguishing feature. The
3154 new name refers directly to the setclassresources() function.
3156 PH/65 Added configuration files for NetBSD3.
3158 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3160 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3161 is preferred over IPv4.
3163 PH/68 The bounce_return_message and bounce_return_body options were not being
3164 honoured for bounces generated during the reception of non-SMTP messages.
3165 In particular, this applied to messages rejected by the ACL. This bug has
3166 been fixed. However, if bounce_return_message is true and bounce_return_
3167 body is false, the headers that are returned for a non-SMTP message
3168 include only those that have been read before the error was detected.
3169 (In the case of an ACL rejection, they have all been read.)
3171 PH/69 The HTML version of the specification is now built in a directory called
3172 spec_html instead of spec.html, because the latter looks like a path with
3173 a MIME-type, and this confuses some software.
3175 PH/70 Catch two compiler warnings in sieve.c.
3177 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3178 function verify_get_ident() calls ip_connect() to connect a socket, but
3179 if the "connect()" function timed out, ip_connect() used to close the
3180 socket. However, verify_get_ident() also closes the socket later, and in
3181 between Exim writes to the log, which may get opened at this point. When
3182 the socket was closed in ip_connect(), the log could get the same file
3183 descriptor number as the socket. This naturally causes chaos. The fix is
3184 not to close the socket in ip_connect(); the socket should be closed by
3185 the function that creates it. There was only one place in the code where
3186 this was missing, in the iplookup router, which I don't think anybody now
3187 uses, but I've fixed it anyway.
3189 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3190 well as to direct DNS lookups. Otherwise the handling of names in host
3191 lists is inconsistent and therefore confusing.
3197 PH/01 Two changes to the default runtime configuration:
3199 (1) Move the checks for relay_from_hosts and authenticated clients from
3200 after to before the (commented out) DNS black list checks.
3202 (2) Add control=submission to the relay_from_hosts and authenticated
3203 clients checks, on the grounds that messages accepted by these
3204 statements are most likely to be submissions.
3206 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3208 (1) Generate an error if the third argument for the ${prvs expansion is
3211 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3214 (3) Reset the variables that are obtained from the first argument of
3215 ${prvscheck and used in the second argument before leaving the code,
3216 because their memory is reclaimed, so using them afterwards may do
3219 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3220 one (it's much easier than Tom thought :-).
3222 (5) Because of (4), we can now allow for the use of $prvscheck_result
3223 inside the third argument.
3225 PH/03 For some reason, the default setting of PATH when running a command from
3226 a pipe transport was just "/usr/bin". I have changed it to
3229 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3230 anything to be listed in the output from -bV.
3232 PH/05 When a filter generated an autoreply, the entire To: header line was
3233 quoted in the delivery log line, like this:
3235 => >A.N.Other <ano@some.domain> <original@ddress> ...
3237 This has been changed so that it extracts the operative address. There
3238 may be more than one such address. If so, they are comma-separated, like
3241 => >ano@some.domain,ona@other.domain <original@ddress> ...
3243 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3244 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3245 not being looked up in the rDNS to get a host name, Exim was showing the
3246 IP address twice in Received: lines, even though the IP addresses were
3247 identical. For example:
3249 Received: from [1.2.3.4] (helo=[1.2.3.4])
3251 However, if the real host name was known, it was omitting the HELO data
3252 if it matched the actual IP address. This has been tidied up so that it
3253 doesn't show the same IP address twice.
3255 PH/07 When both +timestamp and +memory debugging was on, the value given by
3256 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3257 called by the debug printing, thereby overwriting the timestamp buffer.
3258 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3260 PH/08 When the original message was included in an autoreply transport, it
3261 always said "this is a copy of the message, including all the headers",
3262 even if body_only or headers_only was set. It now gives an appropriate
3265 PH/09 Applied a patch from the Sieve maintainer which:
3267 o fixes some comments
3268 o adds the (disabled) notify extension core
3269 o adds some debug output for the result of if/elsif tests
3270 o points to the current vacation draft in the documentation
3271 and documents the missing references header update
3275 o fixes a bug in processing the envelope test (when testing
3276 multiple envelope elements, the last element determined the
3279 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3280 Electronic Mail") by including:
3282 Auto-submitted: auto-generated
3284 in the messages that it generates (bounce messages and others, such as
3285 warnings). In the case of bounce messages for non-SMTP messages, there was
3286 also a typo: it was using "Auto_submitted" (underscore instead of
3287 hyphen). Since every message generated by Exim is necessarily in response
3288 to another message, thes have all been changed to:
3290 Auto-Submitted: auto-replied
3292 in accordance with these statements in the RFC:
3294 The auto-replied keyword:
3296 - SHOULD be used on messages sent in direct response to another
3297 message by an automatic process,
3299 - MUST NOT be used on manually-generated messages,
3301 - MAY be used on Delivery Status Notifications (DSNs) and Message
3302 Disposition Notifications (MDNs),
3304 - MUST NOT be used on messages generated by automatic or periodic
3305 processes, except for messages which are automatic responses to
3308 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3309 to the default Received: header definition.
3311 PH/12 Added log selector acl_warn_skipped (default on).
3313 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3314 variables because (a) they are in the wrong storage pool and (b) even if
3315 they were copied, it wouldn't work properly because of the caching.
3317 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3318 checking when decoding. Apparently there are clients that generate
3319 overlong encoded strings. Why am I not surprised?
3321 PH/15 If the first argument of "${if match_address" was not empty, but did not
3322 contain an "@" character, Exim crashed. Now it writes a panic log message
3323 and treats the condition as false.
3325 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3327 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3328 extension "envelope-auth". The code is finished and in agreement with
3329 other implementations, but there is no documentation so far and in fact,
3330 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3331 not changing the active code.
3333 Print executed "if" and "elsif" statements when debugging is used. This
3334 helps a great deal to understand what a filter does.
3336 Document more things not specified clearly in RFC3028. I had all this
3337 sorted out, when out of a sudden new issues came to my mind. Oops."
3339 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3342 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3343 canonical form (as documented). However, after a host name lookup from
3344 the IP address, check_host() was doing a simple string comparison with
3345 addresses acquired from the DNS when checking that the found name did
3346 have the original IP as one of its addresses. Since any found IPv6
3347 addresses are likely to be in abbreviated form, the comparison could
3348 fail. Luckily, there already exists a function for doing the comparison
3349 by converting both addresses to binary, so now that is used instead of
3350 the text comparison.
3352 PH/20 There was another similar case to PH/19, when a complete host name was
3353 given in a host list; looking up its IP address could give an abbreviated
3354 form, whereas the current host's name might or might not be abbreviated.
3355 The same fix has been applied.
3361 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3362 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3365 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3366 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3368 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3370 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3371 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3372 Also, for "Auto-Submitted", treat anything other than "no" as
3373 non-personal, in accordance with RFC 3834. (Previously it treated
3374 anything starting "auto-" as non-personal.)
3376 TF/01 The control=submission/name=... option had a problem with syntax
3377 errors if the name included a slash character. The /name= option
3378 now slurps the rest of the string, so it can include any characters
3379 but it must come last in the list of options (after /sender_retain
3382 PH/05 Some modifications to the interface to the fake nameserver for the new
3390 TK/01 Added the "success_on_redirect" address verification option. See
3391 NewStuff for rationale and an example.
3393 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3395 PH/02 Patch to exigrep to allow it to work on syslog lines.
3397 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3398 fread() to skip over the body file's header line, because in Cygwin the
3399 header line is locked and is inaccessible.
3401 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3402 co-exist for some time) to make it clear that it is the Exim ID that is
3403 referenced, not the Message-ID: header line.
3405 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3406 string_format() function, because snprintf() does not exist on all
3409 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3410 ${stat: expansion item.
3412 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3413 protocol synchronization error", to keep the pedants happy.
3415 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3416 well as for IRIX systems, when gcc is being used. See the host.c source
3419 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3421 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3424 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3425 transport and to the smtp transport in LMTP mode.
3427 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3429 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3430 run a filter in a subprocess. This could lead to confusion in subsequent
3431 lookups in the parent process. There should also be a search_tidyup() at
3432 the end of the subprocess.
3434 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3435 only if the host matched helo_try_verify_hosts, which caused the
3436 verification to occur when the EHLO/HELO command was issued. The ACL just
3437 tested the remembered result. Now, if a previous verification attempt has
3438 not happened, "verify = helo" does it there and then.
3440 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3442 TK/03 Fix log output including CR from clamd.
3444 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3445 exist provoked a memory error which could cause a segfault.
3447 PH/15 Installed PCRE 6.2
3449 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3451 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3452 of the problem. Specifically, suggested +O2 rather than +O1 for the
3455 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3457 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3458 record, the port was not correctly logged when the outgoing_port log
3459 selector was set (it logged the transort's default port).
3461 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3462 fallback_hosts, and "hosts" in the smtp transport.
3464 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3465 host errors such as "Connection refused".
3467 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3468 authentication with radiusclient 0.4.9:
3470 - Error returned from rc_read_config was caught wrongly
3471 - Username/password not passed on to radius server due to wrong length.
3473 The presumption is that some radiusclient API changes for 4.51/PH/17
3474 were not taken care of correctly. The code is still untested by me (my
3475 Linux distribution still has 0.3.2 of radiusclient), but it was
3476 contributed by a Radius user.
3478 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3479 expanding the "port" option of the smtp transport.
3481 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3482 while reading a MIME header. Thanks to Tom Hughes for a patch.
3484 PH/24 Include config.h inside local_scan.h so that configuration settings are
3487 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3488 that in an ACL for RCPT (for example), you can examine exactly what was
3491 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3492 commands, but it was not correctly comparing the address with the actual
3493 client host address. Thus, it would show the EHLO address in Received:
3494 header lines when this was not necessary.
3496 PH/27 Added the % operator to ${eval:}.
3498 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3499 it should be ignoring failures (because with -C, for example, it has lost
3500 privilege). It wasn't ignoring creation failures other than "already
3503 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3506 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3507 because an input line was too long, either on its own, or by virtue of
3508 too many continuations, the temporary file was not being removed, and the
3509 return code was incorrect.
3511 PH/31 Missing "BOOL" in function definition in filtertest.c.
3513 PH/32 Applied Sieve patches from the maintainer.
3515 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3517 PH/33 Added "verify = not_blind".
3519 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3520 Local/Makefile (with some defaults set). These are used in built scripts
3521 such as exicyclog, but they have never been used in the exim_install
3522 script (though there are many overriding facilities there). I have
3523 arranged that the exim_install script now takes note of these two
3526 PH/35 Installed configuration files for Dragonfly.
3528 PH/36 When a locally submitted message by a trusted user did not contain a
3529 From: header, and the sender address was obtained from -f or from an SMTP
3530 MAIL command, and the trusted user did not use -F to supply a sender
3531 name, $originator_name was incorrectly used when constructing a From:
3532 header. Furthermore, $originator_name was used for submission mode
3533 messages from external hosts without From: headers in a similar way,
3534 which is clearly wrong.
3536 PH/37 Added control=suppress_local_fixups.
3538 PH/38 When log_selector = +received_sender was set, and the addition of the
3539 sender made the log line's construction buffer exactly full, or one byte
3540 less than full, an overflow happened when the terminating "\n" was
3543 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3544 when the result of a list match is failure because a DNS lookup failed.
3546 PH/40 RM_COMMAND is now used in the building process.
3548 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3549 the "build-* directories that it finds.
3551 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3552 address in a domain literal was a prefix of an interface address.
3554 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3555 when verifying a sender address, unless rewrite_headers is false.
3557 PH/44 Wrote a long comment about why errors_to addresses are verified as
3558 recipients, not senders.
3560 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3561 the ratelimit ACL was added.
3563 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3565 PH/46 Added extra information about PostgreSQL errors to the error string.
3567 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3568 avoiding the need to install special zones in a real server. This is
3569 backwards compatible; if it can't find the fake resolver, it drops back.
3570 Thus, both old and new test suites can be run.
3572 TF/02 Added util/ratelimit.pl
3574 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3575 clock is set back in time.
3577 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3578 Candler <B.Candler@pobox.com>.
3580 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3581 OK for addresses that are the result of redirections.
3583 PH/48 A number of further additions for the benefit of the new test suite,
3584 including a fake gethostbyname() that interfaces to the fake DNS resolver
3587 TF/06 The fix for widen_domains has also been applied to qualify_single and
3588 search_parents which are the other dnslookup options that can cause
3591 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3594 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3596 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3597 service" if it encountered an MX record with an empty target hostname.
3598 The message is now "an MX or SRV record indicated no SMTP service".
3600 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3601 if the DNS of the sending site is misconfigured. This is quite a
3602 common situation. This change restores the behaviour of treating a
3603 helo verification defer as a failure.
3605 PH/51 If self=fail was set on a router, the bounce message did not include the
3606 actual error message.
3612 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3614 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3615 command itself ended in error, the underlying message about the transport
3616 filter timeout was being overwritten with the pipe command error. Now the
3617 underlying error message should be appended to the second error message.
3619 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3621 PH/02 Reduce the amount of output that "make" produces by default. Full output
3622 can still be requested.
3624 PH/03 The warning log line about a condition test deferring for a "warn" verb
3625 was being output only once per connection, rather than after each
3626 occurrence (because it was using the same function as for successful
3627 "warn" verbs). This seems wrong, so I have changed it.
3629 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3630 it should not have, which might have caused a crash in the right
3631 circumstances, but probably never did.
3633 PH/04 Installed a modified version of Tony Finch's patch to make submission
3634 mode fix the return path as well as the Sender: header line, and to
3635 add a /name= option so that you can make the user's friendly name appear
3638 TF/03 Added the control = fakedefer ACL modifier.
3640 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3641 Mark Lowes for thorough testing.
3643 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3645 TK/03 Merged latest SRS patch from Miles Wilton.
3647 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3648 with the definition in sysexits.h (which is #included earlier).
3649 Fortunately, Exim does not actually use EX_OK. The code used to try to
3650 preserve the sysexits.h value, by assuming that macro definitions were
3651 scanned for macro replacements. I have been disabused of this notion,
3652 so now the code just undefines EX_OK before #including unistd.h.
3654 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3655 in the smtp transport. When a block could not be written in a single
3656 write() function, the timeout was being re-applied to each part-write.
3657 This seems wrong - if the receiver was accepting one byte at a time it
3658 would take for ever. The timeout is now adjusted when this happens. It
3659 doesn't have to be particularly precise.
3661 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3662 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3664 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3665 without a domain if the check to <postmaster@domain> fails.
3667 SC/01 Eximstats: added -xls and the ability to specify output files
3668 (patch written by Frank Heydlauf).
3670 SC/02 Eximstats: use FileHandles for outputting results.
3672 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3674 SC/04 Eximstats: fixed display of large numbers with -nvr option
3676 SC/05 Eximstats: fixed merging of reports with empty tables.
3678 SC/06 Eximstats: added the -include_original_destination flag
3680 SC/07 Eximstats: removed tabs and trailing whitespace.
3682 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3684 TK/06 MBOX spool code: Add real "From " MBOX separator line
3685 so the .eml file is really in mbox format (even though
3686 most programs do not really care). Patch from Alex Miller.
3688 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3689 The latter is generated from $received_to and is only set if the
3690 message has one envelope recipient. SA can use these headers,
3691 obviously out-of-the-box. Patch from Alex Miller.
3693 PH/08 The ${def test on a variable was returning false if the variable's
3694 value was "0", contrary to what the specification has always said!
3695 The result should be true unless the variable is empty.
3697 PH/09 The syntax error of a character other than { following "${if
3698 def:variable_name" (after optional whitespace) was not being diagnosed.
3699 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3700 accidental colon was present, for example, could give incorrect results.
3702 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3703 result is used (not including appendfile, where other changes are about
3706 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3707 This involved changing a lot of size variables from int to off_t. It
3708 should work with maildirs and everything.
3710 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3711 spamd dying while we are connected to it.
3713 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3716 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3717 function for BDB 4.3.
3719 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3721 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3722 the sources for PCRE have been re-arranged and more files are now
3725 PH/15 The code I had for printing potentially long long variables in PH/11
3726 above was not the best (it lost precision). The length of off_t variables
3727 is now inspected at build time, and an appropriate printing format (%ld
3728 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3729 to be "long long int" or "long int". This is needed for the internal
3730 formatting function string_vformat().
3732 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3733 the configuration file to be ":syslog", then the script "guesses" where
3734 the logs files are, rather than using the compiled in default. In our
3735 case the guess is not the same as the compiled default, so the script
3736 suddenly stopped working when I started to use syslog. The patch checks
3737 to see if log_file_path is "". If so, it attempts to read it from exim
3738 with no configuration file to get the compiled in version, before it
3739 falls back to the previous guessing code."
3741 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3742 implementing BATV in an Exim configuration. See NewStuff for the gory
3745 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3746 Makefile that are specific to HP-UX.
3748 PH/18 If the "use_postmaster" option was set for a recipient callout together
3749 with the "random" option, the postmaster address was used as the MAIL
3750 FROM address for the random test, but not for the subsequent recipient
3751 test. It is now used for both.
3753 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3754 patch removes a few documentation additions to RFC 3028, because the
3755 latest draft now contains them. It adds the new en;ascii-case comparator
3756 and a new error check for 8bit text in MIME parts. Comparator and
3757 require names are now matched exactly. I enabled the subaddress
3758 extension, but it is not well tested yet (read: it works for me)."
3760 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3761 rework some of the code of TK/09 above to avoid the hardwired use of
3762 "%lld" and "long long". Replaced the call to snprintf() with a call to
3765 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3766 records point to non-existent hosts", "retry timeout exceeded", and
3767 "retry time not reached for any host after a long failure period".
3769 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3770 experimental DomainKeys support:
3772 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3773 (2) On an error such as an illegally used "control", the wrong name for
3774 the control was given.
3776 These problems did NOT occur unless DomainKeys support was compiled.
3778 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3780 PH/24 Added ${if match_ip condition.
3782 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3783 databases so that it will be absolutely obvious if a crash occurs in the
3784 DB library. This is a regular occurrence (often caused by mis-matched
3787 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3788 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3789 picked up on a user's system that detects such things. There doesn't seem
3790 to be a gcc warning option for this - only an attribute that has to be
3791 put on the function's prototype. It seems that in Fedora Core 4 they have
3792 set this on a number of new functions. No doubt there will be more in due
3795 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3796 specify a transport. However, if an address that was verified by such a
3797 router was the subject of a callout, Exim crashed because it tried to
3798 read the rcpt_include_affixes from the non-existent transport. Now it
3799 just assumes that the setting of that option is false. This bug was
3800 introduced by 4.51/PH/31.
3802 PH/28 Changed -d+all to exclude +memory, because that information is very
3803 rarely of interest, but it makes the output a lot bigger. People tend to
3804 do -d+all out of habit.
3806 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3807 code in os-type was giving problems when libc.so lives in lib64, like on
3810 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3811 aren't the modern standard, and it seems that some systems' include files
3812 don't always have them. Exim was already checking for some of the newer
3813 ones like T_AAAA, and defining it itself. I've added checks for all the
3814 record types that Exim uses.
3816 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3817 not automatically generating a new one, as it is supposed to. This
3818 prevented TLS from working. If the file did exist, but contained invalid
3819 data, a new version was generated, as expected. It was only the case of a
3820 non-existent file that was broken.
3822 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3823 with a change in libdomainkeys > 0.64.
3825 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3826 from DNS. If the selector record carries the flag, it now has
3827 precedence over the domain-wide flag.
3829 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3831 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3832 the use of a port name that isn't defined in /etc/services) occurred, the
3833 message was deferred as in a normal delivery, and thus remained on the
3834 spool, instead of being failed because of the mua_wrapper setting. This
3835 is now fixed, and I tidied up some of the mua_wrapper messages at the
3838 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3839 the messages in a hash of arrays rather than using individual hashes.
3840 This is a bit cleaner and results in dramatic memory savings, albeit
3841 at a slight CPU cost.
3843 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3844 as requested by Marc Sherman.
3846 SC/10 Eximstats: added histograms for user specified patterns as requested
3849 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3851 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3852 fopen() in the content-scanning modules that did not already have it.
3858 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3859 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3861 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3863 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3865 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3866 to test Sieve filters that use "vacation".
3868 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3869 that changes the way the GnuTLS parameters are stored in the cache file.
3870 The new format can be generated externally. For backward compatibility,
3871 if the data in the cache doesn't make sense, Exim assumes it has read an
3872 old-format file, and it generates new data and writes a new file. This
3873 means that you can't go back to an older release without removing the
3876 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3877 work if there are any delivery delays because "one_time" forces the
3878 parent to be marked "delivered", so its unseen clone is never tried
3879 again. For this reason, Exim now forbids the simultaneous setting of
3882 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3883 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3884 entry if you want to know the details. The fix, however, neglected to
3885 consider the case when local delivery batching is involved. The test for
3886 "previously delivered" was not happening when checking to see if an
3887 address could be batched with a previous (undelivered) one; under
3888 certain circumstances this could lead to multiple deliveries to the same
3891 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3892 in its include files, and this causes problems building Exim.
3894 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3895 header_syntax) but Exim was just ignoring anything given after a slash.
3896 In particular, this caused confusion with an attempt to use "verify =
3897 reverse_host_lookup/defer_ok". An error is now given when options are
3898 supplied for verify items that do not have them. (Maybe reverse_host_
3899 lookup should have a defer_ok option, but that's a different point.)
3901 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3902 defined by RFC 821) to 2048, because there were problems with some AUTH
3903 commands, and RFC 1869 says the size should be increased for extended
3904 SMTP commands that take arguments.
3906 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3909 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3910 "unknown" error; now it says that the functionality isn't in the binary.
3912 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3913 an address' error message when a string expansion fails (syntax or
3914 whatever). Otherwise the password may appear in the log. Following change
3915 PH/42 below, there is no longer a chance of it appearing in a bounce
3918 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3920 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3921 of its addresses was ever tried. (Bugzilla bug #2.)
3923 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3924 the result incorrectly in the debug output. (It correctly added a newline
3925 to what was transported.)
3927 TF/01 Added $received_time.
3929 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3930 commented out examples of how to interface to a virus scanner and to
3931 SpamAssassin. Also added commented examples of av_scanner and
3932 spamd_address settings.
3934 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3935 and controls are allowed in which ACLs. There were a couple of minor
3936 errors. Some of the entries in the conditions table (which is a table of
3937 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3938 negation of where the condition IS allowed.
3940 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3942 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3943 header file does not have a version number, so I've had to invent a new
3944 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3945 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3946 radiusclient), but it was contributed by a Radius user.
3948 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3949 files or directories, for OpenSSL.
3951 PH/19 When an Exim process that is running as root has to create an Exim log
3952 file, it does so in a subprocess that runs as exim:exim so as to get the
3953 ownership right at creation (otherwise, other Exim processes might see
3954 the file with the wrong ownership). There was no test for failure of this
3955 fork() call, which would lead to the process getting stuck as it waited
3956 for a non-existent subprocess. Forks do occasionally fail when resources
3957 run out. I reviewed all the other calls to fork(); they all seem to check
3960 PH/20 When checking for unexpected SMTP input at connect time (before writing
3961 the banner), Exim was not dealing correctly with a non-positive return
3962 from the read() function. If the client had disconnected by this time,
3963 the result was a log entry for a synchronization error with an empty
3964 string after "input=" when read() returned zero. If read() returned -1
3965 (an event I could not check), uninitialized data bytes were printed.
3966 There were reports of junk text (parts of files, etc) appearing after
3969 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3971 PH/22 Added support for macro redefinition, and (re)definition in between
3972 driver and ACL definitions.
3974 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3975 forgetting to use the resulting value; it was using the unexpanded value.
3977 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3978 hadn't been configured. The fix is from Juergen Kreileder, who
3979 understands it better than I do:
3981 "Here's what I see happening with three configured cyrus_sasl
3982 authenticators configured (plain, login, cram-md5):
3984 On startup auth_cyrus_sasl_init() gets called for each of these.
3985 This means three calls to sasl_listmech() without a specified mech_list.
3986 => SASL tests which mechs of all available mechs actually work
3987 => three warnings about OTP not working
3988 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3990 With the patch, sasl_listmech() also gets called three times. But now
3991 SASL's mech_list option is set to the server_mech specified in the the
3992 authenticator. Or in other words, the answer from sasl_listmech()
3993 gets limited to just the mech you're testing for (which is different
3995 => the return list contains just 'plain' or 'login', 'cram-md5' or
3996 nothing depending on the value of ob->server_mech.
3998 I've just tested the patch: Authentication still works fine,
3999 unavailable mechs specified in the exim configuration are still
4000 caught, and the auth.log warnings about OTP are gone."
4002 PH/25 When debugging is enabled, the contents of the command line are added
4003 to the debugging output, even when log_selector=+arguments is not
4006 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4007 answer is "GNU", and only if the return is "GNU/something" is the answer
4010 PH/27 $acl_verify_message is now set immediately after the failure of a
4011 verification in an ACL, and so is available in subsequent modifiers. In
4012 particular, the message can be preserved by coding like this:
4014 warn !verify = sender
4015 set acl_m0 = $acl_verify_message
4017 Previously, $acl_verify_message was set only while expanding "message"
4018 and "log_message" when a very denied access.
4020 PH/28 Modified OS/os.c-Linux with
4022 -#ifndef OS_LOAD_AVERAGE
4023 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4025 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4026 nomenclature these days.)
4028 PH/29 Installed patch from the Sieve maintainer that adds the options
4029 sieve_useraddress and sieve_subaddress to the redirect router.
4031 PH/30 In these circumstances:
4032 . Two addresses routed to the same list of hosts;
4033 . First host does not offer TLS;
4034 . First host accepts first address;
4035 . First host gives temporary error to second address;
4036 . Second host offers TLS and a TLS session is established;
4037 . Second host accepts second address.
4038 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4039 and peerdn, if requested) that were in fact used only for the second
4042 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4043 attention to any local part prefix or suffix that was matched by the
4044 router that accepted the address. It now behaves in the same way as it
4045 does for delivery: the affixes are removed from the local part unless
4046 rcpt_include_affixes is set on the transport.
4048 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4049 timeout during the DATA phase of an incoming message.
4051 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4052 applied a patch sanctioned by the Sieve maintainer.
4054 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4055 the uid or gid is negative. A case of a negative gid caused this to be
4056 noticed. The fix allows for either to be negative.
4058 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4059 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4062 PH/36 The change PH/12 above was broken. Fixed it.
4064 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4065 the grounds that routing the same address twice would always produce the
4066 same answer. This might have been true once, but it is certainly no
4067 longer true now. Routing a child address may depend on the previous
4068 routing that produced that child. Some complicated redirection strategies
4069 went wrong when messages had multiple recipients, and made Exim's
4070 behaviour dependent on the order in which the addresses were given.
4072 I have moved the duplicate checking until after the routing is complete.
4073 Exim scans the addresses that are assigned to local and remote
4074 transports, and removes any duplicates. This means that more work will be
4075 done, as duplicates will always all be routed, but duplicates are
4076 presumably rare, so I don't expect this is of any significance.
4078 For deliveries to pipes, files, and autoreplies, the duplicate checking
4079 still happens during the routing process, since they are not going to be
4082 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4083 It corrects a timeout issue with spamd. This is Ian's comment: "The
4084 background is that sometimes spamd either never reads data from a
4085 connection it has accepted, or it never writes response data. The exiscan
4086 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4087 blindly assumes that writes won't block so it may never time out."
4089 PH/39 Allow G after quota size as well as K and M.
4091 PH/40 The value set for $authenticated_id in an authenticator may not contain
4092 binary zeroes or newlines because the value is written to log lines and
4093 to spool files. There was no check on this. Now the value is run through
4094 the string_printing() function so that such characters are converted to
4095 printable escape sequences.
4097 PH/41 $message_linecount is a new variable that contains the total number of
4098 lines in the message. Compare $body_linecount, which is the count for the
4101 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4102 bounce and delay warning messages, except in certain special cases, which
4105 (a) An SMTP error message from a remote host;
4106 (b) A message specified in a :fail: redirection;
4107 (c) A message specified in a "fail" command in a system filter;
4108 (d) A message specified in a FAIL return from the queryprogram router;
4109 (e) A message specified by the cannot_route_message router option.
4111 In these cases only, Exim does include the error details in bounce and
4112 warning messages. There are also a few cases where bland messages such
4113 as "unrouteable address" or "local delivery error" are given.
4115 PH/43 $value is now also set for the "else" part of a ${run expansion.
4117 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4118 being worked on, but at least Exim now implements the latest version to
4121 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4122 process to complete was treated as a delivery failure, a timeout while
4123 writing the message to the pipe was logged, but erroneously treated as a
4124 successful delivery. Such timeouts include transport filter timeouts. For
4125 consistency with the overall process timeout, these timeouts are now
4126 treated as errors, giving rise to delivery failures by default. However,
4127 there is now a new Boolean option for the pipe transport called
4128 timeout_defer, which, if set TRUE, converts the failures into defers for
4129 both kinds of timeout. A transport filter timeout is now identified in
4132 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4133 systems where "make" and "gmake" are different, calling "gmake" at top
4134 level broke things. I've arranged for the value of $(MAKE) to be passed
4135 from the Makefile to this script so that it can call the same version of
4139 A note about Exim versions 4.44 and 4.50
4140 ----------------------------------------
4142 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4143 changes of various kinds. As a consequence, a big documentation update was
4144 needed. This delayed the release for rather longer than seemed good, especially
4145 in the light of a couple of (minor) security issues. Therefore, the changes
4146 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4147 release. So 4.44 and 4.50 are in effect two different branches that both start
4150 I have left the 4.50 change log unchanged; it contains all the changes since
4151 4.43. The change log for 4.44 is below; many of its items are identical to
4152 those for 4.50. This seems to be the most sensible way to preserve the
4153 historical information.
4159 1. Minor wording change to the doc/README.SIEVE file.
4161 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4162 computation of the current number of files was incorrect.
4164 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4165 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4168 4. Give more explanation in the error message when the command for a transport
4169 filter fails to execute.
4171 5. There are several places where Exim runs a non-Exim command in a
4172 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4173 was being done only for the command run by the queryprogram router. It is
4174 now done for all such subprocesses. The other cases are: ${run, transport
4175 filters, and the commands run by the lmtp and pipe transports.
4177 6. Added CONFIGURE_GROUP build-time option.
4179 7. Some older OS have a limit of 256 on the maximum number of file
4180 descriptors. Exim was using setrlimit() to set 1000 as a large value
4181 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4182 systems. I've change it so that if it can't get 1000, it tries for 256.
4184 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4185 was an oversight, and furthermore, ever since the addition of extra
4186 controls (e.g. 4.43/32), the checks on when to allow different forms of
4187 "control" were broken. There should now be diagnostics for all cases when a
4188 control that does not make sense is encountered.
4190 9. Added the /retain_sender option to "control=submission".
4192 10. $recipients is now available in the predata ACL (oversight).
4194 11. Tidy the search cache before the fork to do a delivery from a message
4195 received from the command line. Otherwise the child will trigger a lookup
4196 failure and thereby defer the delivery if it tries to use (for example) a
4197 cached ldap connection that the parent has called unbind on.
4199 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4200 of $address_data from the recipient verification was clobbered by the
4201 sender verification.
4203 13. The value of address_data from a sender verification is now available in
4204 $sender_address_data in subsequent conditions in the ACL statement.
4206 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4208 15. Added a new option "connect=<time>" to callout options, to set a different
4211 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4212 was its contents. (It was OK if the option was not defined at all.)
4214 17. A "Completed" log line is now written for messages that are removed from
4215 the spool by the -Mrm option.
4217 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4218 information about exactly what failed.
4220 19. Added -dd to debug only the daemon process.
4222 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4223 handles timeouts, both on the server side and network timeouts. Renamed the
4224 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4226 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4227 It is now set to "smtps".
4229 22. $host_address is now set to the target address during the checking of
4230 ignore_target_hosts.
4232 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4233 being passed; this would have caused $sender_host_name to have been used if
4234 matching the list had actually called for a host name (not very likely,
4235 since this list is usually IP addresses). A host name is now passed as
4238 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4239 code that specifies a non-restarting handler (typically sigaction() in
4240 modern systems) in an attempt to fix a rare and obscure crash bug.
4242 25. Narrowed the window for a race in the daemon that could cause it to ignore
4243 SIGCHLD signals. This is not a major problem, because they are used only to
4244 wake it up if nothing else does.
4246 26. A malformed maildirsize file could cause Exim to calculate negative values
4247 for the mailbox size or file count. Odd effects could occur as a result.
4248 The maildirsize information is now recalculated if the size or filecount
4251 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4252 support for a long time. Removed HAVE_SYS_VFS_H.
4254 28. Installed the latest version of exipick from John Jetmore.
4256 29. In an address list, if the pattern was not a regular expression, an empty
4257 subject address (from a bounce message) matched only if the pattern was an
4258 empty string. Non-empty patterns were not even tested. This was the wrong
4259 because it is perfectly reasonable to use an empty address as part of a
4260 database query. An empty address is now tested by patterns that are
4261 lookups. However, all the other forms of pattern expect the subject to
4262 contain a local part and a domain, and therefore, for them, an empty
4263 address still always fails if the pattern is not itself empty.
4265 30. Exim went into a mad DNS loop when attempting to do a callout where the
4266 host was specified on an smtp transport, and looking it up yielded more
4267 than one IP address.
4269 31. Re-factored the code for checking spool and log partition space into a
4270 function that finds that data and another that does the check. The former
4271 is then used to implement four new variables: $spool_space, $log_space,
4272 $spool_inodes, and $log_inodes.
4274 32. The RFC2047 encoding function was originally intended for short strings
4275 such as real names; it was not keeping to the 75-character limit for
4276 encoded words that the RFC imposes. It now respects the limit, and
4277 generates multiple encoded words if necessary. To be on the safe side, I
4278 have increased the buffer size for the ${rfc2047: expansion operator from
4281 33. It is now permitted to omit both strings after an "if" condition; if the
4282 condition is true, the result is "true". As before, when the second string
4283 is omitted, a false condition yields an empty string. This makes it less
4284 cumbersome to write custom ACL and router conditions.
4286 34. Failure to deliver a bounce message always caused it to be frozen, even if
4287 there was an errors_to setting on the router. The errors_to setting is now
4290 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4291 canonical form (fully expanded) before being placed in
4292 $sender_host_address.
4294 36. The table in the code that translates DNS record types into text (T_A to
4295 "A" for instance) was missing entries for NS and CNAME. It is just possible
4296 that this could have caused confusion if both these types were looked up
4297 for the same domain, because the text type is used as part of Exim's
4298 per-process caching. But the chance of anyone hitting this buglet seems
4301 37. The dnsdb lookup has been extended in a number of ways.
4303 (1) There is a new type, "zns", which walks up the domain tree until it
4304 finds some nameserver records. It should be used with care.
4306 (2) There is a new type, "mxh", which is like "mx" except that it returns
4307 just the host names, not the priorities.
4309 (3) It is now possible to give a list of domains (or IP addresses) to be
4310 looked up. The behaviour when one of the lookups defers can be
4311 controlled by a keyword.
4313 (4) It is now possible to specify the separator character for use when
4314 multiple records are returned.
4316 38. The dnslists ACL condition has been extended: it is now possible to supply
4317 a list of IP addresses and/or domains to be looked up in a particular DNS
4320 39. Added log_selector=+queue_time_overall.
4322 40. When running the queue in the test harness, wait just a tad after forking a
4323 delivery process, to get repeatability of debugging output.
4325 41. Include certificate and key file names in error message when GnuTLS fails
4326 to set them up, because the GnuTLS error message doesn't include the name
4327 of the failing file when there is a problem reading it.
4329 42. Allow both -bf and -bF in the same test run.
4331 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4333 44. The "Exiscan patch" is now merged into the mainline Exim source.
4335 45. Sometimes the final signoff response after QUIT could fail to get
4336 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4337 < 0 before doing a fflush(). This bug looks as though it goes back to the
4338 introduction of TLS in release 3.20, but "sometimes" must have been rare
4339 because the tests only now provoked it.
4341 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4342 (this can affect the format of dates).
4344 47. exim_tidydb, when checking for the continued existence of a message for
4345 which it has found a message-specific retry record, was not finding
4346 messages that were in split spool directories. Consequently, it was
4347 deleting retry records that should have stayed in existence.
4349 48. Steve fixed some bugs in eximstats.
4351 49. The SPA authentication driver was not abandoning authentication and moving
4352 on to the next authenticator when an expansion was forced to fail,
4353 contradicting the general specification for all authenticators. Instead it
4354 was generating a temporary error. It now behaves as specified.
4356 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4357 (the order specifies the preference for clients). The order is now AES256,
4358 AES128, 3DES, ARCFOUR128.
4360 51. Small patch to Sieve code - explicitly set From: when generating an
4363 52. Exim crashed if a remote delivery caused a very long error message to be
4364 recorded - for instance if somebody sent an entire SpamAssassin report back
4365 as a large number of 550 error lines. This bug was coincidentally fixed by
4366 increasing the size of one of Exim's internal buffers (big_buffer) that
4367 happened as part of the Exiscan merge. However, to be on the safe side, I
4368 have made the code more robust (and fixed the comments that describe what
4371 53. Now that there can be additional text after "Completed" in log lines (if
4372 the queue_time_overall log selector is set), a one-byte patch to exigrep
4373 was needed to allow it to recognize "Completed" as not the last thing in
4376 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4377 patch that reportedly fixes this has been added. I am not expert enough to
4378 create a test for it. This is what the patch creator wrote:
4380 "I found a little strange behaviour of ldap code when working with
4381 Windows 2003 AD Domain, where users was placed in more than one
4382 Organization Units. When I tried to give exim partial DN, the exit code
4383 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4384 But simultaneously result of request was absolutely normal ldap result,
4385 so I produce this patch..."
4387 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4388 REFERENCE, so I have modified the code to exclude the patch when that macro
4391 55. Some experimental protocols are using DNS PTR records for new purposes. The
4392 keys for these records are domain names, not reversed IP addresses. The
4393 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4394 leaves it alone. Component reversal etc. now happens only for IP addresses.
4397 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4399 57. Double the size of the debug message buffer (to 2048) so that more of very
4400 long debug lines gets shown.
4402 58. The exicyclog utility now does better if the number of log files to keep
4403 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4405 59. Two changes related to the smtp_active_hostname option:
4407 (1) $smtp_active_hostname is now available as a variable.
4408 (2) The default for smtp_banner uses $smtp_active_hostname instead
4409 of $primary_hostname.
4411 60. The host_aton() function is supposed to be passed a string that is known
4412 to be a valid IP address. However, in the case of IPv6 addresses, it was
4413 not checking this. This is a hostage to fortune. Exim now panics and dies
4414 if the condition is not met. A case was found where this could be provoked
4415 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4416 components; fortuitously, this particular loophole had already been fixed
4417 by change 4.50/55 above.
4419 If there are any other similar loopholes, the new check in host_aton()
4420 itself should stop them being exploited. The report I received stated that
4421 data on the command line could provoke the exploit when Exim was running as
4422 exim, but did not say which command line option was involved. All I could
4423 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4424 running as the user.
4427 61. There was a buffer overflow vulnerability in the SPA authentication code
4428 (which came originally from the Samba project). I have added a test to the
4429 spa_base64_to_bits() function which I hope fixes it.
4432 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4433 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4435 63. The daemon start-up calls getloadavg() while still root for those OS that
4436 need the first call to be done as root, but it missed one case: when
4437 deliver_queue_load_max is set with deliver_drop_privilege. This is
4438 necessary for the benefit of the queue runner, because there is no re-exec
4439 when deliver_drop_privilege is set.
4441 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4442 This has been fixed.
4444 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4445 was in use, was not putting the data itself into the right store pool;
4446 consequently, it could be overwritten for a subsequent message in the same
4447 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4450 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4452 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4453 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4454 the function were treating the return as a boolean value, which happened to
4455 work because 0=false and not-0=true, but is not correct code.
4457 68. The host_aton() function was not handling scoped IPv6 addresses (those
4458 with, for example, "%eth0" on the end) correctly.
4460 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4461 negated items (that is, ~something) in unsigned ints. Some compilers
4462 apparently mutter when there is no cast.
4464 70. If an address verification called from an ACL failed, and did not produce a
4465 user-specific message (i.e. there was only a "system" message), nothing was
4466 put in $acl_verify_message. In this situation, it now puts the system
4469 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4470 session; change 4.31/43 added the unwanted input to the log line - except
4471 that it did not do this in the start of session case. It now does.
4473 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4474 This is wrong and can cause the other end to generate a synchronization
4475 error if it is another Exim or anything else that does the synchronization
4476 check. A QUIT command is no longer sent after a timeout.
4478 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4479 during host lookups.
4481 74. The defer_ok option of callout verification was not working if it was used
4482 when verifying addresses in header lines, that is, for this case:
4484 verify = header_sender/callout=defer_ok
4486 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4487 those file descriptors could be used for SMTP connections. If anything
4488 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4489 could be sent to the SMTP client, causing chaos. The daemon now opens
4490 stdin, stdout, and stderr to /dev/null when it puts itself into the
4493 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4494 log by default. The user can override this with suitable Perl magic.
4496 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4497 the log message when discard triggers, was not working for the DATA ACL or
4498 for the non-SMTP ACL.
4500 78. Error message wording change in sieve.c.
4502 79. If smtp_accept_max_per_host was set, the number of connections could be
4503 restricted to fewer than expected, because the daemon was trying to set up
4504 a new connection before checking whether the processes handling previous
4505 connections had finished. The check for completed processes is now done
4506 earlier. On busy systems, this bug wouldn't be noticed because something
4507 else would have woken the daemon, and it would have reaped the completed
4510 80. If a message was submitted locally by a user whose login name contained one
4511 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4512 It caused a spool format error. I have fixed the spool reading code. A
4513 related problem was that the "from" clause in the Received: line became
4514 illegal because of the space(s). It is now covered by ${quote_local_part.
4516 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4519 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4521 83. Patch from Sieve maintainer to fix unterminated string problem in
4522 "vacation" handling.
4524 84. Some minor changes to the Linux configuration files to help with other
4525 OS variants using glibc.
4527 85. One more patch for Sieve to update vacation handling to latest spec.
4530 ----------------------------------------------------
4531 See the note above about the 4.44 and 4.50 releases.
4532 ----------------------------------------------------
4538 1. Change 4.43/35 introduced a bug that caused file counts to be
4539 incorrectly computed when quota_filecount was set in an appendfile
4542 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4543 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4546 3. Give more explanation in the error message when the command for a transport
4547 filter fails to execute.
4549 4. There are several places where Exim runs a non-Exim command in a
4550 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4551 was being done only for the command run by the queryprogram router. It is
4552 now done for all such subprocesses. The other cases are: ${run, transport
4553 filters, and the commands run by the lmtp and pipe transports.
4555 5. Some older OS have a limit of 256 on the maximum number of file
4556 descriptors. Exim was using setrlimit() to set 1000 as a large value
4557 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4558 systems. I've change it so that if it can't get 1000, it tries for 256.
4560 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4561 was an oversight, and furthermore, ever since the addition of extra
4562 controls (e.g. 4.43/32), the checks on when to allow different forms of
4563 "control" were broken. There should now be diagnostics for all cases when a
4564 control that does not make sense is encountered.
4566 7. $recipients is now available in the predata ACL (oversight).
4568 8. Tidy the search cache before the fork to do a delivery from a message
4569 received from the command line. Otherwise the child will trigger a lookup
4570 failure and thereby defer the delivery if it tries to use (for example) a
4571 cached ldap connection that the parent has called unbind on.
4573 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4574 of $address_data from the recipient verification was clobbered by the
4575 sender verification.
4577 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4578 was its contents. (It was OK if the option was not defined at all.)
4580 11. A "Completed" log line is now written for messages that are removed from
4581 the spool by the -Mrm option.
4583 12. $host_address is now set to the target address during the checking of
4584 ignore_target_hosts.
4586 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4587 being passed; this would have caused $sender_host_name to have been used if
4588 matching the list had actually called for a host name (not very likely,
4589 since this list is usually IP addresses). A host name is now passed as
4592 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4593 code that specifies a non-restarting handler (typically sigaction() in
4594 modern systems) in an attempt to fix a rare and obscure crash bug.
4596 15. Narrowed the window for a race in the daemon that could cause it to ignore
4597 SIGCHLD signals. This is not a major problem, because they are used only to
4598 wake it up if nothing else does.
4600 16. A malformed maildirsize file could cause Exim to calculate negative values
4601 for the mailbox size or file count. Odd effects could occur as a result.
4602 The maildirsize information is now recalculated if the size or filecount
4605 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4606 support for a long time. Removed HAVE_SYS_VFS_H.
4608 18. Updated exipick to current release from John Jetmore.
4610 19. Allow an empty sender to be matched against a lookup in an address list.
4611 Previously the only cases considered were a regular expression, or an
4614 20. Exim went into a mad DNS lookup loop when doing a callout where the
4615 host was specified on the transport, if the DNS lookup yielded more than
4618 21. The RFC2047 encoding function was originally intended for short strings
4619 such as real names; it was not keeping to the 75-character limit for
4620 encoded words that the RFC imposes. It now respects the limit, and
4621 generates multiple encoded words if necessary. To be on the safe side, I
4622 have increased the buffer size for the ${rfc2047: expansion operator from
4625 22. Failure to deliver a bounce message always caused it to be frozen, even if
4626 there was an errors_to setting on the router. The errors_to setting is now
4629 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4630 canonical form (fully expanded) before being placed in
4631 $sender_host_address.
4633 24. Updated eximstats to version 1.33
4635 25. Include certificate and key file names in error message when GnuTLS fails
4636 to set them up, because the GnuTLS error message doesn't include the name
4637 of the failing file when there is a problem reading it.
4639 26. Expand error message when OpenSSL has problems setting up cert/key files.
4642 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4643 (this can affect the format of dates).
4645 28. exim_tidydb, when checking for the continued existence of a message for
4646 which it has found a message-specific retry record, was not finding
4647 messages that were in split spool directories. Consequently, it was
4648 deleting retry records that should have stayed in existence.
4650 29. eximstats updated to version 1.35
4651 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4652 1.35 - bugfix such that pie charts by volume are generated correctly
4654 30. The SPA authentication driver was not abandoning authentication and moving
4655 on to the next authenticator when an expansion was forced to fail,
4656 contradicting the general specification for all authenticators. Instead it
4657 was generating a temporary error. It now behaves as specified.
4659 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4660 (the order specifies the preference for clients). The order is now AES256,
4661 AES128, 3DES, ARCFOUR128.
4663 31. Small patch to Sieve code - explicitly set From: when generating an
4666 32. Exim crashed if a remote delivery caused a very long error message to be
4667 recorded - for instance if somebody sent an entire SpamAssassin report back
4668 as a large number of 550 error lines. This bug was coincidentally fixed by
4669 increasing the size of one of Exim's internal buffers (big_buffer) that
4670 happened as part of the Exiscan merge. However, to be on the safe side, I
4671 have made the code more robust (and fixed the comments that describe what
4674 33. Some experimental protocols are using DNS PTR records for new purposes. The
4675 keys for these records are domain names, not reversed IP addresses. The
4676 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4677 leaves it alone. Component reversal etc. now happens only for IP addresses.
4680 34. The host_aton() function is supposed to be passed a string that is known
4681 to be a valid IP address. However, in the case of IPv6 addresses, it was
4682 not checking this. This is a hostage to fortune. Exim now panics and dies
4683 if the condition is not met. A case was found where this could be provoked
4684 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4685 components; fortuitously, this particular loophole had already been fixed
4686 by change 4.50/55 or 4.44/33 above.
4688 If there are any other similar loopholes, the new check in host_aton()
4689 itself should stop them being exploited. The report I received stated that
4690 data on the command line could provoke the exploit when Exim was running as
4691 exim, but did not say which command line option was involved. All I could
4692 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4693 running as the user.
4696 35. There was a buffer overflow vulnerability in the SPA authentication code
4697 (which came originally from the Samba project). I have added a test to the
4698 spa_base64_to_bits() function which I hope fixes it.
4701 36. The daemon start-up calls getloadavg() while still root for those OS that
4702 need the first call to be done as root, but it missed one case: when
4703 deliver_queue_load_max is set with deliver_drop_privilege. This is
4704 necessary for the benefit of the queue runner, because there is no re-exec
4705 when deliver_drop_privilege is set.
4707 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4708 was in use, was not putting the data itself into the right store pool;
4709 consequently, it could be overwritten for a subsequent message in the same
4710 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4713 38. Sometimes the final signoff response after QUIT could fail to get
4714 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4715 < 0 before doing a fflush(). This bug looks as though it goes back to the
4716 introduction of TLS in release 3.20, but "sometimes" must have been rare
4717 because the tests only now provoked it.
4723 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4724 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4725 Now it returns an integer. A number of calls were still expecting a T/F
4726 return. Fortuitously, in all cases, the tests worked in OK situations,
4727 which is the norm. However, things would have gone wrong on any write
4728 failures on the smtp file descriptor. This function is used when sending
4729 messages over SMTP and also when doing verify callouts.
4731 2. When Exim is called to do synchronous delivery of a locally submitted
4732 message (the -odf or -odi options), it no longer closes stderr before doing
4735 3. Implemented the mua_wrapper option.
4737 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4739 5. Implemented the functions header_remove(), header_testname(),
4740 header_add_at_position(), and receive_remove_recipient(), and exported them
4743 6. If an ACL "warn" statement specified the addition of headers, Exim already
4744 inserted X-ACL-Warn: at the start if there was no header name. However, it
4745 was not making this test for the second and subsequent header lines if
4746 there were newlines in the string. This meant that an invalid header could
4747 be inserted if Exim was badly configured.
4749 7. Allow an ACL "warn" statement to add header lines at the start or after all
4750 the Received: headers, as well as at the end.
4752 8. Added the rcpt_4xx retry error code.
4754 9. Added postmaster_mailfrom=xxx to callout verification option.
4756 10. Added mailfrom=xxxx to the callout verification option, for verify=
4759 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4760 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4762 12. Inserted some casts to stop certain compilers warning when using pointer
4763 differences as field lengths or precisions in printf-type calls (mostly
4764 affecting debugging statements).
4766 13. Added optional readline() support for -be (dynamically loaded).
4768 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4769 same clock tick as a message's arrival, so that its received time was the
4770 same as the "first fail" time on the retry record, and that message
4771 remained on the queue past the ultimate address timeout, every queue runner
4772 would try a delivery (because it was past the ultimate address timeout) but
4773 after another failure, the ultimate address timeout, which should have then
4774 bounced the address, did not kick in. This was a "< instead of <=" error;
4775 in most cases the first failure would have been in the next clock tick
4776 after the received time, and all would be well.
4778 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4779 being recognized when the domain list was tested by the match_domain
4780 condition in an expansion string.
4782 16. Added the ${str2b64: operator.
4784 17. Exim was always calling setrlimit() to set a large limit for the number of
4785 processes, without checking whether the existing limit was already
4786 adequate. (It did check for the limit on file descriptors.) Furthermore,
4787 errors from getrlimit() and setrlimit() were being ignored. Now they are
4788 logged to the main and panic logs, but Exim does carry on, to try to do its
4789 job under whatever limits there are.
4791 18. Imported PCRE 5.0.
4793 19. Trivial typo in log message " temporarily refused connection" (the leading
4796 20. If the log selector return_path_on_delivery was set and an address was
4797 redirected to /dev/null, the delivery process crashed because it assumed
4798 that a return path would always be set for a "successful" delivery. In this
4799 case, the whole delivery is bypassed as an optimization, and therefore no
4802 21. Internal re-arrangement: the function for sending a challenge and reading
4803 a response while authentication was assuming a zero-terminated challenge
4804 string. It's now changed to take a pointer and a length, to allow for
4805 binary data in such strings.
4807 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4809 23. Exim was not respecting finduser_retries when seeking the login of the
4810 uid under which it was called; it was always trying 10 times. (The default
4811 setting of finduser_retries is zero.) Also, it was sleeping after the final
4812 failure, which is pointless.
4814 24. Implemented tls_on_connect_ports.
4816 25. Implemented acl_smtp_predata.
4818 26. If the domain in control=submission is set empty, Exim assumes that the
4819 authenticated id is a complete email address when it generates From: or
4820 Sender: header lines.
4822 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4823 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4824 chown and chgrp in /bin and hostname in /usr/bin.
4826 28. Exim was keeping the "process log" file open after each use, just as it
4827 does for the main log. This opens the possibility of it remaining open for
4828 long periods when the USR1 signal hits a daemon. Occasional processlog
4829 errors were reported, that could have been caused by this. Anyway, it seems
4830 much more sensible not to leave this file open at all, so that is what now
4833 29. The long-running daemon process does not normally write to the log once it
4834 has entered its main loop, and it closes the log before doing so. This is
4835 so that log files can straightforwardly be renamed and moved. However,
4836 there are a couple of unusual error situations where the daemon does write
4837 log entries, and I had neglected to close the log afterwards.
4839 30. The text of an SMTP error response that was received during a remote
4840 delivery was being truncated at 512 bytes. This is too short for some of
4841 the long messages that one sometimes sees. I've increased the limit to
4844 31. It is now possible to make retry rules that apply only when a message has a
4845 specific sender, in particular, an empty sender.
4847 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4848 it possible to be selective about when SMTP synchronization is enforced.
4850 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4852 32. Implemented hosts_connection_nolog.
4854 33. Added an ACL for QUIT.
4856 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4859 35. Added mailbox_size and mailbox_filecount to appendfile.
4861 36. Added control = no_multiline_responses to ACLs.
4863 37. There was a bug in the logic of the code that waits for the clock to tick
4864 in the case where the clock went backwards by a substantial amount such
4865 that the microsecond fraction of "now" was more than the microsecond
4866 fraction of "then" (but the whole seconds number was less).
4868 38. Added support for the libradius Radius client library this is found on
4869 FreeBSD (previously only the radiusclient library was supported).
4875 1. When certain lookups returned multiple values in the form name=value, the
4876 quoting of the values was not always being done properly. Specifically:
4877 (a) If the value started with a double quote, but contained no whitespace,
4879 (b) If the value contained whitespace other than a space character (i.e.
4880 tabs or newlines or carriage returns) it was not quoted.
4881 This fix has been applied to the mysql and pgsql lookups by writing a
4882 separate quoting function and calling it from the lookup code. The fix
4883 should probably also be applied to nisplus, ibase and oracle lookups, but
4884 since I cannot test any of those, I have not disturbed their existing code.
4886 2. A hit in the callout cache for a specific address caused a log line with no
4887 reason for rejecting RCPT. Now it says "Previous (cached) callout
4888 verification failure".
4890 3. There was an off-by-one bug in the queryprogram router. An over-long
4891 return line was truncated at 256 instead of 255 characters, thereby
4892 overflowing its buffer with the terminating zero. As well as fixing this, I
4893 have increased the buffer size to 1024 (and made a note to document this).
4895 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4896 when Exim is waiting for an SMTP response from a remote server, Exim
4897 restarts its select() call on the socket, thereby resetting its timeout.
4898 This is not a problem when such interrupts are rare. Somebody set up a cron
4899 job to run exiwhat every 2 minutes, which is less than the normal select()
4900 timeout (5 or 10 minutes). This meant that the select() timeout never
4901 kicked in because it was always reset. I have fixed this by comparing the
4902 time when an interrupt arrives with the time at the start of the first call
4903 to select(). If more time than the timeout has elapsed, the interrupt is
4904 treated as a timeout.
4906 5. Some internal re-factoring in preparation for the addition of Sieve
4907 extensions (by MH). In particular, the "personal" test is moved to a
4908 separate function, and given an option for scanning Cc: and Bcc: (which is
4909 not set for Exim filters).
4911 6. When Exim created an email address using the login of the caller as the
4912 local part (e.g. when creating a From: or Sender: header line), it was not
4913 quoting the local part when it contained special characters such as @.
4915 7. Installed new OpenBSD configuration files.
git://git.exim.org / exim.git / blob