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 Rework error string handling in TLS interface so that the caller in
12 more cases is responsible for logging. This permits library-sourced
13 string to be attached to addresses during delivery, and collapses
14 pairs of long lines into single ones.
16 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
17 during configuration. Wildcards are allowed and expanded.
23 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
24 than -2003 did; needs libidn2 in addition to libidn.
26 JH/02 The path option on a pipe transport is now expanded before use.
28 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
29 Patch provided by "Björn", documentation fix added too.
31 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
32 missing a wire-to-host endian conversion.
34 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
35 close after a BDAT command line could be taken as a following command,
36 giving a synch failure. Fix by only checking for synch immediately
37 before acknowledging the chunk.
39 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
40 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
41 macro. Patches provided by Josh Soref.
43 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
44 Previously we did not; the RFC seems ambiguous and VRFY is not listed
45 by IANA as a service extension. However, John Klensin suggests that we
48 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
49 the dkim code may be unix-mode line endings rather than smtp wire-format
50 CRLF, so prepend a CR to any bare LF.
52 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
53 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
55 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
56 the first were themselves being wrongly included in the feed into dkim
57 processing; with most chunk sizes in use this resulted in an incorrect
58 body hash calculated value.
60 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
61 DKIM signature block, for verification. Although advised against by
62 standards it is specifically not ruled illegal.
64 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
66 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
67 missing a body hash (the bh= tag).
69 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
70 It seems that HAProxy sends the Proxy Protocol information in clear and
71 only then does a TLS startup, so do the same.
73 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
74 TCP connections (such as for Spamd) unless the daemon successfully set
75 Fast Open mode on its listening sockets. This fixes breakage seen on
76 too-old kernels or those not configured for Fast Open, at the cost of
77 requiring both directions being enabled for TFO, and TFO never being used
78 by non-daemon-related Exim processes.
80 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
81 endings, at least on the first header line. Try to canonify any that get
82 past that check, despite the cost.
84 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
85 now limited to an arbitrary five deep, while parsing addresses with the
86 strip_excess_angle_brackets option enabled.
88 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
89 instead leave the unprompted TLS handshake in socket buffer for the
90 TLS library to consume.
92 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
94 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
96 JH/16 Drop variables when they go out of scope. Memory management drops a whole
97 region in one operation, for speed, and this leaves assigned pointers
98 dangling. Add checks run only under the testsuite which checks all
99 variables at a store-reset and panics on a dangling pointer; add code
100 explicitly nulling out all the variables discovered. Fixes one known
101 bug: a transport crash, where a dangling pointer for $sending_ip_address
102 originally assigned in a verify callout, is re-used.
104 PP/06 Drop '.' from @INC in various Perl scripts.
106 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
108 PP/08 Reduce a number of compilation warnings under clang; building with
109 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
110 should be warning-free.
112 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
114 HS/01 Fix portability problems introduced by PP/08 for platforms where
115 realloc(NULL) is not equivalent to malloc() [SunOS et al].
117 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
118 chunk. This allows us to accept broken chunked messages. We need a more
119 general solution here.
125 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
126 supports it and a size is available (ie. the sending peer gave us one).
128 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
129 years of being deprecated). The replacements are the ACLs
130 acl_smtp_mime and acl_not_smtp_mime.
132 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
133 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
134 or even an in-clear connection were permitted. Now, if the host lookup
135 was dnssec and dane was requested then the host is only used if the
136 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
137 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
138 if one fails this test.
139 This means that a poorly-configured remote DNS will make it incommunicado;
140 but it protects against a DNS-interception attack on it.
142 JH/04 Bug 1810: make continued-use of an open smtp transport connection
143 non-noisy when a race steals the message being considered.
145 JH/05 If main configuration option tls_certificate is unset, generate a
146 self-signed certificate for inbound TLS connections.
148 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
149 in rewrites and routers.
151 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
152 and logged a warning sing 4.83; now they are a configuration file error.
154 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
155 (lacking @domain). Apply the same qualification processing as RCPT.
157 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
159 JH/10 Support ${sha256:} applied to a string (as well as the previous
162 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
163 a cutthrough deliver is pending, as we always want to make a connection.
164 This also avoids re-routing the message when later placing the cutthrough
165 connection after a verify cache hit.
166 Do not update it with the verify result either.
168 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
169 when routing results in more than one destination address.
171 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
172 signing (which inhibits the cutthrough capability). Previously only
173 the presence of an option was tested; now an expansion evaluating as
174 empty is permissible (obviously it should depend only on data available
175 when the cutthrough connection is made).
177 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
178 the relevant preceding SMTP command did not note the pipelining mode.
180 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
181 Previously they were not counted.
183 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
184 as one having no matching records. Previously we deferred the message
185 that needed the lookup.
187 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
188 distinguished as "(=".
190 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
191 for missing MX records. Previously it only worked for missing A records.
193 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
195 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
196 after the data-go-ahead and data-ack. Patch from Jason Betts.
198 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
199 even for a "none" policy. Patch from Tony Meyer.
201 JH/22 Fix continued use of a connection for further deliveries. If a port was
202 specified by a router, it must also match for the delivery to be
205 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
206 When one of the recipients of a message was unsuitable for the connection
207 (has no matching addresses), we lost track of needing to mark it
208 deferred. As a result mail would be lost.
210 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
212 JH/25 Decoding ACL controls is now done using a binary search; the source code
213 takes up less space and should be simpler to maintain. Merge the ACL
214 condition decode tables also, with similar effect.
216 JH/26 Fix problem with one_time used on a redirect router which returned the
217 parent address unchanged. A retry would see the parent address marked as
218 delivered, so not attempt the (identical) child. As a result mail would
221 JH/27 Fix a possible security hole, wherein a process operating with the Exim
222 UID can gain a root shell. Credit to http://www.halfdog.net/ for
223 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
226 JH/28 Enable {spool,log} filesystem space and inode checks as default.
227 Main config options check_{log,spool}_{inodes,space} are now
228 100 inodes, 10MB unless set otherwise in the configuration.
230 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
231 Previously it only applied to the main-section connection policy
234 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
236 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
237 by me. Added RFC7919 DH primes as an alternative.
239 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
240 are not in the system include path.
242 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
243 GnuTLS, when a session startup failed (eg because the client disconnected)
244 Exim did stdio operations after fclose. This was exposed by a recent
245 change which nulled out the file handle after the fclose.
247 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
248 signed directly by the cert-signing cert, rather than an intermediate
249 OCSP-signing cert. This is the model used by LetsEncrypt.
251 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
253 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
254 an incoming connection.
256 HS/02 Bug 1802: Do not half-close the connection after sending a request
259 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
260 fallback to "prime256v1".
262 JH/34 SECURITY: Use proper copy of DATA command in error message.
263 Could leak key material. Remotely exploitable. CVE-2016-9963.
269 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
270 and 3.4.4 - once the server is enabled to respond to an OCSP request
271 it does even when not requested, resulting in a stapling non-aware
272 client dropping the TLS connection.
274 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
275 support variable-length bit vectors. No functional change.
277 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
278 The I= interface field on outgoing lines is now after the H= remote
279 host field, same as incoming lines. There is a separate
280 outgoing_interface log selector which allows you to disable the
283 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
284 If not running log_selector +smtp_connection the mainlog would be held
285 open indefinitely after a "too many connections" event, including to a
286 deleted file after a log rotate. Leave the per net connection logging
287 leaving it open for efficiency as that will be quickly detected by the
288 check on the next write.
290 HS/01 Bug 1671: Fix post transport crash.
291 Processing the wait-<transport> messages could crash the delivery
292 process if the message IDs didn't exist for some reason. When
293 using 'split_spool_directory=yes' the construction of the spool
294 file name failed already, exposing the same netto behaviour.
296 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
297 mime_regex ACL conditions.
299 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
300 to DSN fail messages (bounces): remote IP, remote greeting, remote response
301 to HELO, local diagnostic string.
303 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
304 log line to debug. Even when configured with a tls authenticator many
305 client connections are expected to not authenticate in this way, so
306 an authenticate fail is not an error.
308 HS/02 Add the Exim version string to the process info. This way exiwhat
309 gives some more detail about the running daemon.
311 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
312 matter for fast-change records such as DNSBLs.
314 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
315 retry record, even if constant. There may be multiple transports with
316 different interface settings and the retry behaviour needs to be kept
319 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
321 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
323 JH/10 Bug 840: fix log_defer_output option of pipe transport
325 JH/11 Bug 830: use same host for all RCPTS of a message, even under
326 hosts_randomize. This matters a lot when combined with mua_wrapper.
328 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
329 ${quote_pgsql:<string>} operator.
331 JH/13 Bug 1708: avoid misaligned access in cached lookup.
333 JH/14 Change header file name for freeradius-client. Relevant if compiling
334 with Radius support; from the Gentoo tree and checked under Fedora.
336 JH/15 Bug 1712: Introduce $prdr_requested flag variable
338 JH/16 Bug 1714: Permit an empty string as expansion result for transport
339 option transport_filter, meaning no filtering.
341 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
343 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
344 defaults to "*" (all hosts). The variable is now available when not built
345 with TLS, default unset, mainly to enable keeping the testsuite sane.
346 If a server certificate is not supplied (via tls_certificate) an error is
347 logged, and clients will find TLS connections fail on startup. Presumably
348 they will retry in-clear.
349 Packagers of Exim are strongly encouraged to create a server certificate
350 at installation time.
352 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
353 with the $config_file variable.
355 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
356 in transport context, after the attempt, and per-recipient. The latter type
357 is per host attempted. The event data is the error message, and the errno
358 information encodes the lookup type (A vs. MX) used for the (first) host,
359 and the trailing two digits of the smtp 4xx response.
361 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
362 to write to mainlog (or rejectlog, paniclog) in the window between file
363 creation and permissions/ownership being changed. Particularly affects
364 installations where exicyclog is run as root, rather than exim user;
365 result is that the running daemon panics and dies.
367 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
369 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
370 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
371 "pri" and "weight". Note that the previous implicit priority given by the
372 list order is no longer honoured.
374 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
377 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
378 by defining SUPPORT_SOCKS.
380 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
381 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
382 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
383 variables are renamed to proxy_{local,external}_{address,port}.
385 JH/27 Move Internationalisation support from Experimental to mainline, enabled
386 for a build by defining SUPPORT_I18N
388 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
389 of the query string, and make ${quote_redis:} do that quoting.
391 JH/29 Move Events support from Experimental to mainline, enabled by default
392 and removable for a build by defining DISABLE_EVENT.
394 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
396 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
397 cached by the daemon.
399 JH/32 Move Redis support from Experimental to mainline, enabled for a build
400 by defining LOOKUP_REDIS. The libhiredis library is required.
402 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
403 keys are given for lookup.
405 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
406 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
407 only supported when built with TLS support. The PolarSSL SHA routines
408 are still used when the TLS library is too old for convenient support.
410 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
411 openssl_options), for security. OpenSSL forces this from version 1.1.0
412 server-side so match that on older versions.
414 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
415 allocation for $value could be released as the expansion processing
416 concluded, but leaving the global pointer active for it.
418 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
419 and to use the domains and local_parts ACL conditions.
421 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
422 incorrectly not doubled on cutthrough transmission, hence seen as a
423 body-termination at the receiving system - resulting in truncated mails.
424 Commonly the sender saw a TCP-level error, and retransmitted the message
425 via the normal store-and-forward channel. This could result in duplicates
426 received - but deduplicating mailstores were liable to retain only the
427 initial truncated version.
429 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
431 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
433 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
434 we're in there, support oversigning also; bug 1309.
436 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
438 HS/04 Add support for keep_environment and add_environment options.
440 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
441 either intentional arithmetic overflow during PRNG, or testing config-
444 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
445 delivery resulted in actual delivery. Cancel cutthrough before DATA
448 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
449 rejects a recipient: pass the reject to the originator.
451 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
452 Many were false-positives and ignorable, but it's worth fixing the
455 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
456 for the new environment-manipulation done at startup. Move the routines
457 from being local to tls.c to being global via the os.c file.
459 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
460 an extract embedded as result-arg for a map, the first arg for extract
461 is unavailable so we cannot tell if this is a numbered or keyed
462 extraction. Accept either.
468 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
471 JH/02 The smtp transport option "multi_domain" is now expanded.
473 JH/03 The smtp transport now requests PRDR by default, if the server offers
476 JH/04 Certificate name checking on server certificates, when exim is a client,
477 is now done by default. The transport option tls_verify_cert_hostnames
478 can be used to disable this per-host. The build option
479 EXPERIMENTAL_CERTNAMES is withdrawn.
481 JH/05 The value of the tls_verify_certificates smtp transport and main options
482 default to the word "system" to access the system default CA bundle.
483 For GnuTLS, only version 3.0.20 or later.
485 JH/06 Verification of the server certificate for a TLS connection is now tried
486 (but not required) by default. The verification status is now logged by
487 default, for both outbound TLS and client-certificate supplying inbound
490 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
493 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
494 Status Notification (bounce) messages are now MIME format per RFC 3464.
495 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
496 under the control of the dsn_advertise_hosts option, and routers may
497 have a dsn_lasthop option.
499 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
500 default, modifiable by a malware= option. The list separator for
501 the options can now be changed in the usual way. Bug 68.
503 JH/10 The smtp_receive_timeout main option is now expanded before use.
505 JH/11 The incoming_interface log option now also enables logging of the
506 local interface on delivery outgoing connections.
508 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
509 if the interface and destination host and port all match.
511 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
514 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
515 Patch from Andrew Lewis.
517 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
518 now supports optional time-restrictions, weighting, and priority
519 modifiers per server. Patch originally by <rommer@active.by>.
521 JH/16 The spamd_address main option now supports a mixed list of local
522 and remote servers. Remote servers can be IPv6 addresses, and
523 specify a port-range.
525 JH/17 Bug 68: The spamd_address main option now supports an optional
526 timeout value per server.
528 JH/18 Bug 1581: Router and transport options headers_add/remove can
529 now have the list separator specified.
531 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
534 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
537 JH/21 Support for the A6 type of dns record is withdrawn.
539 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
540 rather than the verbs used.
542 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
543 from 255 to 1024 chars.
545 JH/24 Verification callouts now attempt to use TLS by default.
547 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
548 are generic router options now. The defaults didn't change.
550 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
551 Original patch from Alexander Shikoff, worked over by JH.
553 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
554 files if ZCAT_COMMAND is not executable.
556 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
558 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
560 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
561 Normally benign, it bites when the pair was led to by a CNAME;
562 modern usage is to not canonicalize the domain to a CNAME target
563 (and we were inconsistent anyway for A-only vs AAAA+A).
565 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
567 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
568 when evaluating $sender_host_dnssec.
570 JH/31 Check the HELO verification lookup for DNSSEC, adding new
571 $sender_helo_dnssec variable.
573 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
575 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
577 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
579 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
580 documented as working, but never had. Support all but $spam_report.
582 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
583 added for tls authenticator.
585 HS/03 Add perl_taintmode main config option
591 TL/01 When running the test suite, the README says that variables such as
592 no_msglog_check are global and can be placed anywhere in a specific
593 test's script, however it was observed that placement needed to be near
594 the beginning for it to behave that way. Changed the runtest perl
595 script to read through the entire script once to detect and set these
596 variables, reset to the beginning of the script, and then run through
597 the script parsing/test process like normal.
599 TL/02 The BSD's have an arc4random API. One of the functions to induce
600 adding randomness was arc4random_stir(), but it has been removed in
601 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
602 function when detected.
604 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
605 cause callback expansion.
607 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
608 syntax errors in an expansion can be treated as a string instead of
609 logging or causing an error, due to the internal use of bool_lax
610 instead of bool when processing it.
612 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
613 server certificates when making smtp deliveries.
615 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
617 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
619 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
621 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
622 Merged patch from Sebastian Wiedenroth.
624 JH/05 Fix results-pipe from transport process. Several recipients, combined
625 with certificate use, exposed issues where response data items split
626 over buffer boundaries were not parsed properly. This eventually
627 resulted in duplicates being sent. This issue only became common enough
628 to notice due to the introduction of connection certificate information,
629 the item size being so much larger. Found and fixed by Wolfgang Breyha.
631 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
632 size buffer was used, resulting in syntax errors when an expansion
635 JH/07 Add support for directories of certificates when compiled with a GnuTLS
636 version 3.3.6 or later.
638 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
639 is EXPERIMENTAL_EVENT, the main-configuration and transport options
640 both become "event_action", the variables become $event_name, $event_data
641 and $event_defer_errno. There is a new variable $verify_mode, usable in
642 routers, transports and related events. The tls:cert event is now also
643 raised for inbound connections, if the main configuration event_action
646 TL/06 In test suite, disable OCSP for old versions of openssl which contained
647 early OCSP support, but no stapling (appears to be less than 1.0.0).
649 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
650 server certificate names available under the smtp transport option
651 "tls_verify_cert_hostname" now do not permit multi-component wildcard
654 JH/10 Time-related extraction expansions from certificates now use the main
655 option "timezone" setting for output formatting, and are consistent
656 between OpenSSL and GnuTLS compilations. Bug 1541.
658 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
659 encoded parameter in the incoming message. Bug 1558.
661 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
662 include certificate info, eximon was claiming there were spoolfile
665 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
667 JH/14 Log delivery-related information more consistently, using the sequence
668 "H=<name> [<ip>]" wherever possible.
670 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
671 are problematic for Debian distribution, omit them from the release
674 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
676 JH/16 Fix string representation of time values on 64bit time_t architectures.
679 JH/17 Fix a null-indirection in certextract expansions when a nondefault
680 output list separator was used.
685 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
686 checkers that were complaining about end of non-void function with no
689 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
690 This was a regression introduced in 4.83 by another bugfix.
692 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
694 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
695 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
701 TF/01 Correctly close the server side of TLS when forking for delivery.
703 When a message was received over SMTP with TLS, Exim failed to clear up
704 the incoming connection properly after forking off the child process to
705 deliver the message. In some situations the subsequent outgoing
706 delivery connection happened to have the same fd number as the incoming
707 connection previously had. Exim would try to use TLS and fail, logging
708 a "Bad file descriptor" error.
710 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
711 utilities have not been installed.
713 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
714 temporary space as the ACL may create new global variables.
716 TL/01 LDAP support uses per connection or global context settings, depending
717 upon the detected version of the libraries at build time.
719 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
720 to extract and use the src ip:port in logging and expansions as if it
721 were a direct connection from the outside internet. PPv2 support was
722 updated based on HAProxy spec change in May 2014.
724 JH/02 Add ${listextract {number}{list}{success}{fail}}.
726 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
727 Properly escape header and check for NULL return.
729 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
732 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
734 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
735 characters in header names, implemented as a verify condition.
736 Contributed by Michael Fischer v. Mollard.
738 TL/05 Rename SPF condition results err_perm and err_temp to standardized
739 results permerror and temperror. Previous values are deprecated but
740 still accepted. In a future release, err_perm and err_temp will be
741 completely removed, which will be a backward incompatibility if the
742 ACL tests for either of these two old results. Patch contributed by
743 user bes-internal on the mailing list.
745 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
747 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
748 selectors, in both main and reject logs.
750 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
753 JH/07 Add malware type "sock" for talking to simple daemon.
755 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
757 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
758 routers/transports under cutthrough routing.
760 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
761 numbers. Touch up "bool" conditional to keep the same definition.
763 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
765 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
767 JH/12 Expand items in router/transport headers_add or headers_remove lists
768 individually rather than the list as a whole. Bug 1452.
770 Required for reasonable handling of multiple headers_ options when
771 they may be empty; requires that headers_remove items with embedded
772 colons must have them doubled (or the list-separator changed).
774 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
775 view the policy declared in the DMARC record. Currently, $dmarc_status
776 is a combined value of both the record presence and the result of the
779 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
781 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
782 dnslookup router and the smtp transport (applying to the forward
785 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
786 of ldap servers used for a specific lookup. Patch provided by Heiko
789 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
790 New variable $lookup_dnssec_authenticated for observability.
792 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
793 Patch submitted by Lars Timman.
795 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
797 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
798 Requires trusted mode and valid format message id, aborts otherwise.
799 Patch contributed by Heiko Schlichting.
801 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
802 certextract with support for various fields. Bug 1358.
804 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
805 is requested by default, modifiable by smtp transport option
808 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
809 operate on certificate variables to give certificate fingerprints
810 Also new ${sha256:cert_variable}.
812 JH/23 The PRDR feature is moved from being Experimental into the mainline.
814 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
815 Christian Aistleitner.
817 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
819 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
820 file. Patch from Wolfgang Breyha.
822 JH/25 Expand the coverage of the delivery $host and $host_address to
823 client authenticators run in verify callout. Bug 1476.
825 JH/26 Port service names are now accepted for tls_on_connect_ports, to
826 align with daemon_smtp_ports. Bug 72.
828 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
829 support and error reporting did not work properly.
831 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
832 and is readable. Patch from Andrew Colin Kissa.
834 TL/14 Enhance documentation of ${run expansion and how it parses the
835 commandline after expansion, particularly in the case when an
836 unquoted variable expansion results in an empty value.
838 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
840 PP/02 Fix internal collision of T_APL on systems which support RFC3123
841 by renaming away from it. Addresses GH issue 15, reported by
844 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
846 TL/15 SECURITY: prevent double expansion in math comparison functions
847 (can expand unsanitized data). Not remotely exploitable.
854 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
856 PP/02 Make -n do something, by making it not do something.
857 When combined with -bP, the name of an option is not output.
859 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
862 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
863 $sender_host_name and config options to manage this, and basic check
866 PP/05 DSCP support for outbound connections and control modifier for inbound.
868 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
869 (Only plugin which currently uses this is kerberos4, which nobody should
870 be using, but we should make it available and other future plugins might
871 conceivably use it, even though it would break NAT; stuff *should* be
872 using channel bindings instead).
874 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
875 name; added for Sendmail compatibility; requires admin caller.
876 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
877 just ignore it); requires trusted caller.
878 Also parse but ignore: -Ac -Am -X<logfile>
881 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
883 TL/02 Add +smtp_confirmation as a default logging option.
885 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
886 Patch by Magnus Holmgren from 2007-02-20.
888 TL/04 Bugzilla 1281 - Spec typo.
889 Bugzilla 1283 - Spec typo.
890 Bugzilla 1290 - Spec grammar fixes.
892 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
894 TL/06 Add Experimental DMARC support using libopendmarc libraries.
896 TL/07 Fix an out of order global option causing a segfault. Reported to dev
897 mailing list by by Dmitry Isaikin.
899 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
901 JH/02 Support "G" suffix to numbers in ${if comparisons.
903 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
905 NM/01 Bugzilla 1197 - Spec typo
906 Bugzilla 1196 - Spec examples corrections
908 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
910 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
911 gnutls_enable_pkcs11, but renamed to more accurately indicate its
914 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
915 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
917 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
918 "acl {{name}{arg}...}", and optional args on acl condition
921 JH/05 Permit multiple router/transport headers_add/remove lines.
923 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
925 JH/07 Avoid using a waiting database for a single-message-only transport.
926 Performance patch from Paul Fisher. Bugzilla 1262.
928 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
931 JH/09 Add $headers_added variable, with content from use of ACL modifier
932 add_header (but not yet added to the message). Bugzilla 199.
934 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
935 Pulled from Bugzilla 817 by Wolfgang Breyha.
937 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
939 (nb: this is the same fix as in Exim 4.80.1)
941 JH/11 Add A= logging on delivery lines, and a client_set_id option on
944 JH/12 Add optional authenticated_sender logging to A= and a log_selector
947 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
949 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
950 advertise SMTP AUTH mechanism to us, instead of a generic
951 protocol violation error. Also, make Exim more robust to bad
952 data from the Dovecot auth socket.
954 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
956 When a queue runner is handling a message, Exim first routes the
957 recipient addresses, during which it prunes them based on the retry
958 hints database. After that it attempts to deliver the message to
959 any remaining recipients. It then updates the hints database using
962 So if a recipient address works intermittently, it can get repeatedly
963 deferred at routing time. The retry hints record remains fresh so the
964 address never reaches the final cutoff time.
966 This is a fairly common occurrence when a user is bumping up against
967 their storage quota. Exim had some logic in its local delivery code
968 to deal with this. However it did not apply to per-recipient defers
969 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
971 This change adds a proper retry rule check during routing so that the
972 final cutoff time is checked against the message's age. We only do
973 this check if there is an address retry record and there is not a
974 domain retry record; this implies that previous attempts to handle
975 the address had the retry_use_local_parts option turned on. We use
976 this as an approximation for the destination being like a local
977 delivery, as in LMTP.
979 I suspect this new check makes the old local delivery cutoff check
980 redundant, but I have not verified this so I left the code in place.
982 TF/02 Correct gecos expansion when From: is a prefix of the username.
984 Test 0254 submits a message to Exim with the header
988 When I ran the test suite under the user fanf2, Exim expanded
989 the header to contain my full name, whereas it should have added
990 a Resent-Sender: header. It erroneously treats any prefix of the
991 username as equal to the username.
993 This change corrects that bug.
995 GF/01 DCC debug and logging tidyup
996 Error conditions log to paniclog rather than rejectlog.
997 Debug lines prefixed by "DCC: " to remove any ambiguity.
999 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1001 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1002 Bug spotted by Jeremy Harris; was flawed since initial commit.
1003 Would have resulted in OCSP responses post-SNI triggering an Exim
1004 NULL dereference and crash.
1006 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1008 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1009 Bug detection, analysis and fix by Samuel Thibault.
1010 Bugzilla 1331, Debian bug #698092.
1012 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1014 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1015 Server implementation by Todd Lyons, client by JH.
1016 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1017 config variable "prdr_enable" controls whether the server
1018 advertises the facility. If the client requests PRDR a new
1019 acl_data_smtp_prdr ACL is called once for each recipient, after
1020 the body content is received and before the acl_smtp_data ACL.
1021 The client is controlled by both of: a hosts_try_prdr option
1022 on the smtp transport, and the server advertisement.
1023 Default client logging of deliveries and rejections involving
1024 PRDR are flagged with the string "PRDR".
1026 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1027 fclose(). Diagnosis by Todd Lyons.
1029 PP/17 Update configure.default to handle IPv6 localhost better.
1030 Patch by Alain Williams (plus minor tweaks).
1033 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1034 This is now consistent with GnuTLS, and is now documented: the
1035 previous undocumented portable approach to treating the option as
1036 unset was to force an expansion failure. That still works, and
1037 an empty string is now equivalent.
1039 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1040 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1041 not performing validation itself.
1043 PP/20 Added force_command boolean option to pipe transport.
1044 Patch from Nick Koston, of cPanel Inc.
1046 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1049 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1051 PP/21 Fix eximon continuous updating with timestamped log-files.
1052 Broken in a format-string cleanup in 4.80, missed when I repaired the
1053 other false fix of the same issue.
1054 Report and fix from Heiko Schlichting.
1057 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1058 Report from Prashanth Katuri.
1060 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1061 It's SecureTransport, so affects any MacOS clients which use the
1062 system-integrated TLS libraries, including email clients.
1064 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1065 using a MIME ACL for non-SMTP local injection.
1066 Report and assistance in diagnosis by Warren Baker.
1068 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1070 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1072 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1073 last id that failed so it may be referenced in subsequent ACL's.
1075 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1078 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1079 ldap library initialization, allowing self-signed CA's to be
1080 used. Also properly sets require_cert option later in code by
1081 using NULL (global ldap config) instead of ldap handle (per
1082 session). Bug diagnosis and testing by alxgomz.
1084 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1085 the src/util/ subdirectory.
1087 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1088 renamed to Transport Post Delivery Action by Jeremy Harris, as
1091 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1092 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1093 redis_servers = needs to be configured which will be used by the redis
1094 lookup. Patch from Warren Baker, of The Packet Hub.
1096 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1098 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1099 hostname or reverse DNS when processing a host list. Used suggestions
1100 from multiple comments on this bug.
1102 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1104 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1105 Missed a few lines, added it to make the runtest require no keyboard
1108 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1109 contains upper case chars. Make router use caseful_local_part.
1111 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1112 support when GnuTLS has been built with p11-kit.
1118 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1120 This, or similar/improved, will also be change PP/11 of 4.82.
1126 PP/01 Handle short writes when writing local log-files.
1127 In practice, only affects FreeBSD (8 onwards).
1128 Bugzilla 1053, with thanks to Dmitry Isaikin.
1130 NM/01 Bugzilla 949 - Documentation tweak
1132 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1135 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1137 PP/02 Implemented gsasl authenticator.
1139 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1141 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1142 `pkg-config foo` for cflags/libs.
1144 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1145 with rest of GSASL and with heimdal_gssapi.
1147 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1148 `pkg-config foo` for cflags/libs for the TLS implementation.
1150 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1151 properties get this fed in as external SSF. A number of robustness
1152 and debugging improvements to the cyrus_sasl authenticator.
1154 PP/08 cyrus_sasl server now expands the server_realm option.
1156 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1157 Patch by Jeremy Harris.
1159 PP/10 Added dbmjz lookup type.
1161 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1163 PP/12 MAIL args handles TAB as well as SP, for better interop with
1164 non-compliant senders.
1165 Analysis and variant patch by Todd Lyons.
1167 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1168 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1169 Patch from Dirk Mueller <dmueller@suse.com>
1171 PP/13 tls_peerdn now print-escaped for spool files.
1172 Observed some $tls_peerdn in wild which contained \n, which resulted
1173 in spool file corruption.
1175 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1176 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1177 or write after TLS renegotiation, which otherwise led to messages
1180 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1181 as a tracking header (ie: a signed header comes before the signature).
1182 Patch from Wolfgang Breyha.
1184 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1185 comma-sep list; embedded commas doubled.
1187 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1189 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1191 Report and patch from Dmitry Banschikov.
1193 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1194 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1195 failures. We appear to now support TLS1.1+ with Exim.
1197 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1198 lets Exim select keys and certificates based upon TLS SNI from client.
1199 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1200 before an outbound SMTP session. New log_selector, +tls_sni.
1202 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1203 NULL dereference. Report and patch from Alun Jones.
1205 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1206 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1207 Not seeing resolver debug output on NetBSD, but suspect this is a
1208 resolver implementation change.
1210 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1211 Left warnings. Added "eximon gdb" invocation mode.
1213 PP/21 Defaulting "accept_8bitmime" to true, not false.
1215 PP/22 Added -bw for inetd wait mode support.
1217 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1218 locate the relevant includes and libraries. Made this the default.
1220 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1221 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1223 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1224 This may cause build issues on older platforms.
1226 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1227 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1228 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1229 Added SNI support via GnuTLS too.
1230 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1232 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1234 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1235 Applied second patch from Janne, implementing suggestion to default
1236 multiple-strings-in-record handling to match SPF spec.
1238 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1240 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1241 read-only, out of scope).
1242 Patch from Wolfgang Breyha, report from Stuart Northfield.
1244 PP/29 Fix three issues highlighted by clang analyser static analysis.
1245 Only crash-plausible issue would require the Cambridge-specific
1246 iplookup router and a misconfiguration.
1247 Report from Marcin Mirosław.
1249 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1251 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1252 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1253 As part of this, removing so much warning spew let me fix some minor
1254 real issues in debug logging.
1256 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1257 assignment on my part. Fixed.
1259 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1260 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1261 Janne Snabb (who went above and beyond: thank you).
1263 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1264 string otherwise requires a connection and a bunch more work and it's
1265 relatively easy to get wrong. Should also expose TLS library linkage
1268 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1269 64-bit ${eval} (JH/03).
1271 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1272 GNU libc to support some of the 64-bit stuff, should not lead to
1273 conflicts. Defined before os.h is pulled in, so if a given platform
1274 needs to override this, it can.
1276 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1277 protection layer was required, which is not implemented.
1278 Bugzilla 1254, patch from Wolfgang Breyha.
1280 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1281 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1282 tls_dhparam take prime identifiers. Also unbreak combination of
1283 OpenSSL+DH_params+TLSSNI.
1285 PP/39 Disable SSLv2 by default in OpenSSL support.
1291 PP/01 Solaris build fix for Oracle's LDAP libraries.
1292 Bugzilla 1109, patch from Stephen Usher.
1294 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1296 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1299 TF/02 Fix a couple more cases where we did not log the error message
1300 when unlink() failed. See also change 4.74-TF/03.
1302 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1303 lock up or crash if it happened to be inside a call to libc when it
1304 got a SIGUSR1 from exiwhat.
1306 The SIGUSR1 handler appends the current process status to the process
1307 log which is later printed by exiwhat. It used to use the general
1308 purpose logging code to do this, but several functions it calls are
1309 not safe for signals.
1311 The new output code in the SIGUSR1 handler is specific to the process
1312 log, and simple enough that it's easy to inspect for signal safety.
1313 Removing some special cases also simplifies the general logging code.
1314 Removing the spurious timestamps from the process log simplifies
1317 TF/04 Improved ratelimit ACL condition.
1319 The /noupdate option has been deprecated in favour of /readonly which
1320 has clearer semantics. The /leaky, /strict, and /readonly update modes
1321 are mutually exclusive. The update mode is no longer included in the
1322 database key; it just determines when the database is updated. (This
1323 means that when you upgrade Exim will forget old rate measurements.)
1325 Exim now checks that the per_* options are used with an update mode that
1326 makes sense for the current ACL. For example, when Exim is processing a
1327 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1328 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1329 must specify per_mail/readonly. If you omit the update mode it defaults to
1330 /leaky where that makes sense (as before) or /readonly where required.
1332 The /noupdate option is now undocumented but still supported for
1333 backwards compatibility. It is equivalent to /readonly except that in
1334 ACLs where /readonly is required you may specify /leaky/noupdate or
1335 /strict/noupdate which are treated the same as /readonly.
1337 A useful new feature is the /count= option. This is a generalization
1338 of the per_byte option, so that you can measure the throughput of other
1339 aggregate values. For example, the per_byte option is now equivalent
1340 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1342 The per_rcpt option has been generalized using the /count= mechanism
1343 (though it's more complicated than the per_byte equivalence). When it is
1344 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1345 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1346 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1347 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1348 Note that using per_rcpt with a non-readonly update mode in more than
1349 one ACL will cause the recipients to be double-counted. (The per_mail
1350 and per_byte options don't have this problem.)
1352 The handling of very low rates has changed slightly. If the computed rate
1353 is less than the event's count (usually one) then this event is the first
1354 after a long gap. In this case the rate is set to the same as this event's
1355 count, so that the first message of a spam run is counted properly.
1357 The major new feature is a mechanism for counting the rate of unique
1358 events. The new per_addr option counts the number of different
1359 recipients that someone has sent messages to in the last time period. It
1360 behaves like per_rcpt if all the recipient addresses are different, but
1361 duplicate recipient addresses do not increase the measured rate. Like
1362 the /count= option this is a general mechanism, so the per_addr option
1363 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1364 example, measure the rate that a client uses different sender addresses
1365 with the options per_mail/unique=$sender_address. There are further
1366 details in the main documentation.
1368 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1370 TF/06 Removed a few PCRE remnants.
1372 TF/07 Automatically extract Exim's version number from tags in the git
1373 repository when doing development or release builds.
1375 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1376 Bugzilla 879. Patch from Paul Fisher.
1378 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1379 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1382 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1384 PP/05 Variable $av_failed, true if the AV scanner deferred.
1385 Bugzilla 1078. Patch from John Horne.
1387 PP/06 Stop make process more reliably on build failure.
1388 Bugzilla 1087. Patch from Heiko Schlittermann.
1390 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1391 Bugzilla 1089. Patch from Heiko Schlittermann.
1393 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1394 Bugzilla 1131. Patch from Holger Weiß.
1396 PP/09 Handle IPv6 addresses with SPF.
1397 Bugzilla 860. Patch from Wolfgang Breyha.
1399 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1401 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1404 PP/11 match_* no longer expand right-hand-side by default.
1405 New compile-time build option, EXPAND_LISTMATCH_RHS.
1406 New expansion conditions, "inlist", "inlisti".
1408 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1410 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1412 PP/14 fix log_write() format string regression from TF/03.
1413 Bugzilla 1152. Patch from Dmitry Isaikin.
1419 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1421 PP/02 Harmonised TLS library version reporting; only show if debugging.
1422 Layout now matches that introduced for other libraries in 4.74 PP/03.
1424 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1426 PP/04 New "dns_use_edns0" global option.
1428 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1431 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1432 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1434 TK/01 Updated PolarSSL code to 0.14.2.
1435 Bugzilla 1097. Patch from Andreas Metzler.
1437 PP/07 Catch divide-by-zero in ${eval:...}.
1438 Fixes bugzilla 1102.
1440 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1443 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1444 format-string attack -- SECURITY: remote arbitrary code execution.
1446 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1447 time unintentionally subject to list matching rules, letting the header
1448 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1449 arbitrary string expansion). This allowed for information disclosure.
1451 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1452 INT_MIN/-1 -- value coerced to INT_MAX.
1458 NM/01 Workaround for PCRE version dependency in version reporting
1461 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1462 This fixes portability to compilers other than gcc, notably
1463 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1465 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1466 makefiles for portability to HP-UX and POSIX correctness.
1468 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1469 Also via indented variable definition in the Makefile.
1470 (Debugging by Oliver Heesakkers).
1472 PP/02 Restore caching of spamd results with expanded spamd_address.
1473 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1475 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1476 Improves build reliability. Fix from: Frank Elsner
1478 NM/02 Fix wide character breakage in the rfc2047 coding
1479 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1481 NM/03 Allow underscore in dnslist lookups
1482 Fixes bug 1026. Patch from Graeme Fowler
1484 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1485 Code patches from Adam Ciarcinski of NetBSD.
1487 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1490 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1491 is logged, to avoid truncation. Patch from John Horne.
1493 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1494 Patch from Jakob Hirsch.
1496 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1497 SQL string expansion failure details.
1498 Patch from Andrey Oktyabrski.
1500 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1501 Patch from Simon Arlott.
1503 PP/09 New lookups functionality failed to compile on old gcc which rejects
1504 extern declarations in function scope.
1505 Patch from Oliver Fleischmann
1507 PP/10 Use sig_atomic_t for flags set from signal handlers.
1508 Check getgroups() return and improve debugging.
1509 Fixed developed for diagnosis in bug 927 (which turned out to be
1512 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1513 Patch from Mark Zealey.
1515 PP/12 Bugzilla 1056: Improved spamd server selection.
1516 Patch from Mark Zealey.
1518 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1519 Based on patch from Heiko Schlittermann.
1521 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1522 Patch from Uwe Doering, sign-off by Michael Haardt.
1524 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1525 variable declaration deep within a block. Bug and patch from
1528 PP/15 lookups-Makefile IRIX compatibility coercion.
1530 PP/16 Make DISABLE_DKIM build knob functional.
1532 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1533 Patch by Simon Arlott
1535 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1536 variable argument macros. Our copy now differs from upstream.
1542 TF/01 Failure to get a lock on a hints database can have serious
1543 consequences so log it to the panic log.
1545 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1546 controlled using the smtp_confirmation log selector.
1548 TF/03 Include the error message when we fail to unlink a spool file.
1550 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1551 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1552 for maintaining out-of-tree patches for some time.
1554 PP/01 Bugzilla 139: Documentation and portability issues.
1555 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1556 Handle per-OS dynamic-module compilation flags.
1558 PP/02 Let /dev/null have normal permissions.
1559 The 4.73 fixes were a little too stringent and complained about the
1560 permissions on /dev/null. Exempt it from some checks.
1561 Reported by Andreas M. Kirchwitz.
1563 PP/03 Report version information for many libraries, including
1564 Exim version information for dynamically loaded libraries. Created
1565 version.h, now support a version extension string for distributors
1566 who patch heavily. Dynamic module ABI change.
1568 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1569 privilege escalation vulnerability whereby the Exim run-time user
1570 can cause root to append content of the attacker's choosing to
1573 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1576 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1577 If dropping privileges for untrusted macros, we disabled normal logging
1578 on the basis that it would fail; for the Exim run-time user, this is not
1579 the case, and it resulted in successful deliveries going unlogged.
1580 Fixed. Reported by Andreas Metzler.
1586 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1587 only prepend for the Resent-* case. Fixes regression introduced in
1588 Exim 4.70 by NM/22 for Bugzilla 607.
1590 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1591 increasing numbers of administrators be bitten by this.
1593 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1595 PP/03 Bugzilla 994: added openssl_options main configuration option.
1597 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1599 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1601 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1603 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1604 without a peer certificate, leading to a segfault because of an
1605 assumption that peers always have certificates. Be a little more
1606 paranoid. Problem reported by Martin Tscholak.
1608 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1609 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1610 NB: ClamAV planning to remove STREAM in "middle of 2010".
1611 CL also introduces -bmalware, various -d+acl logging additions and
1612 more caution in buffer sizes.
1614 PP/09 Implemented reverse_ip expansion operator.
1616 PP/10 Bugzilla 937: provide a "debug" ACL control.
1618 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1620 PP/12 Bugzilla 973: Implement --version.
1622 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1624 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1626 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1628 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1629 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1630 ignore trailing whitespace.
1632 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1634 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1637 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1638 Notification from Dr Andrew Aitchison.
1640 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1641 ExtendedDetectionInfo response format.
1642 Notification from John Horne.
1644 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1647 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1648 XSL and documented dependency on system catalogs, with examples of how
1651 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1654 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1655 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1656 configuration file which is writeable by the Exim user or group.
1658 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1659 of configuration files to cover files specified with the -C option if
1660 they are going to be used with root privileges, not just the default
1663 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1664 option (effectively making it always true).
1666 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1667 files to be used while preserving root privileges.
1669 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1670 that rogue child processes cannot use them.
1672 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1673 run-time user, instead of root.
1675 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1676 Exim run-time user without dropping privileges.
1678 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1679 result string, instead of calling string_vformat() twice with the same
1682 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1683 for other users. Others should always drop root privileges if they use
1684 -C on the command line, even for a whitelisted configure file.
1686 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1688 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1694 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1695 $data_path, and $header_path variables; fixed documentation bugs and
1698 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1699 exipick to access non-standard spools, including the "frozen" queue
1702 NM/01 Bugzilla 965: Support mysql stored procedures.
1703 Patch from Alain Williams
1705 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1707 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1708 Patch from Andreas Metzler
1710 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1711 Patch from Kirill Miazine
1713 NM/05 Bugzilla 671: Added umask to procmail example.
1715 JJ/03 installed exipick 20100323.0, fixing doc bug
1717 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1718 directory. Notification and patch from Dan Rosenberg.
1720 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1722 TK/02 Improve log output when DKIM signing operation fails.
1724 MH/01 Treat the transport option dkim_domain as a colon separated
1725 list, not as a single string, and sign the message with each element,
1726 omitting multiple occurences of the same signer.
1728 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1729 Bugzilla 985, 986. Patch by Simon Arlott
1731 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1732 Patch by Simon Arlott
1734 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1735 MBX locking. Notification from Dan Rosenberg.
1741 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1743 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1745 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1747 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1749 NM/04 Bugzilla 915: Flag broken perl installation during build.
1755 TK/01 Added patch by Johannes Berg that expands the main option
1756 "spamd_address" if it starts with a dollar sign.
1758 TK/02 Write list of recipients to X-Envelope-Sender header when building
1759 the mbox-format spool file for content scanning (suggested by Jakob
1762 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1763 (http://www.dcc-servers.net/) support via dccifd. Activated by
1764 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1766 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1767 by Mark Daniel Reidel <mr@df.eu>.
1769 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1770 When building exim an external PCRE library is now needed -
1771 PCRE is a system library on the majority of modern systems.
1772 See entry on PCRE_LIBS in EDITME file.
1774 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1775 conversation. Added nologin parameter to request.
1776 Patch contributed by Kirill Miazine.
1778 TF/01 Do not log submission mode rewrites if they do not change the address.
1780 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1782 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1783 log files in place. Contributed by Roberto Lima.
1785 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1787 TF/03 Bugzilla 615: When checking the local_parts router precondition
1788 after a local_part_suffix or local_part_prefix option, Exim now
1789 does not use the address's named list lookup cache, since this
1790 contains cached lookups for the whole local part.
1792 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1793 Robert Millan. Documentation is in experimental-spec.txt.
1795 TF/04 Bugzilla 668: Fix parallel build (make -j).
1797 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1799 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1800 Patch provided by Jan Srzednicki.
1802 TF/05 Leading white space used to be stripped from $spam_report which
1803 wrecked the formatting. Now it is preserved.
1805 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1806 that they are available at delivery time.
1808 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1810 TF/08 TLS error reporting now respects the incoming_interface and
1811 incoming_port log selectors.
1813 TF/09 Produce a more useful error message if an SMTP transport's hosts
1814 setting expands to an empty string.
1816 NM/06 Bugzilla 744: EXPN did not work under TLS.
1817 Patch provided by Phil Pennock.
1819 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1820 Patch provided by Richard Godbee.
1822 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1823 acl_smtp_notquit, added index entry.
1825 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1826 Patch provided by Eugene Bujak.
1828 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1829 accept(). Patch provided by Maxim Dounin.
1831 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1832 Patch provided by Phil Pennock.
1834 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1836 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1837 Patch provided by Brad "anomie" Jorsch.
1839 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1840 Patch provided by Dean Brooks.
1842 TK/05 Add native DKIM support (does not depend on external libraries).
1844 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1845 Patch provided by Graeme Fowler.
1847 NM/16 Bugzilla 851: Documentation example syntax fix.
1849 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1851 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1854 NM/19 Bugzilla 745: TLS version reporting.
1855 Patch provided by Phil Pennock.
1857 NM/20 Bugzilla 167: bool: condition support.
1858 Patch provided by Phil Pennock.
1860 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1861 clients. Patch provided by Phil Pennock.
1863 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1864 Patch provided by Brad "anomie" Jorsch.
1866 NM/23 Bugzilla 687: Fix misparses in eximstats.
1867 Patch provided by Heiko Schlittermann.
1869 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1870 Patch provided by Heiko Schlittermann.
1872 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1873 plus update to original patch.
1875 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1877 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1878 Patch provided by David Brownlee.
1880 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1882 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1884 NM/30 Bugzilla 888: TLS documentation bugfixes.
1886 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1888 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1889 Unlike the original bugzilla I have changed all shell scripts in src tree.
1891 NM/33 Bugzilla 898: Transport filter timeout fix.
1892 Patch by Todd Rinaldo.
1894 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1895 Patch by Serge Demonchaux.
1897 NM/35 Bugzilla 39: Base64 decode bug fixes.
1898 Patch by Jakob Hirsch.
1900 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1902 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1904 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1906 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1912 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1913 ALT-N's libdkim that I have put here:
1914 http://duncanthrax.net/exim-experimental/
1916 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1917 were called 'true' and it seems that C99 defines that as a reserved
1918 keyword to be used with 'bool' variable types. That means you could
1919 not include C99-style headers which use bools without triggering
1920 build errors in sieve.c.
1922 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1923 as mailq or other aliases. Changed the --help handling significantly
1924 to do whats expected. exim_usage() emits usage/help information.
1926 SC/01 Added the -bylocaldomain option to eximstats.
1928 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1930 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1932 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1938 PH/01 Another patch from the Sieve maintainer.
1940 PH/02 When an IPv6 address is converted to a string for single-key lookup
1941 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1942 dots are used instead of colons so that keys in lsearch files need not
1943 contain colons. This was done some time before quoting was made available
1944 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1945 (notated using the quote facility) so as to distinguish them from IPv4
1946 keys. This meant that lookups for IP addresses in host lists did not work
1947 for iplsearch lookups.
1949 This has been fixed by arranging for IPv6 addresses to be expressed with
1950 colons if the lookup type is iplsearch. This is not incompatible, because
1951 previously such lookups could never work.
1953 The situation is now rather anomalous, since one *can* have colons in
1954 ordinary lsearch keys. However, making the change in all cases is
1955 incompatible and would probably break a number of configurations.
1957 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1960 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1961 conversion specification without a maximum field width, thereby enabling
1962 a rogue spamd server to cause a buffer overflow. While nobody in their
1963 right mind would setup Exim to query an untrusted spamd server, an
1964 attacker that gains access to a server running spamd could potentially
1965 exploit this vulnerability to run arbitrary code as the Exim user.
1967 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1968 $primary_hostname instead of what libspf2 thinks the hosts name is.
1970 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1971 a directory entry by the name of the lookup key. Previously, if a
1972 symlink pointed to a non-existing file or a file in a directory that
1973 Exim lacked permissions to read, a lookup for a key matching that
1974 symlink would fail. Now it is enough that a matching directory entry
1975 exists, symlink or not. (Bugzilla 503.)
1977 PH/03 The body_linecount and body_zerocount variables are now exported in the
1980 PH/04 Added the $dnslist_matched variable.
1982 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1983 This means they are set thereafter only if the connection becomes
1986 PH/06 Added the client_condition to authenticators so that some can be skipped
1987 by clients under certain conditions.
1989 PH/07 The error message for a badly-placed control=no_multiline_responses left
1990 "_responses" off the end of the name.
1992 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1994 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1995 (without spaces) instead of just copying the configuration text.
1997 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1999 PH/11 Added $max_received_linelength.
2001 PH/12 Added +ignore_defer and +include_defer to host lists.
2003 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2004 way in which PCRE > 7.0 is built.
2006 PH/14 Implemented queue_only_load_latch.
2008 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2009 MAIL command. The effect was to mangle the value on 64-bit systems.
2011 PH/16 Another patch from the Sieve maintainer.
2013 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2015 PH/18 If a system quota error occurred while trying to create the file for
2016 a maildir delivery, the message "Mailbox is full" was not appended to the
2017 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2018 only to a quota excession during the actual writing of the file.
2020 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2021 characters?) which causes problems in log lines. The DN values are now
2022 passed through string_printing() before being added to log lines.
2024 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2025 and InterBase are left for another time.)
2027 PH/21 Added message_body_newlines option.
2029 PH/22 Guard against possible overflow in moan_check_errorcopy().
2031 PH/23 POSIX allows open() to be a macro; guard against that.
2033 PH/24 If the recipient of an error message contained an @ in the local part
2034 (suitably quoted, of course), incorrect values were put in $domain and
2035 $local_part during the evaluation of errors_copy.
2041 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2042 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2045 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2046 issue a MAIL command.
2048 PH/02 In an ACL statement such as
2050 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2052 if a client was not listed at all, or was listed with a value other than
2053 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2054 the condition was not true (as it should be), so access was not denied.
2055 The bug was that the ! inversion was incorrectly passed on to the second
2056 item. This has been fixed.
2058 PH/03 Added additional dnslists conditions == and =& which are different from
2059 = and & when the dns lookup returns more than one IP address.
2061 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2062 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2064 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2065 FSYNC, which compiles an option called disable_fsync that allows for
2066 bypassing fsync(). The documentation is heavily laced with warnings.
2068 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2070 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2071 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2072 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2073 including adding "make clean"; (3) Added -fPIC when compiling the test
2074 dynamically loaded module, to get rid of a warning.
2076 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2077 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2078 = 0s. The bug is otherwise harmless.
2080 PH/07 There was a bug in the dovecot authenticator such that the value of
2081 $auth1 could be overwritten, and so not correctly preserved, after a
2082 successful authentication. This usually meant that the value preserved by
2083 the server_setid option was incorrect.
2085 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2087 PH/09 Installed PCRE release 7.0.
2089 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2090 run for batched SMTP input. It is now run at the start of every message
2091 in the batch. While fixing this I discovered that the process information
2092 (output by running exiwhat) was not always getting set for -bs and -bS
2093 input. This is fixed, and it now also says "batched" for BSMTP.
2095 PH/11 Added control=no_pipelining.
2097 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2098 patch, slightly modified), and move the expansion of helo_data till after
2099 the connection is made in the smtp transport (so it can use these
2102 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2104 PH/14 Added log_selector = +pid.
2106 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2108 PH/16 Add ${if forany and ${if forall.
2110 PH/17 Added dsn_from option to vary the From: line in DSNs.
2112 PH/18 Flush SMTP output before performing a callout, unless control =
2113 no_callout_flush is set.
2115 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2116 was true (the default) a successful delivery failed to delete the retry
2117 item, thus causing premature timeout of the address. The bug is now
2120 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2122 PH/21 Long custom messages for fakedefer and fakereject are now split up
2123 into multiline responses in the same way that messages for "deny" and
2124 other ACL rejections are.
2126 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2127 with slight modification.
2129 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2130 draft, changing the syntax and factoring some duplicate code".
2132 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2133 for deliveries of the second and subsequent messages over the same SMTP
2136 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2137 ${reduce, with only minor "tidies".
2139 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2141 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2142 expansion side effects.
2144 PH/27 When a message times out after an over-quota error from an Exim-imposed
2145 quota, the bounce message says "mailbox is full". This message was not
2146 being given when it was a system quota that was exceeded. It now should
2149 MH/03 Made $recipients available in local_scan(). local_scan() already has
2150 better access to the recipient list through recipients_list[], but
2151 $recipients can be useful in postmaster-provided expansion strings.
2153 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2154 in the case of a MAIL command with additional options following the
2155 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2156 were accidentally chopped off.
2158 PH/29 SMTP synchronization checks are implemented when a command is read -
2159 there is a check that no more input is waiting when there shouldn't be
2160 any. However, for some commands, a delay in an ACL can mean that it is
2161 some time before the response is written. In this time, more input might
2162 arrive, invalidly. So now there are extra checks after an ACL has run for
2163 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2164 pipelining has not been advertised.
2166 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2168 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2169 returned in the SMTP response when smtp_return_error_details was false.
2170 This has been fixed.
2172 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2173 instead of the C I/O that was originally supplied, because problems were
2174 reported on Solaris.
2176 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2177 Exim which did not show up earlier: it was assuming that a call to
2178 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2179 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2180 was a macro that became an assignment, so it seemed to work. This has
2181 changed to a proper function call with a void return, hence the compile
2182 error. Exim's code has been fixed.
2184 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2187 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2189 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2191 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2192 because it turns out to be too big or there is a timeout) there is no
2193 "Completed" line in the log. When some messages of this type were
2194 selected by exigrep, they were listed as "not completed". Others were
2195 picked up by some special patterns. I have improved the selection
2196 criteria to be more general.
2198 PH/38 The host_find_failed option in the manualroute router can now be set
2199 to "ignore", to completely ignore a host whose IP address cannot be
2200 found. If all hosts are ignored, the behaviour is controlled by the new
2201 host_all_ignored option.
2203 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2204 homing or because of multiple MX records with /mx) generated more than
2205 one IP address, and the following item turned out to be the local host,
2206 all the secondary addresses of the first item were incorrectly removed
2207 from the list, along with the local host and any following hosts (which
2208 is what is supposed to happen).
2210 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2211 whoever called Exim into the -H file. In the case of the daemon it was
2212 behaving confusingly. When first started, it used values for whoever
2213 started the daemon, but after a SIGHUP it used the Exim user (because it
2214 calls itself on a restart). I have changed the code so that it now always
2217 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2218 message are rejected with the same error (e.g. no authentication or bad
2219 sender address), and a DATA command is nevertheless sent (as can happen
2220 with PIPELINING or a stupid MUA), the error message that was given to the
2221 RCPT commands is included in the rejection of the DATA command. This is
2222 intended to be helpful for MUAs that show only the final error to their
2225 PH/42 Another patch from the Sieve maintainer.
2227 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2228 Eximstats - Fixed some broken HTML links and added missing column headers
2230 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2231 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2233 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2235 PH/43 Yet another patch from the Sieve maintainer.
2237 PH/44 I found a way to check for a TCP/IP connection going away before sending
2238 the response to the final '.' that terminates a message, but only in the
2239 case where the client has not sent further data following the '.'
2240 (unfortunately, this is allowed). However, in many cases there won't be
2241 any further data because there won't be any more messages to send. A call
2242 to select() can be used: if it shows that the input is "ready", there is
2243 either input waiting, or the socket has been closed. An attempt to read
2244 the next input character can distinguish the two cases. Previously, Exim
2245 would have sent an OK response which the client would never have see.
2246 This could lead to message repetition. This fix should cure that, at
2247 least in a lot of common cases.
2249 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2250 advertised in response to EHLO.
2256 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2257 fixed by 4.65/MH/01 (is this a record?) are fixed:
2259 (i) An empty string was always treated as zero by the numeric comparison
2260 operators. This behaviour has been restored.
2262 (ii) It is documented that the numeric comparison operators always treat
2263 their arguments as decimal numbers. This was broken in that numbers
2264 starting with 0 were being interpreted as octal.
2266 While fixing these problems I realized that there was another issue that
2267 hadn't been noticed. Values of message_size_limit (both the global option
2268 and the transport option) were treated as octal if they started with 0.
2269 The documentation was vague. These values are now always treated as
2270 decimal, and I will make that clear in the documentation.
2276 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2277 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2280 MH/01 Don't check that the operands of numeric comparison operators are
2281 integers when their expansion is in "skipping" mode (fixes bug
2282 introduced by 4.64-PH/07).
2284 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2285 child addresses, Exim now panics and dies. Previously, because the count
2286 is held in a short int, deliveries were likely to be lost. As such a
2287 large number of recipients for a single message is ridiculous
2288 (performance will be very, very poor), I have chosen to impose a limit
2289 rather than extend the field.
2295 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2296 leftover -K file (the existence of which was triggered by #402).
2297 While we were at it, introduced process PID as part of the -K
2298 filename. This should rule out race conditions when creating
2301 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2302 processing considerably. Previous code took too long for large mails,
2303 triggering a timeout which in turn triggers #401.
2305 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2306 in the DK code in transports.c. sendfile() is not really portable,
2307 hence the _LINUX specificness.
2309 TF/01 In the add_headers option to the mail command in an Exim filter,
2310 there was a bug that Exim would claim a syntax error in any
2311 header after the first one which had an odd number of characters
2314 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2315 callout verification, Exim cached a "reject" for the entire domain. This
2316 is correct for most verifications, but it is not correct for a recipient
2317 verification with use_sender or use_postmaster set, because in that case
2318 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2319 case of MAIL FROM:<> rejection from other early rejections (e.g.
2320 rejection of HELO). When verifying a recipient using a non-null MAIL
2321 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2322 Whatever the result of the callout, the value of the domain cache is
2323 left unchanged (for any other kind of callout, getting as far as trying
2324 RCPT means that the domain itself is ok).
2326 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2329 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2330 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2331 session, the daemon ignores ECONNECT errors and logs others; it now
2332 ignores EPIPE as well.
2334 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2335 (quoted-printable decoding).
2337 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2338 later the small subsequent patch to fix an introduced bug.
2340 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2342 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2344 PH/08 An error is now given if message_size_limit is specified negative.
2346 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2347 to be given (somewhat) arbitrary names.
2349 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2352 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2353 miscellaneous code fixes
2355 PH/10 Added the log_reject_target ACL modifier to specify where to log
2358 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2359 hostname. This is wrong, because it relates to the incoming message (and
2360 probably the interface on which it is arriving) and not to the outgoing
2361 callout (which could be using a different interface). This has been
2362 changed to use the value of the helo_data option from the smtp transport
2363 instead - this is what is used when a message is actually being sent. If
2364 there is no remote transport (possible with a router that sets up host
2365 addresses), $smtp_active_hostname is used.
2367 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2368 tweaks were necessary in order to get it to work (see also 21 below):
2369 (a) The code assumed that strncpy() returns a negative number on buffer
2370 overflow, which isn't the case. Replaced with Exim's string_format()
2372 (b) There were several signed/unsigned issues. I just did the minimum
2373 hacking in of casts. There is scope for a larger refactoring.
2374 (c) The code used strcasecmp() which is not a standard C function.
2375 Replaced with Exim's strcmpic() function.
2376 (d) The code set only $1; it now sets $auth1 as well.
2377 (e) A simple test gave the error "authentication client didn't specify
2378 service in request". It would seem that Dovecot has changed its
2379 interface. Fortunately there's a specification; I followed it and
2380 changed what the client sends and it appears to be working now.
2382 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2385 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2386 address A is aliased to B and C, where B exists and C does not. Without
2387 -v the output is "A verified" because verification stops after a
2388 successful redirection if more than one address is generated. However,
2389 with -v the child addresses are also verified. Exim was outputting "A
2390 failed to verify" and then showing the successful verification for C,
2391 with its parentage. It now outputs "B failed to verify", showing B's
2392 parentage before showing the successful verification of C.
2394 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2395 look up a TXT record in a specific list after matching in a combined
2398 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2399 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2400 they consult the DNS. I had assumed they would set it the way they
2401 wanted; and indeed my experiments on Linux seem to show that in some
2402 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2403 To be on the safe side, however, I have now made the interface to
2404 host_find_byname() similar to host_find_bydns(), with an argument
2405 containing the DNS resolver options. The host_find_byname() function now
2406 sets these options at its start, just as host_find_bydns() does. The smtp
2407 transport options dns_qualify_single and dns_search_parents are passed to
2408 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2409 of host_find_byname() use the default settings of RES_DEFNAMES
2410 (qualify_single) but not RES_DNSRCH (search_parents).
2412 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2413 spool_read_header() do less string testing, by means of a preliminary
2414 switch on the second character of optional "-foo" lines. (This is
2415 overdue, caused by the large number of possibilities that now exist.
2416 Originally there were few.) While I was there, I also converted the
2417 str(n)cmp tests so they don't re-test the leading "-" and the first
2418 character, in the hope this might squeeze out yet more improvement.
2420 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2421 flag allowing group syntax was set by the header_syntax check but not
2422 turned off, possible causing trouble later; (2) The flag was not being
2423 set at all for the header_verify test, causing "group"-style headers to
2424 be rejected. I have now set it in this case, and also caused header_
2425 verify to ignore an empty address taken from a group. While doing this, I
2426 came across some other cases where the code for allowing group syntax
2427 while scanning a header line wasn't quite right (mostly, not resetting
2428 the flag correctly in the right place). These bugs could have caused
2429 trouble for malformed header lines. I hope it is now all correct.
2431 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2432 with the "reply" argument non-NULL. The code, however (which originally
2433 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2434 but it didn't always do it. This confused somebody who was copying the
2435 code for some other use. I have removed all the tests.
2437 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2438 feature that was used to support insecure browsers during the U.S. crypto
2439 embargo. It requires special client support, and Exim is probably the
2440 only MTA that supported it -- and would never use it because real RSA is
2441 always available. This code has been removed, because it had the bad
2442 effect of slowing Exim down by computing (never used) parameters for the
2443 RSA_EXPORT functionality.
2445 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2446 authenticator to fail if there's a tab character in the incoming data
2447 (there should never be unless someone is messing about, as it's supposed
2448 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2449 if the connection is using TLS or if the remote IP is the same as the
2450 local IP, and the "valid-client-cert option" if a client certificate has
2453 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2454 authenticators. This can be used for authorization after authentication
2455 succeeds. (In the case of plaintext, it servers for both authentication
2458 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2459 if any retry times were supplied.
2461 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2462 connection, where there is no client IP address to check. In this
2463 situation, the verify now always succeeds.
2465 PH/25 Applied John Jetmore's -Mset patch.
2467 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2469 PH/27 In a string expansion for a processed (not raw) header when multiple
2470 headers of the same name were present, leading whitespace was being
2471 removed from all of them, but trailing whitespace was being removed only
2472 from the last one. Now trailing whitespace is removed from each header
2473 before concatenation. Completely empty headers in a concatenation (as
2474 before) are ignored.
2476 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2477 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2479 PH/29 [Removed. This was a change that I later backed out, and forgot to
2480 correct the ChangeLog entry (that I had efficiently created) before
2481 committing the later change.]
2483 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2484 address errors (4xx response to RCPT) over the same connection as other
2485 messages routed to the same hosts. Such deliveries are always "forced",
2486 so retry times are not inspected. This resulted in far too many retries
2487 for the affected addresses. The effect occurred only when there were more
2488 hosts than the hosts_max_try setting in the smtp transport when it had
2489 the 4xx errors. Those hosts that it had tried were not added to the list
2490 of hosts for which the message was waiting, so if all were tried, there
2491 was no problem. Two fixes have been applied:
2493 (i) If there are any address or message errors in an SMTP delivery, none
2494 of the hosts (tried or untried) are now added to the list of hosts
2495 for which the message is waiting, so the message should not be a
2496 candidate for sending over the same connection that was used for a
2497 successful delivery of some other message. This seems entirely
2498 reasonable: after all the message is NOT "waiting for some host".
2499 This is so "obvious" that I'm not sure why it wasn't done
2500 previously. Hope I haven't missed anything, but it can't do any
2501 harm, as the worst effect is to miss an optimization.
2503 (ii) If, despite (i), such a delivery is accidentally attempted, the
2504 routing retry time is respected, so at least it doesn't keep
2505 hammering the server.
2507 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2508 in ${readsocket because some servers need this prod.
2510 PH/32 Added some extra debug output when updating a wait-xxx database.
2512 PH/33 The hint "could be header name not terminated by colon", which has been
2513 given for certain expansion errors for a long time, was not being given
2514 for the ${if def:h_colon_omitted{... case.
2516 PH/34 The spec says: "With one important exception, whenever a domain list is
2517 being scanned, $domain contains the subject domain." There was at least
2518 one case where this was not true.
2520 PH/35 The error "getsockname() failed: connection reset by peer" was being
2521 written to the panic log as well as the main log, but it isn't really
2522 panic-worthy as it just means the connection died rather early on. I have
2523 removed the panic log writing for the ECONNRESET error when getsockname()
2526 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2527 runs only) independently of the message's sender address. This meant
2528 that, if the 4xx error was in fact related to the sender, a different
2529 message to the same recipient with a different sender could confuse
2530 things. In particular, this can happen when sending to a greylisting
2531 server, but other circumstances could also provoke similar problems.
2532 I have changed the default so that the retry time for these errors is now
2533 based a combination of the sender and recipient addresses. This change
2534 can be overridden by setting address_retry_include_sender=false in the
2537 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2538 remote server are returned as part of bounce messages. This was not
2539 happening for LMTP over a pipe (the lmtp transport), but now it is the
2540 same for both kinds of LMTP.
2542 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2543 in header lines that were in fact CNAMEs. This is no longer the case.
2545 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2546 and queue runs started by the daemon processed all messages. This has
2547 been fixed so that -R and -S can now usefully be given with -q<time>.
2549 PH/40 Import PCRE release 6.7 (fixes some bugs).
2551 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2553 PH/42 Give an error if -q is specified more than once.
2555 PH/43 Renamed the variables $interface_address and $interface_port as
2556 $received_ip_address and $received_port, to make it clear that these
2557 values apply to message reception, and not to the outgoing interface when
2558 a message is delivered. (The old names remain recognized, of course.)
2560 PH/44 There was no timeout on the connect() call when using a Unix domain
2561 socket in the ${readsocket expansion. There now is.
2563 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2564 be meaningful with "accept".
2566 SC/01 Eximstats V1.43
2567 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2569 SC/02 Eximstats V1.44
2570 Use a glob alias rather than an array ref in the generated
2571 parser. This improves both readability and performance.
2573 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2574 Collect SpamAssassin and rejection statistics.
2575 Don't display local sender or destination tables unless
2576 there is data to show.
2577 Added average volumes into the top table text output.
2579 SC/04 Eximstats V1.46
2580 Collect data on the number of addresses (recipients)
2581 as well as the number of messages.
2583 SC/05 Eximstats V1.47
2584 Added 'Message too big' to the list of mail rejection
2585 reasons (thanks to Marco Gaiarin).
2587 SC/06 Eximstats V1.48
2588 Mainlog lines which have GMT offsets and are too short to
2589 have a flag are now skipped.
2591 SC/07 Eximstats V1.49 (Alain Williams)
2592 Added the -emptyok flag.
2594 SC/08 Eximstats V1.50
2595 Fixes for obtaining the IP address from reject messages.
2597 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2598 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2599 whitespace changes from 4.64-PH/27
2601 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2604 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2605 are found, allow negative numbers in numeric criteria)
2607 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2609 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2610 to match changes made in 4.64-PH/43
2612 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2614 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2615 "domain" is actually an IP address, and force a failure. This locks out
2616 those revolvers/nameservers that support "A-for-A" lookups, in
2617 contravention of the specifications.
2619 PH/48 When a host name was looked up from an IP address, and the subsequent
2620 forward lookup of the name timed out, the host name was left in
2621 $sender_host_name, contrary to the specification.
2623 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2624 restricted to single-key lookups, Exim was not diagnosing an error if
2625 * or *@ was used with a query-style lookup.
2627 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2629 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2630 long ago, but noone interested enough thought of it. Let's just say that
2631 the "1.1" means that there are some new functions that weren't there at
2632 some point in the past.
2634 PH/51 Error processing for expansion failure of helo_data from an smtp
2635 transport during callout processing was broken.
2637 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2638 tested/used via the -bh/-bhc/-bs options.
2640 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2641 bug, fixed in subsequent PCRE releases).
2643 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2644 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2646 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2652 SC/01 Use a glob alias rather than an array ref in eximstats generated
2653 parser. This improves both readability and performance.
2655 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2656 Don't display local sender or destination tables in eximstats unless
2657 there is data to show.
2658 Added average volumes into the eximstats top table text output.
2660 SC/03 Collect data on the number of addresses (recipients) as well
2661 as the number of messages in eximstats.
2663 TF/01 Correct an error in the documentation for the redirect router. Exim
2664 does (usually) call initgroups() when daemonizing.
2666 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2667 with consistent privilege compared to when running as a daemon.
2669 TF/03 Note in the spec that $authenticated_id is not set for local
2670 submissions from trusted users.
2672 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2673 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2675 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2676 by adding some example configuration directives to the default
2677 configuration file. A little bit of work is required to uncomment the
2678 directives and define how usernames and passwords are checked, but
2679 there is now a framework to start from.
2681 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2682 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2683 without this. I don't know how relevant this is to other LDAP libraries.
2685 PH/02 Add the verb name to the "unknown ACL verb" error.
2687 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2689 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2691 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2692 directory not expanded when it should be if an expanded home directory
2693 was set for the address (which is overridden by the transport).
2695 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2698 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2699 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2700 because it is too late at that time, and has no effect.
2702 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2703 security issue with \' (bugzilla #107). I could not use the
2704 PQescapeStringConn() function, because it needs a PGconn value as one of
2707 PH/08 When testing addresses using -bt, indicate those final addresses that
2708 are duplicates that would not cause an additional delivery. At least one
2709 person was confused, thinking that -bt output corresponded to deliveries.
2710 (Suppressing duplicates isn't a good idea as you lose the information
2711 about possibly different redirections that led to the duplicates.)
2713 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2714 systems where poll() doesn't work, in particular OS X.
2716 PH/10 Added more information to debugging output for retry time not reached.
2718 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2719 operations in malware.c.
2721 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2724 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2725 syslog_duplication set false, log lines that would normally be written
2726 both the the main log and to the reject log were not written to syslog at
2729 PH/14 In the default configuration, change the use of "message" in ACL warn
2730 statements to "add_header".
2732 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2733 not followed by a command (e.g. "seen endif").
2735 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2736 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2739 PH/17 Added extra conditions to the default value of delay_warning_condition
2743 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2744 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2745 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2748 The Auto-Submitted: and various List- headers are standardised, whereas I
2749 don't think Precedence: ever was.
2751 PH/18 Refactored debugging code in route_finduser() to show more information,
2752 in particular, the error code if getpwnam() issues one.
2754 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2755 This is apparently needed in addition to the PH/07 change above to avoid
2756 any possible encoding problems.
2758 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2759 but not after initializing Perl.
2761 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2762 output them only if debugging. By default they are written stderr,
2763 apparently, which is not desirable.
2765 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2768 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2771 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2773 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2774 authenticated or an ident call has been made. Suppress the default
2775 values for $authenticated_id and $authenticated_sender (but permit -oMai
2776 and -oMas) when testing with -bh.
2778 PH/24 Re-jigged the order of the tests in the default configuration so that the
2779 tests for valid domains and recipients precede the DNS black list and CSA
2780 tests, on the grounds that those ones are more expensive.
2782 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2783 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2784 This bug exists in every version of Exim that I still have, right back to
2787 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2788 However, an attempt to turn on case-sensitivity in a regex key by
2789 including (?-i) didn't work because the subject string was already
2790 lowercased, and the effects were non-intuitive. It turns out that a
2791 one-line patch can be used to allow (?-i) to work as expected.
2797 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2798 other effects) broke the use of negated acl sub-conditions.
2800 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2803 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2804 "Deny" causes Exim to reject the incoming connection with a 554 error.
2805 Unfortunately, if there is a major crisis, such as a disk failure,
2806 tcp-wrappers gives "deny", whereas what one would like would be some
2807 kind of temporary error. A kludge has been added to help with this.
2808 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2809 554 error is used if errno is still zero or contains ENOENT (which occurs
2810 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2813 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2815 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2816 errors. Otherwise a message that provokes a temporary error (when other
2817 messages do not) can cause a whole host to time out.
2819 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2820 addresses were routed directly to files or pipes from a redirect router.
2821 File deliveries just didn't batch; pipe deliveries might have suffered
2824 PH/06 A failure to get a lock for a hints database would erroneously always say
2825 "Failed to get write lock", even when it was really a read lock.
2827 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2828 of 0600. This has been changed to use the value of the lockfile_mode
2829 option (which defaults to 0600).
2831 PH/08 Applied small patch from the Sieve maintainer.
2833 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2834 folder from quota calculations, a direct delivery into this folder messed
2835 up the contents of the maildirsize file. This was because the regex was
2836 used only to exclude .Trash (or whatever) when the size of the mailbox
2837 was calculated. There was no check that a delivery was happening into an
2838 excluded directory. This bug has been fixed by ignoring all quota
2839 processing for deliveries into excluded directories.
2841 PH/10 Added the maildirfolder_create_regex option to appendfile.
2847 PH/01 The code for finding all the local interface addresses on a FreeBSD
2848 system running IPv6 was broken. This may well have applied to all BSD
2849 systems, as well as to others that have similar system calls. The broken
2850 code found IPv4 interfaces correctly, but gave incorrect values for the
2851 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2852 that it would not match correctly against @[] and not recognize the IPv6
2855 PH/02 The ipliteral router was not recognizing addresses of the form user@
2856 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2858 PH/03 Added disable_ipv6.
2860 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2861 decoded form, because it is most often used to construct To: headers
2862 lines in autoreplies, and the decoded form may well be syntactically
2863 invalid. However, $reply_address has leading white space removed, and all
2864 newlines turned into spaces so that the autoreply transport does not
2867 PH/05 If group was specified without a user on a router, and no group or user
2868 was specified on a transport, the group from the router was ignored.
2870 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2871 for visible compile-time settings that can be used to change these
2872 numbers, for those that want even more. Backwards compatibility with old
2873 spool files has been maintained. However, going back to a previous Exim
2874 release will lost any variables that are in spool files.
2876 PH/07 Two small changes when running in the test harness: increase delay when
2877 passing a TCP/IP connection to a new process, in case the original
2878 process has to generate a bounce, and remove special handling of
2879 127.0.0.2 (sic), which is no longer necessary.
2881 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2882 be the same on different OS.
2884 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2887 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2888 whether --show-vars was specified or not
2890 JJ/02 exipick: Added support for new ACL variable spool format introduced
2893 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2894 syntactically invalid From: or Reply-to: line, and a filter used this to
2895 generate an autoreply, and therefore failed to obtain an address for the
2896 autoreply, Exim could try to deliver to a non-existent relative file
2897 name, causing unrelated and misleading errors. What now happens is that
2898 it logs this as a hard delivery error, but does not attempt to create a
2901 PH/11 The exinext utility has a -C option for testing purposes, but although
2902 the given file was scanned by exinext itself; it wasn't being passed on
2903 when Exim was called.
2905 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2906 an end-of-file indication when reading a command response.
2908 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2909 compiled. In many other places in Exim, IPv6 addresses are always
2910 recognized, so I have changed this. It also means that IPv4 domain
2911 literals of the form [IPV4:n.n.n.n] are now always recognized.
2913 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2914 used if the router is not running as root, for example, when verifying at
2915 ACL time, or when using -bh. The debugging output from this situation was
2916 non-existent - all you got was a failure to exec. I have made two
2919 (a) Failures to set uid/gid, the current directory, or a process leader
2920 in a subprocess such as that created by queryprogram now generate
2921 suitable debugging output when -d is set.
2923 (b) The queryprogram router detects when it is not running as root,
2924 outputs suitable debugging information if -d is set, and then runs
2925 the subprocess without attempting to change uid/gid.
2927 PH/15 Minor change to Makefile for building test_host (undocumented testing
2930 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2931 additional section of a DNS packet that returns MX or SRV records.
2932 Instead, it always explicitly searches for A/AAAA records. This avoids
2933 major problems that occur when a DNS server includes only records of one
2934 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2935 fixed another bug: if SRV records were looked up and the corresponding
2936 address records were *not* found in the additional section, the port
2937 values from the SRV records were lost.
2939 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2940 using the correct key (the original address) when searching the retry
2941 rules in order to find which one to use for generating the retry hint.
2943 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2944 adding the default one. Similarly, if it contains a Reply-To: header, the
2945 errors_reply_to option, if set, is not used.
2947 PH/19 When calculating a retry time, Exim used to measure the "time since
2948 failure" by looking at the "first failed" field in the retry record. Now
2949 it does not use this if it is later than than the arrival time of the
2950 message. Instead it uses the arrival time. This makes for better
2951 behaviour in cases where some deliveries succeed, thus re-setting the
2952 "first failed" field. An example is a quota failure for a huge message
2953 when small messages continue to be delivered. Without this change, the
2954 "time since failure" will always be short, possible causing more frequent
2955 delivery attempts for the huge message than are intended.
2956 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2958 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2959 $1, $2, $3) because the numerical variables can be reset during some
2960 expansion items (e.g. "match"), thereby losing the authentication data.
2962 PH/21 Make -bV show the size of off_t variables so that the test suite can
2963 decide whether to run tests for quotas > 2G.
2965 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2966 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2967 filecount value is greater than 2G or if a quota value is greater than 2G
2968 on a system where the size of off_t is not greater than 4, a panic error
2971 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2972 never match. The debug and -bh output now contains an explicit error
2973 message indicating a malformed IPv4 address or mask.
2975 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2976 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2977 PH/23 above applies.
2979 PH/25 Do not write to syslog when running in the test harness. The only
2980 occasion when this arises is a failure to open the main or panic logs
2981 (for which there is an explicit test).
2983 PH/26 Added the /no_tell option to "control=freeze".
2985 PH/27 If a host name lookup failed very early in a connection, for example, if
2986 the IP address matched host_lookup and the reverse lookup yielded a name
2987 that did not have a forward lookup, an error message of the form "no IP
2988 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2989 could be logged. Now it outputs the IP address instead of "NULL".
2991 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2992 allows the sender and the authenticated sender to be set when
2993 submitting a message from within Exim. Since child_open_exim() is
2994 documented for local_scan(), the new function should be too.
2996 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2997 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2998 results in an empty string is now treated as unset.
3000 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3002 PH/31 Added sender_verify_fail logging option.
3004 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3005 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3006 tidied the source and removed it altogether.
3008 PH/33 When a queue run was abandoned because the load average was too high, a
3009 log line was always written; now it is written only if the queue_run log
3010 selector is set. In addition, the log line for abandonment now contains
3011 information about the queue run such as the pid. This is always present
3012 in "start" and "stop" lines but was omitted from the "abandon" line.
3014 PH/34 Omit spaces between a header name and the colon in the error message that
3015 is given when verify = headers_syntax fails (if there are lots of them,
3016 the message gets confusing).
3018 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3019 names, as there are now some PTR records that contain slashes. This check
3020 is only to protect against broken name servers that fall over on strange
3021 characters, so the fact that it applies to all lookups doesn't matter.
3023 PH/36 Now that the new test suite is complete, we can remove some of the
3024 special code in Exim that was needed for the old test suite. For example,
3025 sorting DNS records because real resolvers return them in an arbitrary
3026 order. The new test suite's fake resolver always returns records in the
3029 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3030 bounces) except when queue_only is set, to avoid logging races between
3031 the different processes.
3033 PH/38 Panic-die if .include specifies a non-absolute path.
3035 PH/39 A tweak to the "H" retry rule from its user.
3037 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3038 a label. They prevented compilation on older perls.
3040 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3041 a warning to be raised on newish perls.
3043 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3044 on queue. Changes to match documented behaviour of showing count of
3045 messages matching specified criteria.
3047 PH/40 Changed the default ident timeout from 30s to 5s.
3049 PH/41 Added support for the use of login_cap features, on those BSD systems
3050 that have them, for controlling the resources used by pipe deliveries.
3052 PH/42 The content-scanning code uses fopen() to create files in which to put
3053 message data. Previously it was not paying any attention to the mode of
3054 the files. Exim runs with umask(0) because the rest of the code creates
3055 files with open(), and sets the required mode explicitly. Thus, these
3056 files were ending up world-writeable. This was not a big issue, because,
3057 being within the spool directory, they were not world-accessible. I have
3058 created a function called modefopen, which takes an additional mode
3059 argument. It sets umask(777), creates the file, chmods it to the required
3060 mode, then resets the umask. All the relevant calls to fopen() in the
3061 content scanning code have been changed to use this function.
3063 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3064 to 24 hours. This avoids potential overflow problems when processing G
3065 and H retry rules. I suspect nobody ever tinkers with this value.
3067 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3069 PH/45 When the plaintext authenticator is running as a client, the server's
3070 challenges are checked to ensure they are valid base64 strings. By
3071 default, the authentication attempt is cancelled if an invalid string is
3072 received. Setting client_ignore_invalid_base64 true ignores these errors.
3073 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3074 they are received. Thus, the responses can be made to depend on the
3075 challenges. If an invalid string is ignored, an empty string is placed in
3078 PH/46 Messages that are created by the autoreply transport now contains a
3079 References: header, in accordance with RFCs 2822 and 3834.
3081 PH/47 Added authenticated_sender_force to the smtp transport.
3083 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3085 PH/49 Installed latest patch from the Sieve maintainer.
3087 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3088 was also set, the appendfile transport was unnecessarily scanning a
3089 directory of message files (e.g. for maildir delivery) to find the count
3090 of files (along with the size), even though it did not need this
3091 information. It now does the scan only if it needs to find either the
3092 size of the count of files.
3094 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3096 PH/52 Two bugs concerned with error handling when the smtp transport is
3099 (i) Exim was not creating retry information for temporary errors given
3100 for individual recipients after the DATA command when the smtp transport
3101 was used in LMTP mode. This meant that they could be retried too
3102 frequently, and not timed out correctly.
3104 (ii) Exim was setting the flag that allows error details to be returned
3105 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3106 recipients that were returned after the DATA command.
3108 PH/53 This is related to PH/52, but is more general: for any failing address,
3109 when detailed error information was permitted to be returned to the
3110 sender, but the error was temporary, then after the final timeout, only
3111 "retry timeout exceeded" was returned. Now it returns the full error as
3112 well as "retry timeout exceeded".
3114 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3115 do this, and (what is worse) MTAs that accept it.
3117 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3118 will now be deprecated.
3120 PH/56 New os.c-cygwin from the Cygwin maintainer.
3122 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3123 formats (previously only available in exim formats via -bpr, -bpru,
3124 and -bpra. Now also available in native and exiqgrep formats)
3126 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3127 with very large, slow to parse queues
3129 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3131 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3133 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3134 responses to authentication challenges, though it was showing the
3135 challenges; (ii) I've removed the CR characters from the debug output for
3138 PH/58 Allow for the insertion of a newline as well as a space when a string
3139 is turned into more than one encoded-word during RFC 2047 encoding. The
3140 Sieve code now uses this.
3142 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3143 data_4xx, lost_connection, tls_required.
3145 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3146 message was being sent as an SMTP response.
3148 PH/61 Add -l and -k options to exicyclog.
3150 PH/62 When verifying, if an address was redirected to one new address, so that
3151 verification continued, and the new address failed or deferred after
3152 having set something in $address_data, the value of $address_data was not
3153 passed back to the ACL. This was different to the case when no
3154 redirection occurred. The value is now passed back in both cases.
3156 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3157 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3158 use login_cap.h, so on its own it isn't the distinguishing feature. The
3159 new name refers directly to the setclassresources() function.
3161 PH/65 Added configuration files for NetBSD3.
3163 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3165 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3166 is preferred over IPv4.
3168 PH/68 The bounce_return_message and bounce_return_body options were not being
3169 honoured for bounces generated during the reception of non-SMTP messages.
3170 In particular, this applied to messages rejected by the ACL. This bug has
3171 been fixed. However, if bounce_return_message is true and bounce_return_
3172 body is false, the headers that are returned for a non-SMTP message
3173 include only those that have been read before the error was detected.
3174 (In the case of an ACL rejection, they have all been read.)
3176 PH/69 The HTML version of the specification is now built in a directory called
3177 spec_html instead of spec.html, because the latter looks like a path with
3178 a MIME-type, and this confuses some software.
3180 PH/70 Catch two compiler warnings in sieve.c.
3182 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3183 function verify_get_ident() calls ip_connect() to connect a socket, but
3184 if the "connect()" function timed out, ip_connect() used to close the
3185 socket. However, verify_get_ident() also closes the socket later, and in
3186 between Exim writes to the log, which may get opened at this point. When
3187 the socket was closed in ip_connect(), the log could get the same file
3188 descriptor number as the socket. This naturally causes chaos. The fix is
3189 not to close the socket in ip_connect(); the socket should be closed by
3190 the function that creates it. There was only one place in the code where
3191 this was missing, in the iplookup router, which I don't think anybody now
3192 uses, but I've fixed it anyway.
3194 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3195 well as to direct DNS lookups. Otherwise the handling of names in host
3196 lists is inconsistent and therefore confusing.
3202 PH/01 Two changes to the default runtime configuration:
3204 (1) Move the checks for relay_from_hosts and authenticated clients from
3205 after to before the (commented out) DNS black list checks.
3207 (2) Add control=submission to the relay_from_hosts and authenticated
3208 clients checks, on the grounds that messages accepted by these
3209 statements are most likely to be submissions.
3211 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3213 (1) Generate an error if the third argument for the ${prvs expansion is
3216 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3219 (3) Reset the variables that are obtained from the first argument of
3220 ${prvscheck and used in the second argument before leaving the code,
3221 because their memory is reclaimed, so using them afterwards may do
3224 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3225 one (it's much easier than Tom thought :-).
3227 (5) Because of (4), we can now allow for the use of $prvscheck_result
3228 inside the third argument.
3230 PH/03 For some reason, the default setting of PATH when running a command from
3231 a pipe transport was just "/usr/bin". I have changed it to
3234 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3235 anything to be listed in the output from -bV.
3237 PH/05 When a filter generated an autoreply, the entire To: header line was
3238 quoted in the delivery log line, like this:
3240 => >A.N.Other <ano@some.domain> <original@ddress> ...
3242 This has been changed so that it extracts the operative address. There
3243 may be more than one such address. If so, they are comma-separated, like
3246 => >ano@some.domain,ona@other.domain <original@ddress> ...
3248 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3249 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3250 not being looked up in the rDNS to get a host name, Exim was showing the
3251 IP address twice in Received: lines, even though the IP addresses were
3252 identical. For example:
3254 Received: from [1.2.3.4] (helo=[1.2.3.4])
3256 However, if the real host name was known, it was omitting the HELO data
3257 if it matched the actual IP address. This has been tidied up so that it
3258 doesn't show the same IP address twice.
3260 PH/07 When both +timestamp and +memory debugging was on, the value given by
3261 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3262 called by the debug printing, thereby overwriting the timestamp buffer.
3263 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3265 PH/08 When the original message was included in an autoreply transport, it
3266 always said "this is a copy of the message, including all the headers",
3267 even if body_only or headers_only was set. It now gives an appropriate
3270 PH/09 Applied a patch from the Sieve maintainer which:
3272 o fixes some comments
3273 o adds the (disabled) notify extension core
3274 o adds some debug output for the result of if/elsif tests
3275 o points to the current vacation draft in the documentation
3276 and documents the missing references header update
3280 o fixes a bug in processing the envelope test (when testing
3281 multiple envelope elements, the last element determined the
3284 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3285 Electronic Mail") by including:
3287 Auto-submitted: auto-generated
3289 in the messages that it generates (bounce messages and others, such as
3290 warnings). In the case of bounce messages for non-SMTP messages, there was
3291 also a typo: it was using "Auto_submitted" (underscore instead of
3292 hyphen). Since every message generated by Exim is necessarily in response
3293 to another message, thes have all been changed to:
3295 Auto-Submitted: auto-replied
3297 in accordance with these statements in the RFC:
3299 The auto-replied keyword:
3301 - SHOULD be used on messages sent in direct response to another
3302 message by an automatic process,
3304 - MUST NOT be used on manually-generated messages,
3306 - MAY be used on Delivery Status Notifications (DSNs) and Message
3307 Disposition Notifications (MDNs),
3309 - MUST NOT be used on messages generated by automatic or periodic
3310 processes, except for messages which are automatic responses to
3313 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3314 to the default Received: header definition.
3316 PH/12 Added log selector acl_warn_skipped (default on).
3318 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3319 variables because (a) they are in the wrong storage pool and (b) even if
3320 they were copied, it wouldn't work properly because of the caching.
3322 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3323 checking when decoding. Apparently there are clients that generate
3324 overlong encoded strings. Why am I not surprised?
3326 PH/15 If the first argument of "${if match_address" was not empty, but did not
3327 contain an "@" character, Exim crashed. Now it writes a panic log message
3328 and treats the condition as false.
3330 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3332 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3333 extension "envelope-auth". The code is finished and in agreement with
3334 other implementations, but there is no documentation so far and in fact,
3335 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3336 not changing the active code.
3338 Print executed "if" and "elsif" statements when debugging is used. This
3339 helps a great deal to understand what a filter does.
3341 Document more things not specified clearly in RFC3028. I had all this
3342 sorted out, when out of a sudden new issues came to my mind. Oops."
3344 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3347 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3348 canonical form (as documented). However, after a host name lookup from
3349 the IP address, check_host() was doing a simple string comparison with
3350 addresses acquired from the DNS when checking that the found name did
3351 have the original IP as one of its addresses. Since any found IPv6
3352 addresses are likely to be in abbreviated form, the comparison could
3353 fail. Luckily, there already exists a function for doing the comparison
3354 by converting both addresses to binary, so now that is used instead of
3355 the text comparison.
3357 PH/20 There was another similar case to PH/19, when a complete host name was
3358 given in a host list; looking up its IP address could give an abbreviated
3359 form, whereas the current host's name might or might not be abbreviated.
3360 The same fix has been applied.
3366 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3367 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3370 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3371 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3373 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3375 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3376 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3377 Also, for "Auto-Submitted", treat anything other than "no" as
3378 non-personal, in accordance with RFC 3834. (Previously it treated
3379 anything starting "auto-" as non-personal.)
3381 TF/01 The control=submission/name=... option had a problem with syntax
3382 errors if the name included a slash character. The /name= option
3383 now slurps the rest of the string, so it can include any characters
3384 but it must come last in the list of options (after /sender_retain
3387 PH/05 Some modifications to the interface to the fake nameserver for the new
3395 TK/01 Added the "success_on_redirect" address verification option. See
3396 NewStuff for rationale and an example.
3398 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3400 PH/02 Patch to exigrep to allow it to work on syslog lines.
3402 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3403 fread() to skip over the body file's header line, because in Cygwin the
3404 header line is locked and is inaccessible.
3406 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3407 co-exist for some time) to make it clear that it is the Exim ID that is
3408 referenced, not the Message-ID: header line.
3410 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3411 string_format() function, because snprintf() does not exist on all
3414 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3415 ${stat: expansion item.
3417 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3418 protocol synchronization error", to keep the pedants happy.
3420 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3421 well as for IRIX systems, when gcc is being used. See the host.c source
3424 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3426 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3429 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3430 transport and to the smtp transport in LMTP mode.
3432 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3434 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3435 run a filter in a subprocess. This could lead to confusion in subsequent
3436 lookups in the parent process. There should also be a search_tidyup() at
3437 the end of the subprocess.
3439 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3440 only if the host matched helo_try_verify_hosts, which caused the
3441 verification to occur when the EHLO/HELO command was issued. The ACL just
3442 tested the remembered result. Now, if a previous verification attempt has
3443 not happened, "verify = helo" does it there and then.
3445 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3447 TK/03 Fix log output including CR from clamd.
3449 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3450 exist provoked a memory error which could cause a segfault.
3452 PH/15 Installed PCRE 6.2
3454 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3456 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3457 of the problem. Specifically, suggested +O2 rather than +O1 for the
3460 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3462 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3463 record, the port was not correctly logged when the outgoing_port log
3464 selector was set (it logged the transort's default port).
3466 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3467 fallback_hosts, and "hosts" in the smtp transport.
3469 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3470 host errors such as "Connection refused".
3472 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3473 authentication with radiusclient 0.4.9:
3475 - Error returned from rc_read_config was caught wrongly
3476 - Username/password not passed on to radius server due to wrong length.
3478 The presumption is that some radiusclient API changes for 4.51/PH/17
3479 were not taken care of correctly. The code is still untested by me (my
3480 Linux distribution still has 0.3.2 of radiusclient), but it was
3481 contributed by a Radius user.
3483 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3484 expanding the "port" option of the smtp transport.
3486 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3487 while reading a MIME header. Thanks to Tom Hughes for a patch.
3489 PH/24 Include config.h inside local_scan.h so that configuration settings are
3492 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3493 that in an ACL for RCPT (for example), you can examine exactly what was
3496 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3497 commands, but it was not correctly comparing the address with the actual
3498 client host address. Thus, it would show the EHLO address in Received:
3499 header lines when this was not necessary.
3501 PH/27 Added the % operator to ${eval:}.
3503 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3504 it should be ignoring failures (because with -C, for example, it has lost
3505 privilege). It wasn't ignoring creation failures other than "already
3508 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3511 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3512 because an input line was too long, either on its own, or by virtue of
3513 too many continuations, the temporary file was not being removed, and the
3514 return code was incorrect.
3516 PH/31 Missing "BOOL" in function definition in filtertest.c.
3518 PH/32 Applied Sieve patches from the maintainer.
3520 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3522 PH/33 Added "verify = not_blind".
3524 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3525 Local/Makefile (with some defaults set). These are used in built scripts
3526 such as exicyclog, but they have never been used in the exim_install
3527 script (though there are many overriding facilities there). I have
3528 arranged that the exim_install script now takes note of these two
3531 PH/35 Installed configuration files for Dragonfly.
3533 PH/36 When a locally submitted message by a trusted user did not contain a
3534 From: header, and the sender address was obtained from -f or from an SMTP
3535 MAIL command, and the trusted user did not use -F to supply a sender
3536 name, $originator_name was incorrectly used when constructing a From:
3537 header. Furthermore, $originator_name was used for submission mode
3538 messages from external hosts without From: headers in a similar way,
3539 which is clearly wrong.
3541 PH/37 Added control=suppress_local_fixups.
3543 PH/38 When log_selector = +received_sender was set, and the addition of the
3544 sender made the log line's construction buffer exactly full, or one byte
3545 less than full, an overflow happened when the terminating "\n" was
3548 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3549 when the result of a list match is failure because a DNS lookup failed.
3551 PH/40 RM_COMMAND is now used in the building process.
3553 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3554 the "build-* directories that it finds.
3556 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3557 address in a domain literal was a prefix of an interface address.
3559 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3560 when verifying a sender address, unless rewrite_headers is false.
3562 PH/44 Wrote a long comment about why errors_to addresses are verified as
3563 recipients, not senders.
3565 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3566 the ratelimit ACL was added.
3568 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3570 PH/46 Added extra information about PostgreSQL errors to the error string.
3572 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3573 avoiding the need to install special zones in a real server. This is
3574 backwards compatible; if it can't find the fake resolver, it drops back.
3575 Thus, both old and new test suites can be run.
3577 TF/02 Added util/ratelimit.pl
3579 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3580 clock is set back in time.
3582 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3583 Candler <B.Candler@pobox.com>.
3585 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3586 OK for addresses that are the result of redirections.
3588 PH/48 A number of further additions for the benefit of the new test suite,
3589 including a fake gethostbyname() that interfaces to the fake DNS resolver
3592 TF/06 The fix for widen_domains has also been applied to qualify_single and
3593 search_parents which are the other dnslookup options that can cause
3596 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3599 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3601 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3602 service" if it encountered an MX record with an empty target hostname.
3603 The message is now "an MX or SRV record indicated no SMTP service".
3605 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3606 if the DNS of the sending site is misconfigured. This is quite a
3607 common situation. This change restores the behaviour of treating a
3608 helo verification defer as a failure.
3610 PH/51 If self=fail was set on a router, the bounce message did not include the
3611 actual error message.
3617 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3619 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3620 command itself ended in error, the underlying message about the transport
3621 filter timeout was being overwritten with the pipe command error. Now the
3622 underlying error message should be appended to the second error message.
3624 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3626 PH/02 Reduce the amount of output that "make" produces by default. Full output
3627 can still be requested.
3629 PH/03 The warning log line about a condition test deferring for a "warn" verb
3630 was being output only once per connection, rather than after each
3631 occurrence (because it was using the same function as for successful
3632 "warn" verbs). This seems wrong, so I have changed it.
3634 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3635 it should not have, which might have caused a crash in the right
3636 circumstances, but probably never did.
3638 PH/04 Installed a modified version of Tony Finch's patch to make submission
3639 mode fix the return path as well as the Sender: header line, and to
3640 add a /name= option so that you can make the user's friendly name appear
3643 TF/03 Added the control = fakedefer ACL modifier.
3645 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3646 Mark Lowes for thorough testing.
3648 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3650 TK/03 Merged latest SRS patch from Miles Wilton.
3652 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3653 with the definition in sysexits.h (which is #included earlier).
3654 Fortunately, Exim does not actually use EX_OK. The code used to try to
3655 preserve the sysexits.h value, by assuming that macro definitions were
3656 scanned for macro replacements. I have been disabused of this notion,
3657 so now the code just undefines EX_OK before #including unistd.h.
3659 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3660 in the smtp transport. When a block could not be written in a single
3661 write() function, the timeout was being re-applied to each part-write.
3662 This seems wrong - if the receiver was accepting one byte at a time it
3663 would take for ever. The timeout is now adjusted when this happens. It
3664 doesn't have to be particularly precise.
3666 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3667 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3669 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3670 without a domain if the check to <postmaster@domain> fails.
3672 SC/01 Eximstats: added -xls and the ability to specify output files
3673 (patch written by Frank Heydlauf).
3675 SC/02 Eximstats: use FileHandles for outputting results.
3677 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3679 SC/04 Eximstats: fixed display of large numbers with -nvr option
3681 SC/05 Eximstats: fixed merging of reports with empty tables.
3683 SC/06 Eximstats: added the -include_original_destination flag
3685 SC/07 Eximstats: removed tabs and trailing whitespace.
3687 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3689 TK/06 MBOX spool code: Add real "From " MBOX separator line
3690 so the .eml file is really in mbox format (even though
3691 most programs do not really care). Patch from Alex Miller.
3693 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3694 The latter is generated from $received_to and is only set if the
3695 message has one envelope recipient. SA can use these headers,
3696 obviously out-of-the-box. Patch from Alex Miller.
3698 PH/08 The ${def test on a variable was returning false if the variable's
3699 value was "0", contrary to what the specification has always said!
3700 The result should be true unless the variable is empty.
3702 PH/09 The syntax error of a character other than { following "${if
3703 def:variable_name" (after optional whitespace) was not being diagnosed.
3704 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3705 accidental colon was present, for example, could give incorrect results.
3707 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3708 result is used (not including appendfile, where other changes are about
3711 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3712 This involved changing a lot of size variables from int to off_t. It
3713 should work with maildirs and everything.
3715 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3716 spamd dying while we are connected to it.
3718 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3721 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3722 function for BDB 4.3.
3724 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3726 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3727 the sources for PCRE have been re-arranged and more files are now
3730 PH/15 The code I had for printing potentially long long variables in PH/11
3731 above was not the best (it lost precision). The length of off_t variables
3732 is now inspected at build time, and an appropriate printing format (%ld
3733 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3734 to be "long long int" or "long int". This is needed for the internal
3735 formatting function string_vformat().
3737 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3738 the configuration file to be ":syslog", then the script "guesses" where
3739 the logs files are, rather than using the compiled in default. In our
3740 case the guess is not the same as the compiled default, so the script
3741 suddenly stopped working when I started to use syslog. The patch checks
3742 to see if log_file_path is "". If so, it attempts to read it from exim
3743 with no configuration file to get the compiled in version, before it
3744 falls back to the previous guessing code."
3746 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3747 implementing BATV in an Exim configuration. See NewStuff for the gory
3750 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3751 Makefile that are specific to HP-UX.
3753 PH/18 If the "use_postmaster" option was set for a recipient callout together
3754 with the "random" option, the postmaster address was used as the MAIL
3755 FROM address for the random test, but not for the subsequent recipient
3756 test. It is now used for both.
3758 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3759 patch removes a few documentation additions to RFC 3028, because the
3760 latest draft now contains them. It adds the new en;ascii-case comparator
3761 and a new error check for 8bit text in MIME parts. Comparator and
3762 require names are now matched exactly. I enabled the subaddress
3763 extension, but it is not well tested yet (read: it works for me)."
3765 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3766 rework some of the code of TK/09 above to avoid the hardwired use of
3767 "%lld" and "long long". Replaced the call to snprintf() with a call to
3770 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3771 records point to non-existent hosts", "retry timeout exceeded", and
3772 "retry time not reached for any host after a long failure period".
3774 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3775 experimental DomainKeys support:
3777 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3778 (2) On an error such as an illegally used "control", the wrong name for
3779 the control was given.
3781 These problems did NOT occur unless DomainKeys support was compiled.
3783 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3785 PH/24 Added ${if match_ip condition.
3787 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3788 databases so that it will be absolutely obvious if a crash occurs in the
3789 DB library. This is a regular occurrence (often caused by mis-matched
3792 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3793 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3794 picked up on a user's system that detects such things. There doesn't seem
3795 to be a gcc warning option for this - only an attribute that has to be
3796 put on the function's prototype. It seems that in Fedora Core 4 they have
3797 set this on a number of new functions. No doubt there will be more in due
3800 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3801 specify a transport. However, if an address that was verified by such a
3802 router was the subject of a callout, Exim crashed because it tried to
3803 read the rcpt_include_affixes from the non-existent transport. Now it
3804 just assumes that the setting of that option is false. This bug was
3805 introduced by 4.51/PH/31.
3807 PH/28 Changed -d+all to exclude +memory, because that information is very
3808 rarely of interest, but it makes the output a lot bigger. People tend to
3809 do -d+all out of habit.
3811 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3812 code in os-type was giving problems when libc.so lives in lib64, like on
3815 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3816 aren't the modern standard, and it seems that some systems' include files
3817 don't always have them. Exim was already checking for some of the newer
3818 ones like T_AAAA, and defining it itself. I've added checks for all the
3819 record types that Exim uses.
3821 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3822 not automatically generating a new one, as it is supposed to. This
3823 prevented TLS from working. If the file did exist, but contained invalid
3824 data, a new version was generated, as expected. It was only the case of a
3825 non-existent file that was broken.
3827 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3828 with a change in libdomainkeys > 0.64.
3830 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3831 from DNS. If the selector record carries the flag, it now has
3832 precedence over the domain-wide flag.
3834 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3836 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3837 the use of a port name that isn't defined in /etc/services) occurred, the
3838 message was deferred as in a normal delivery, and thus remained on the
3839 spool, instead of being failed because of the mua_wrapper setting. This
3840 is now fixed, and I tidied up some of the mua_wrapper messages at the
3843 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3844 the messages in a hash of arrays rather than using individual hashes.
3845 This is a bit cleaner and results in dramatic memory savings, albeit
3846 at a slight CPU cost.
3848 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3849 as requested by Marc Sherman.
3851 SC/10 Eximstats: added histograms for user specified patterns as requested
3854 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3856 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3857 fopen() in the content-scanning modules that did not already have it.
3863 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3864 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3866 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3868 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3870 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3871 to test Sieve filters that use "vacation".
3873 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3874 that changes the way the GnuTLS parameters are stored in the cache file.
3875 The new format can be generated externally. For backward compatibility,
3876 if the data in the cache doesn't make sense, Exim assumes it has read an
3877 old-format file, and it generates new data and writes a new file. This
3878 means that you can't go back to an older release without removing the
3881 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3882 work if there are any delivery delays because "one_time" forces the
3883 parent to be marked "delivered", so its unseen clone is never tried
3884 again. For this reason, Exim now forbids the simultaneous setting of
3887 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3888 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3889 entry if you want to know the details. The fix, however, neglected to
3890 consider the case when local delivery batching is involved. The test for
3891 "previously delivered" was not happening when checking to see if an
3892 address could be batched with a previous (undelivered) one; under
3893 certain circumstances this could lead to multiple deliveries to the same
3896 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3897 in its include files, and this causes problems building Exim.
3899 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3900 header_syntax) but Exim was just ignoring anything given after a slash.
3901 In particular, this caused confusion with an attempt to use "verify =
3902 reverse_host_lookup/defer_ok". An error is now given when options are
3903 supplied for verify items that do not have them. (Maybe reverse_host_
3904 lookup should have a defer_ok option, but that's a different point.)
3906 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3907 defined by RFC 821) to 2048, because there were problems with some AUTH
3908 commands, and RFC 1869 says the size should be increased for extended
3909 SMTP commands that take arguments.
3911 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3914 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3915 "unknown" error; now it says that the functionality isn't in the binary.
3917 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3918 an address' error message when a string expansion fails (syntax or
3919 whatever). Otherwise the password may appear in the log. Following change
3920 PH/42 below, there is no longer a chance of it appearing in a bounce
3923 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3925 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3926 of its addresses was ever tried. (Bugzilla bug #2.)
3928 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3929 the result incorrectly in the debug output. (It correctly added a newline
3930 to what was transported.)
3932 TF/01 Added $received_time.
3934 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3935 commented out examples of how to interface to a virus scanner and to
3936 SpamAssassin. Also added commented examples of av_scanner and
3937 spamd_address settings.
3939 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3940 and controls are allowed in which ACLs. There were a couple of minor
3941 errors. Some of the entries in the conditions table (which is a table of
3942 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3943 negation of where the condition IS allowed.
3945 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3947 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3948 header file does not have a version number, so I've had to invent a new
3949 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3950 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3951 radiusclient), but it was contributed by a Radius user.
3953 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3954 files or directories, for OpenSSL.
3956 PH/19 When an Exim process that is running as root has to create an Exim log
3957 file, it does so in a subprocess that runs as exim:exim so as to get the
3958 ownership right at creation (otherwise, other Exim processes might see
3959 the file with the wrong ownership). There was no test for failure of this
3960 fork() call, which would lead to the process getting stuck as it waited
3961 for a non-existent subprocess. Forks do occasionally fail when resources
3962 run out. I reviewed all the other calls to fork(); they all seem to check
3965 PH/20 When checking for unexpected SMTP input at connect time (before writing
3966 the banner), Exim was not dealing correctly with a non-positive return
3967 from the read() function. If the client had disconnected by this time,
3968 the result was a log entry for a synchronization error with an empty
3969 string after "input=" when read() returned zero. If read() returned -1
3970 (an event I could not check), uninitialized data bytes were printed.
3971 There were reports of junk text (parts of files, etc) appearing after
3974 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3976 PH/22 Added support for macro redefinition, and (re)definition in between
3977 driver and ACL definitions.
3979 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3980 forgetting to use the resulting value; it was using the unexpanded value.
3982 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3983 hadn't been configured. The fix is from Juergen Kreileder, who
3984 understands it better than I do:
3986 "Here's what I see happening with three configured cyrus_sasl
3987 authenticators configured (plain, login, cram-md5):
3989 On startup auth_cyrus_sasl_init() gets called for each of these.
3990 This means three calls to sasl_listmech() without a specified mech_list.
3991 => SASL tests which mechs of all available mechs actually work
3992 => three warnings about OTP not working
3993 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3995 With the patch, sasl_listmech() also gets called three times. But now
3996 SASL's mech_list option is set to the server_mech specified in the the
3997 authenticator. Or in other words, the answer from sasl_listmech()
3998 gets limited to just the mech you're testing for (which is different
4000 => the return list contains just 'plain' or 'login', 'cram-md5' or
4001 nothing depending on the value of ob->server_mech.
4003 I've just tested the patch: Authentication still works fine,
4004 unavailable mechs specified in the exim configuration are still
4005 caught, and the auth.log warnings about OTP are gone."
4007 PH/25 When debugging is enabled, the contents of the command line are added
4008 to the debugging output, even when log_selector=+arguments is not
4011 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4012 answer is "GNU", and only if the return is "GNU/something" is the answer
4015 PH/27 $acl_verify_message is now set immediately after the failure of a
4016 verification in an ACL, and so is available in subsequent modifiers. In
4017 particular, the message can be preserved by coding like this:
4019 warn !verify = sender
4020 set acl_m0 = $acl_verify_message
4022 Previously, $acl_verify_message was set only while expanding "message"
4023 and "log_message" when a very denied access.
4025 PH/28 Modified OS/os.c-Linux with
4027 -#ifndef OS_LOAD_AVERAGE
4028 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4030 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4031 nomenclature these days.)
4033 PH/29 Installed patch from the Sieve maintainer that adds the options
4034 sieve_useraddress and sieve_subaddress to the redirect router.
4036 PH/30 In these circumstances:
4037 . Two addresses routed to the same list of hosts;
4038 . First host does not offer TLS;
4039 . First host accepts first address;
4040 . First host gives temporary error to second address;
4041 . Second host offers TLS and a TLS session is established;
4042 . Second host accepts second address.
4043 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4044 and peerdn, if requested) that were in fact used only for the second
4047 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4048 attention to any local part prefix or suffix that was matched by the
4049 router that accepted the address. It now behaves in the same way as it
4050 does for delivery: the affixes are removed from the local part unless
4051 rcpt_include_affixes is set on the transport.
4053 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4054 timeout during the DATA phase of an incoming message.
4056 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4057 applied a patch sanctioned by the Sieve maintainer.
4059 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4060 the uid or gid is negative. A case of a negative gid caused this to be
4061 noticed. The fix allows for either to be negative.
4063 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4064 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4067 PH/36 The change PH/12 above was broken. Fixed it.
4069 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4070 the grounds that routing the same address twice would always produce the
4071 same answer. This might have been true once, but it is certainly no
4072 longer true now. Routing a child address may depend on the previous
4073 routing that produced that child. Some complicated redirection strategies
4074 went wrong when messages had multiple recipients, and made Exim's
4075 behaviour dependent on the order in which the addresses were given.
4077 I have moved the duplicate checking until after the routing is complete.
4078 Exim scans the addresses that are assigned to local and remote
4079 transports, and removes any duplicates. This means that more work will be
4080 done, as duplicates will always all be routed, but duplicates are
4081 presumably rare, so I don't expect this is of any significance.
4083 For deliveries to pipes, files, and autoreplies, the duplicate checking
4084 still happens during the routing process, since they are not going to be
4087 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4088 It corrects a timeout issue with spamd. This is Ian's comment: "The
4089 background is that sometimes spamd either never reads data from a
4090 connection it has accepted, or it never writes response data. The exiscan
4091 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4092 blindly assumes that writes won't block so it may never time out."
4094 PH/39 Allow G after quota size as well as K and M.
4096 PH/40 The value set for $authenticated_id in an authenticator may not contain
4097 binary zeroes or newlines because the value is written to log lines and
4098 to spool files. There was no check on this. Now the value is run through
4099 the string_printing() function so that such characters are converted to
4100 printable escape sequences.
4102 PH/41 $message_linecount is a new variable that contains the total number of
4103 lines in the message. Compare $body_linecount, which is the count for the
4106 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4107 bounce and delay warning messages, except in certain special cases, which
4110 (a) An SMTP error message from a remote host;
4111 (b) A message specified in a :fail: redirection;
4112 (c) A message specified in a "fail" command in a system filter;
4113 (d) A message specified in a FAIL return from the queryprogram router;
4114 (e) A message specified by the cannot_route_message router option.
4116 In these cases only, Exim does include the error details in bounce and
4117 warning messages. There are also a few cases where bland messages such
4118 as "unrouteable address" or "local delivery error" are given.
4120 PH/43 $value is now also set for the "else" part of a ${run expansion.
4122 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4123 being worked on, but at least Exim now implements the latest version to
4126 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4127 process to complete was treated as a delivery failure, a timeout while
4128 writing the message to the pipe was logged, but erroneously treated as a
4129 successful delivery. Such timeouts include transport filter timeouts. For
4130 consistency with the overall process timeout, these timeouts are now
4131 treated as errors, giving rise to delivery failures by default. However,
4132 there is now a new Boolean option for the pipe transport called
4133 timeout_defer, which, if set TRUE, converts the failures into defers for
4134 both kinds of timeout. A transport filter timeout is now identified in
4137 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4138 systems where "make" and "gmake" are different, calling "gmake" at top
4139 level broke things. I've arranged for the value of $(MAKE) to be passed
4140 from the Makefile to this script so that it can call the same version of
4144 A note about Exim versions 4.44 and 4.50
4145 ----------------------------------------
4147 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4148 changes of various kinds. As a consequence, a big documentation update was
4149 needed. This delayed the release for rather longer than seemed good, especially
4150 in the light of a couple of (minor) security issues. Therefore, the changes
4151 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4152 release. So 4.44 and 4.50 are in effect two different branches that both start
4155 I have left the 4.50 change log unchanged; it contains all the changes since
4156 4.43. The change log for 4.44 is below; many of its items are identical to
4157 those for 4.50. This seems to be the most sensible way to preserve the
4158 historical information.
4164 1. Minor wording change to the doc/README.SIEVE file.
4166 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4167 computation of the current number of files was incorrect.
4169 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4170 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4173 4. Give more explanation in the error message when the command for a transport
4174 filter fails to execute.
4176 5. There are several places where Exim runs a non-Exim command in a
4177 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4178 was being done only for the command run by the queryprogram router. It is
4179 now done for all such subprocesses. The other cases are: ${run, transport
4180 filters, and the commands run by the lmtp and pipe transports.
4182 6. Added CONFIGURE_GROUP build-time option.
4184 7. Some older OS have a limit of 256 on the maximum number of file
4185 descriptors. Exim was using setrlimit() to set 1000 as a large value
4186 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4187 systems. I've change it so that if it can't get 1000, it tries for 256.
4189 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4190 was an oversight, and furthermore, ever since the addition of extra
4191 controls (e.g. 4.43/32), the checks on when to allow different forms of
4192 "control" were broken. There should now be diagnostics for all cases when a
4193 control that does not make sense is encountered.
4195 9. Added the /retain_sender option to "control=submission".
4197 10. $recipients is now available in the predata ACL (oversight).
4199 11. Tidy the search cache before the fork to do a delivery from a message
4200 received from the command line. Otherwise the child will trigger a lookup
4201 failure and thereby defer the delivery if it tries to use (for example) a
4202 cached ldap connection that the parent has called unbind on.
4204 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4205 of $address_data from the recipient verification was clobbered by the
4206 sender verification.
4208 13. The value of address_data from a sender verification is now available in
4209 $sender_address_data in subsequent conditions in the ACL statement.
4211 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4213 15. Added a new option "connect=<time>" to callout options, to set a different
4216 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4217 was its contents. (It was OK if the option was not defined at all.)
4219 17. A "Completed" log line is now written for messages that are removed from
4220 the spool by the -Mrm option.
4222 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4223 information about exactly what failed.
4225 19. Added -dd to debug only the daemon process.
4227 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4228 handles timeouts, both on the server side and network timeouts. Renamed the
4229 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4231 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4232 It is now set to "smtps".
4234 22. $host_address is now set to the target address during the checking of
4235 ignore_target_hosts.
4237 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4238 being passed; this would have caused $sender_host_name to have been used if
4239 matching the list had actually called for a host name (not very likely,
4240 since this list is usually IP addresses). A host name is now passed as
4243 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4244 code that specifies a non-restarting handler (typically sigaction() in
4245 modern systems) in an attempt to fix a rare and obscure crash bug.
4247 25. Narrowed the window for a race in the daemon that could cause it to ignore
4248 SIGCHLD signals. This is not a major problem, because they are used only to
4249 wake it up if nothing else does.
4251 26. A malformed maildirsize file could cause Exim to calculate negative values
4252 for the mailbox size or file count. Odd effects could occur as a result.
4253 The maildirsize information is now recalculated if the size or filecount
4256 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4257 support for a long time. Removed HAVE_SYS_VFS_H.
4259 28. Installed the latest version of exipick from John Jetmore.
4261 29. In an address list, if the pattern was not a regular expression, an empty
4262 subject address (from a bounce message) matched only if the pattern was an
4263 empty string. Non-empty patterns were not even tested. This was the wrong
4264 because it is perfectly reasonable to use an empty address as part of a
4265 database query. An empty address is now tested by patterns that are
4266 lookups. However, all the other forms of pattern expect the subject to
4267 contain a local part and a domain, and therefore, for them, an empty
4268 address still always fails if the pattern is not itself empty.
4270 30. Exim went into a mad DNS loop when attempting to do a callout where the
4271 host was specified on an smtp transport, and looking it up yielded more
4272 than one IP address.
4274 31. Re-factored the code for checking spool and log partition space into a
4275 function that finds that data and another that does the check. The former
4276 is then used to implement four new variables: $spool_space, $log_space,
4277 $spool_inodes, and $log_inodes.
4279 32. The RFC2047 encoding function was originally intended for short strings
4280 such as real names; it was not keeping to the 75-character limit for
4281 encoded words that the RFC imposes. It now respects the limit, and
4282 generates multiple encoded words if necessary. To be on the safe side, I
4283 have increased the buffer size for the ${rfc2047: expansion operator from
4286 33. It is now permitted to omit both strings after an "if" condition; if the
4287 condition is true, the result is "true". As before, when the second string
4288 is omitted, a false condition yields an empty string. This makes it less
4289 cumbersome to write custom ACL and router conditions.
4291 34. Failure to deliver a bounce message always caused it to be frozen, even if
4292 there was an errors_to setting on the router. The errors_to setting is now
4295 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4296 canonical form (fully expanded) before being placed in
4297 $sender_host_address.
4299 36. The table in the code that translates DNS record types into text (T_A to
4300 "A" for instance) was missing entries for NS and CNAME. It is just possible
4301 that this could have caused confusion if both these types were looked up
4302 for the same domain, because the text type is used as part of Exim's
4303 per-process caching. But the chance of anyone hitting this buglet seems
4306 37. The dnsdb lookup has been extended in a number of ways.
4308 (1) There is a new type, "zns", which walks up the domain tree until it
4309 finds some nameserver records. It should be used with care.
4311 (2) There is a new type, "mxh", which is like "mx" except that it returns
4312 just the host names, not the priorities.
4314 (3) It is now possible to give a list of domains (or IP addresses) to be
4315 looked up. The behaviour when one of the lookups defers can be
4316 controlled by a keyword.
4318 (4) It is now possible to specify the separator character for use when
4319 multiple records are returned.
4321 38. The dnslists ACL condition has been extended: it is now possible to supply
4322 a list of IP addresses and/or domains to be looked up in a particular DNS
4325 39. Added log_selector=+queue_time_overall.
4327 40. When running the queue in the test harness, wait just a tad after forking a
4328 delivery process, to get repeatability of debugging output.
4330 41. Include certificate and key file names in error message when GnuTLS fails
4331 to set them up, because the GnuTLS error message doesn't include the name
4332 of the failing file when there is a problem reading it.
4334 42. Allow both -bf and -bF in the same test run.
4336 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4338 44. The "Exiscan patch" is now merged into the mainline Exim source.
4340 45. Sometimes the final signoff response after QUIT could fail to get
4341 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4342 < 0 before doing a fflush(). This bug looks as though it goes back to the
4343 introduction of TLS in release 3.20, but "sometimes" must have been rare
4344 because the tests only now provoked it.
4346 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4347 (this can affect the format of dates).
4349 47. exim_tidydb, when checking for the continued existence of a message for
4350 which it has found a message-specific retry record, was not finding
4351 messages that were in split spool directories. Consequently, it was
4352 deleting retry records that should have stayed in existence.
4354 48. Steve fixed some bugs in eximstats.
4356 49. The SPA authentication driver was not abandoning authentication and moving
4357 on to the next authenticator when an expansion was forced to fail,
4358 contradicting the general specification for all authenticators. Instead it
4359 was generating a temporary error. It now behaves as specified.
4361 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4362 (the order specifies the preference for clients). The order is now AES256,
4363 AES128, 3DES, ARCFOUR128.
4365 51. Small patch to Sieve code - explicitly set From: when generating an
4368 52. Exim crashed if a remote delivery caused a very long error message to be
4369 recorded - for instance if somebody sent an entire SpamAssassin report back
4370 as a large number of 550 error lines. This bug was coincidentally fixed by
4371 increasing the size of one of Exim's internal buffers (big_buffer) that
4372 happened as part of the Exiscan merge. However, to be on the safe side, I
4373 have made the code more robust (and fixed the comments that describe what
4376 53. Now that there can be additional text after "Completed" in log lines (if
4377 the queue_time_overall log selector is set), a one-byte patch to exigrep
4378 was needed to allow it to recognize "Completed" as not the last thing in
4381 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4382 patch that reportedly fixes this has been added. I am not expert enough to
4383 create a test for it. This is what the patch creator wrote:
4385 "I found a little strange behaviour of ldap code when working with
4386 Windows 2003 AD Domain, where users was placed in more than one
4387 Organization Units. When I tried to give exim partial DN, the exit code
4388 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4389 But simultaneously result of request was absolutely normal ldap result,
4390 so I produce this patch..."
4392 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4393 REFERENCE, so I have modified the code to exclude the patch when that macro
4396 55. Some experimental protocols are using DNS PTR records for new purposes. The
4397 keys for these records are domain names, not reversed IP addresses. The
4398 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4399 leaves it alone. Component reversal etc. now happens only for IP addresses.
4402 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4404 57. Double the size of the debug message buffer (to 2048) so that more of very
4405 long debug lines gets shown.
4407 58. The exicyclog utility now does better if the number of log files to keep
4408 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4410 59. Two changes related to the smtp_active_hostname option:
4412 (1) $smtp_active_hostname is now available as a variable.
4413 (2) The default for smtp_banner uses $smtp_active_hostname instead
4414 of $primary_hostname.
4416 60. The host_aton() function is supposed to be passed a string that is known
4417 to be a valid IP address. However, in the case of IPv6 addresses, it was
4418 not checking this. This is a hostage to fortune. Exim now panics and dies
4419 if the condition is not met. A case was found where this could be provoked
4420 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4421 components; fortuitously, this particular loophole had already been fixed
4422 by change 4.50/55 above.
4424 If there are any other similar loopholes, the new check in host_aton()
4425 itself should stop them being exploited. The report I received stated that
4426 data on the command line could provoke the exploit when Exim was running as
4427 exim, but did not say which command line option was involved. All I could
4428 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4429 running as the user.
4432 61. There was a buffer overflow vulnerability in the SPA authentication code
4433 (which came originally from the Samba project). I have added a test to the
4434 spa_base64_to_bits() function which I hope fixes it.
4437 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4438 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4440 63. The daemon start-up calls getloadavg() while still root for those OS that
4441 need the first call to be done as root, but it missed one case: when
4442 deliver_queue_load_max is set with deliver_drop_privilege. This is
4443 necessary for the benefit of the queue runner, because there is no re-exec
4444 when deliver_drop_privilege is set.
4446 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4447 This has been fixed.
4449 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4450 was in use, was not putting the data itself into the right store pool;
4451 consequently, it could be overwritten for a subsequent message in the same
4452 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4455 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4457 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4458 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4459 the function were treating the return as a boolean value, which happened to
4460 work because 0=false and not-0=true, but is not correct code.
4462 68. The host_aton() function was not handling scoped IPv6 addresses (those
4463 with, for example, "%eth0" on the end) correctly.
4465 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4466 negated items (that is, ~something) in unsigned ints. Some compilers
4467 apparently mutter when there is no cast.
4469 70. If an address verification called from an ACL failed, and did not produce a
4470 user-specific message (i.e. there was only a "system" message), nothing was
4471 put in $acl_verify_message. In this situation, it now puts the system
4474 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4475 session; change 4.31/43 added the unwanted input to the log line - except
4476 that it did not do this in the start of session case. It now does.
4478 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4479 This is wrong and can cause the other end to generate a synchronization
4480 error if it is another Exim or anything else that does the synchronization
4481 check. A QUIT command is no longer sent after a timeout.
4483 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4484 during host lookups.
4486 74. The defer_ok option of callout verification was not working if it was used
4487 when verifying addresses in header lines, that is, for this case:
4489 verify = header_sender/callout=defer_ok
4491 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4492 those file descriptors could be used for SMTP connections. If anything
4493 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4494 could be sent to the SMTP client, causing chaos. The daemon now opens
4495 stdin, stdout, and stderr to /dev/null when it puts itself into the
4498 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4499 log by default. The user can override this with suitable Perl magic.
4501 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4502 the log message when discard triggers, was not working for the DATA ACL or
4503 for the non-SMTP ACL.
4505 78. Error message wording change in sieve.c.
4507 79. If smtp_accept_max_per_host was set, the number of connections could be
4508 restricted to fewer than expected, because the daemon was trying to set up
4509 a new connection before checking whether the processes handling previous
4510 connections had finished. The check for completed processes is now done
4511 earlier. On busy systems, this bug wouldn't be noticed because something
4512 else would have woken the daemon, and it would have reaped the completed
4515 80. If a message was submitted locally by a user whose login name contained one
4516 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4517 It caused a spool format error. I have fixed the spool reading code. A
4518 related problem was that the "from" clause in the Received: line became
4519 illegal because of the space(s). It is now covered by ${quote_local_part.
4521 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4524 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4526 83. Patch from Sieve maintainer to fix unterminated string problem in
4527 "vacation" handling.
4529 84. Some minor changes to the Linux configuration files to help with other
4530 OS variants using glibc.
4532 85. One more patch for Sieve to update vacation handling to latest spec.
4535 ----------------------------------------------------
4536 See the note above about the 4.44 and 4.50 releases.
4537 ----------------------------------------------------
4543 1. Change 4.43/35 introduced a bug that caused file counts to be
4544 incorrectly computed when quota_filecount was set in an appendfile
4547 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4548 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4551 3. Give more explanation in the error message when the command for a transport
4552 filter fails to execute.
4554 4. There are several places where Exim runs a non-Exim command in a
4555 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4556 was being done only for the command run by the queryprogram router. It is
4557 now done for all such subprocesses. The other cases are: ${run, transport
4558 filters, and the commands run by the lmtp and pipe transports.
4560 5. Some older OS have a limit of 256 on the maximum number of file
4561 descriptors. Exim was using setrlimit() to set 1000 as a large value
4562 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4563 systems. I've change it so that if it can't get 1000, it tries for 256.
4565 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4566 was an oversight, and furthermore, ever since the addition of extra
4567 controls (e.g. 4.43/32), the checks on when to allow different forms of
4568 "control" were broken. There should now be diagnostics for all cases when a
4569 control that does not make sense is encountered.
4571 7. $recipients is now available in the predata ACL (oversight).
4573 8. Tidy the search cache before the fork to do a delivery from a message
4574 received from the command line. Otherwise the child will trigger a lookup
4575 failure and thereby defer the delivery if it tries to use (for example) a
4576 cached ldap connection that the parent has called unbind on.
4578 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4579 of $address_data from the recipient verification was clobbered by the
4580 sender verification.
4582 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4583 was its contents. (It was OK if the option was not defined at all.)
4585 11. A "Completed" log line is now written for messages that are removed from
4586 the spool by the -Mrm option.
4588 12. $host_address is now set to the target address during the checking of
4589 ignore_target_hosts.
4591 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4592 being passed; this would have caused $sender_host_name to have been used if
4593 matching the list had actually called for a host name (not very likely,
4594 since this list is usually IP addresses). A host name is now passed as
4597 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4598 code that specifies a non-restarting handler (typically sigaction() in
4599 modern systems) in an attempt to fix a rare and obscure crash bug.
4601 15. Narrowed the window for a race in the daemon that could cause it to ignore
4602 SIGCHLD signals. This is not a major problem, because they are used only to
4603 wake it up if nothing else does.
4605 16. A malformed maildirsize file could cause Exim to calculate negative values
4606 for the mailbox size or file count. Odd effects could occur as a result.
4607 The maildirsize information is now recalculated if the size or filecount
4610 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4611 support for a long time. Removed HAVE_SYS_VFS_H.
4613 18. Updated exipick to current release from John Jetmore.
4615 19. Allow an empty sender to be matched against a lookup in an address list.
4616 Previously the only cases considered were a regular expression, or an
4619 20. Exim went into a mad DNS lookup loop when doing a callout where the
4620 host was specified on the transport, if the DNS lookup yielded more than
4623 21. The RFC2047 encoding function was originally intended for short strings
4624 such as real names; it was not keeping to the 75-character limit for
4625 encoded words that the RFC imposes. It now respects the limit, and
4626 generates multiple encoded words if necessary. To be on the safe side, I
4627 have increased the buffer size for the ${rfc2047: expansion operator from
4630 22. Failure to deliver a bounce message always caused it to be frozen, even if
4631 there was an errors_to setting on the router. The errors_to setting is now
4634 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4635 canonical form (fully expanded) before being placed in
4636 $sender_host_address.
4638 24. Updated eximstats to version 1.33
4640 25. Include certificate and key file names in error message when GnuTLS fails
4641 to set them up, because the GnuTLS error message doesn't include the name
4642 of the failing file when there is a problem reading it.
4644 26. Expand error message when OpenSSL has problems setting up cert/key files.
4647 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4648 (this can affect the format of dates).
4650 28. exim_tidydb, when checking for the continued existence of a message for
4651 which it has found a message-specific retry record, was not finding
4652 messages that were in split spool directories. Consequently, it was
4653 deleting retry records that should have stayed in existence.
4655 29. eximstats updated to version 1.35
4656 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4657 1.35 - bugfix such that pie charts by volume are generated correctly
4659 30. The SPA authentication driver was not abandoning authentication and moving
4660 on to the next authenticator when an expansion was forced to fail,
4661 contradicting the general specification for all authenticators. Instead it
4662 was generating a temporary error. It now behaves as specified.
4664 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4665 (the order specifies the preference for clients). The order is now AES256,
4666 AES128, 3DES, ARCFOUR128.
4668 31. Small patch to Sieve code - explicitly set From: when generating an
4671 32. Exim crashed if a remote delivery caused a very long error message to be
4672 recorded - for instance if somebody sent an entire SpamAssassin report back
4673 as a large number of 550 error lines. This bug was coincidentally fixed by
4674 increasing the size of one of Exim's internal buffers (big_buffer) that
4675 happened as part of the Exiscan merge. However, to be on the safe side, I
4676 have made the code more robust (and fixed the comments that describe what
4679 33. Some experimental protocols are using DNS PTR records for new purposes. The
4680 keys for these records are domain names, not reversed IP addresses. The
4681 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4682 leaves it alone. Component reversal etc. now happens only for IP addresses.
4685 34. The host_aton() function is supposed to be passed a string that is known
4686 to be a valid IP address. However, in the case of IPv6 addresses, it was
4687 not checking this. This is a hostage to fortune. Exim now panics and dies
4688 if the condition is not met. A case was found where this could be provoked
4689 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4690 components; fortuitously, this particular loophole had already been fixed
4691 by change 4.50/55 or 4.44/33 above.
4693 If there are any other similar loopholes, the new check in host_aton()
4694 itself should stop them being exploited. The report I received stated that
4695 data on the command line could provoke the exploit when Exim was running as
4696 exim, but did not say which command line option was involved. All I could
4697 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4698 running as the user.
4701 35. There was a buffer overflow vulnerability in the SPA authentication code
4702 (which came originally from the Samba project). I have added a test to the
4703 spa_base64_to_bits() function which I hope fixes it.
4706 36. The daemon start-up calls getloadavg() while still root for those OS that
4707 need the first call to be done as root, but it missed one case: when
4708 deliver_queue_load_max is set with deliver_drop_privilege. This is
4709 necessary for the benefit of the queue runner, because there is no re-exec
4710 when deliver_drop_privilege is set.
4712 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4713 was in use, was not putting the data itself into the right store pool;
4714 consequently, it could be overwritten for a subsequent message in the same
4715 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4718 38. Sometimes the final signoff response after QUIT could fail to get
4719 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4720 < 0 before doing a fflush(). This bug looks as though it goes back to the
4721 introduction of TLS in release 3.20, but "sometimes" must have been rare
4722 because the tests only now provoked it.
4728 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4729 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4730 Now it returns an integer. A number of calls were still expecting a T/F
4731 return. Fortuitously, in all cases, the tests worked in OK situations,
4732 which is the norm. However, things would have gone wrong on any write
4733 failures on the smtp file descriptor. This function is used when sending
4734 messages over SMTP and also when doing verify callouts.
4736 2. When Exim is called to do synchronous delivery of a locally submitted
4737 message (the -odf or -odi options), it no longer closes stderr before doing
4740 3. Implemented the mua_wrapper option.
4742 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4744 5. Implemented the functions header_remove(), header_testname(),
4745 header_add_at_position(), and receive_remove_recipient(), and exported them
4748 6. If an ACL "warn" statement specified the addition of headers, Exim already
4749 inserted X-ACL-Warn: at the start if there was no header name. However, it
4750 was not making this test for the second and subsequent header lines if
4751 there were newlines in the string. This meant that an invalid header could
4752 be inserted if Exim was badly configured.
4754 7. Allow an ACL "warn" statement to add header lines at the start or after all
4755 the Received: headers, as well as at the end.
4757 8. Added the rcpt_4xx retry error code.
4759 9. Added postmaster_mailfrom=xxx to callout verification option.
4761 10. Added mailfrom=xxxx to the callout verification option, for verify=
4764 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4765 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4767 12. Inserted some casts to stop certain compilers warning when using pointer
4768 differences as field lengths or precisions in printf-type calls (mostly
4769 affecting debugging statements).
4771 13. Added optional readline() support for -be (dynamically loaded).
4773 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4774 same clock tick as a message's arrival, so that its received time was the
4775 same as the "first fail" time on the retry record, and that message
4776 remained on the queue past the ultimate address timeout, every queue runner
4777 would try a delivery (because it was past the ultimate address timeout) but
4778 after another failure, the ultimate address timeout, which should have then
4779 bounced the address, did not kick in. This was a "< instead of <=" error;
4780 in most cases the first failure would have been in the next clock tick
4781 after the received time, and all would be well.
4783 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4784 being recognized when the domain list was tested by the match_domain
4785 condition in an expansion string.
4787 16. Added the ${str2b64: operator.
4789 17. Exim was always calling setrlimit() to set a large limit for the number of
4790 processes, without checking whether the existing limit was already
4791 adequate. (It did check for the limit on file descriptors.) Furthermore,
4792 errors from getrlimit() and setrlimit() were being ignored. Now they are
4793 logged to the main and panic logs, but Exim does carry on, to try to do its
4794 job under whatever limits there are.
4796 18. Imported PCRE 5.0.
4798 19. Trivial typo in log message " temporarily refused connection" (the leading
4801 20. If the log selector return_path_on_delivery was set and an address was
4802 redirected to /dev/null, the delivery process crashed because it assumed
4803 that a return path would always be set for a "successful" delivery. In this
4804 case, the whole delivery is bypassed as an optimization, and therefore no
4807 21. Internal re-arrangement: the function for sending a challenge and reading
4808 a response while authentication was assuming a zero-terminated challenge
4809 string. It's now changed to take a pointer and a length, to allow for
4810 binary data in such strings.
4812 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4814 23. Exim was not respecting finduser_retries when seeking the login of the
4815 uid under which it was called; it was always trying 10 times. (The default
4816 setting of finduser_retries is zero.) Also, it was sleeping after the final
4817 failure, which is pointless.
4819 24. Implemented tls_on_connect_ports.
4821 25. Implemented acl_smtp_predata.
4823 26. If the domain in control=submission is set empty, Exim assumes that the
4824 authenticated id is a complete email address when it generates From: or
4825 Sender: header lines.
4827 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4828 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4829 chown and chgrp in /bin and hostname in /usr/bin.
4831 28. Exim was keeping the "process log" file open after each use, just as it
4832 does for the main log. This opens the possibility of it remaining open for
4833 long periods when the USR1 signal hits a daemon. Occasional processlog
4834 errors were reported, that could have been caused by this. Anyway, it seems
4835 much more sensible not to leave this file open at all, so that is what now
4838 29. The long-running daemon process does not normally write to the log once it
4839 has entered its main loop, and it closes the log before doing so. This is
4840 so that log files can straightforwardly be renamed and moved. However,
4841 there are a couple of unusual error situations where the daemon does write
4842 log entries, and I had neglected to close the log afterwards.
4844 30. The text of an SMTP error response that was received during a remote
4845 delivery was being truncated at 512 bytes. This is too short for some of
4846 the long messages that one sometimes sees. I've increased the limit to
4849 31. It is now possible to make retry rules that apply only when a message has a
4850 specific sender, in particular, an empty sender.
4852 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4853 it possible to be selective about when SMTP synchronization is enforced.
4855 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4857 32. Implemented hosts_connection_nolog.
4859 33. Added an ACL for QUIT.
4861 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4864 35. Added mailbox_size and mailbox_filecount to appendfile.
4866 36. Added control = no_multiline_responses to ACLs.
4868 37. There was a bug in the logic of the code that waits for the clock to tick
4869 in the case where the clock went backwards by a substantial amount such
4870 that the microsecond fraction of "now" was more than the microsecond
4871 fraction of "then" (but the whole seconds number was less).
4873 38. Added support for the libradius Radius client library this is found on
4874 FreeBSD (previously only the radiusclient library was supported).
4880 1. When certain lookups returned multiple values in the form name=value, the
4881 quoting of the values was not always being done properly. Specifically:
4882 (a) If the value started with a double quote, but contained no whitespace,
4884 (b) If the value contained whitespace other than a space character (i.e.
4885 tabs or newlines or carriage returns) it was not quoted.
4886 This fix has been applied to the mysql and pgsql lookups by writing a
4887 separate quoting function and calling it from the lookup code. The fix
4888 should probably also be applied to nisplus, ibase and oracle lookups, but
4889 since I cannot test any of those, I have not disturbed their existing code.
4891 2. A hit in the callout cache for a specific address caused a log line with no
4892 reason for rejecting RCPT. Now it says "Previous (cached) callout
4893 verification failure".
4895 3. There was an off-by-one bug in the queryprogram router. An over-long
4896 return line was truncated at 256 instead of 255 characters, thereby
4897 overflowing its buffer with the terminating zero. As well as fixing this, I
4898 have increased the buffer size to 1024 (and made a note to document this).
4900 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4901 when Exim is waiting for an SMTP response from a remote server, Exim
4902 restarts its select() call on the socket, thereby resetting its timeout.
4903 This is not a problem when such interrupts are rare. Somebody set up a cron
4904 job to run exiwhat every 2 minutes, which is less than the normal select()
4905 timeout (5 or 10 minutes). This meant that the select() timeout never
4906 kicked in because it was always reset. I have fixed this by comparing the
4907 time when an interrupt arrives with the time at the start of the first call
4908 to select(). If more time than the timeout has elapsed, the interrupt is
4909 treated as a timeout.
4911 5. Some internal re-factoring in preparation for the addition of Sieve
4912 extensions (by MH). In particular, the "personal" test is moved to a
4913 separate function, and given an option for scanning Cc: and Bcc: (which is
4914 not set for Exim filters).
4916 6. When Exim created an email address using the login of the caller as the
4917 local part (e.g. when creating a From: or Sender: header line), it was not
4918 quoting the local part when it contained special characters such as @.
4920 7. Installed new OpenBSD configuration files.
4922 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4923 try to make them clearer.
4925 9. Callout options, other than the timeout value, were being ignored when
4926 verifying sender addresses in header lines. For example, when using
4928 verify = header_sender/callout=no_cache
4930 the cache was (incorrectly) being used.
4932 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4933 only the Cygwin environment.
4935 11. When return_path_on_delivery was set as a log selector, if different remote
4936 addresses in the same message used different return paths and parallel
4937 remote delivery occurred, the wrong values would sometimes be logged.
4938 (Whenever a remote delivery process finished, the return path value from
4939 the most recently started remote delivery process was logged.)
4941 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4942 lines when AUTH and/or TLS are in use. This is the "received protocol"
4943 field. Exim used to use "asmtp" for authenticated SMTP, without any
4944 indication (in the protocol name) for TLS use. Now it follows the RFC and
4945 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4946 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4947 names appear in log lines as well as in Received: header lines.
4949 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4950 extensions, and comparison tests, and to fix some bugs.
4952 14. Changes to the "personal" filter test:
4954 (1) The test was buggy in that it was just doing the equivalent of
4955 "contains" tests on header lines. For example, if a user's address was
4956 anne@some.where, the "personal" test would incorrectly be true for
4958 To: susanne@some.where
4960 This test is now done by extracting each address from the header in turn,
4961 and checking the entire address. Other tests that are part of "personal"
4962 are now done using regular expressions (for example, to check local parts
4963 of addresses in From: header lines).
4965 (2) The list of non-personal local parts in From: addresses has been
4966 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4967 taken from the Sieve specification recommendations.
4969 (3) If the message contains any header line starting with "List-" it is
4970 treated as non-personal.
4972 (4) The test for "circular" in the Subject: header line has been removed
4973 because it now seems ill-conceived.
4975 15. Minor typos in src/EDITME comments corrected.
4977 16. Installed latest exipick from John Jetmore.
4979 17. If headers_add on a router specified a text string that was too long for
4980 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4981 of string_sprintf() is now avoided.
4983 18. $message_body_size was not set (it was always zero) when running the DATA
4984 ACL and the local_scan() function.
4986 19. For the "mail" command in an Exim filter, no default was being set for
4987 the once_repeat time, causing a random time value to be used if "once" was
4988 specified. (If the value happened to be <= 0, no repeat happened.) The
4989 default is now 0s, meaning "never repeat". The "vacation" command was OK
4990 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4991 (I found it when inspecting the code).
4993 20. There is now an overall timeout for performing a callout verification. It
4994 defaults to 4 times the callout timeout, which applies to individual SMTP
4995 commands during the callout. The overall timeout applies when there is more
4996 than one host that can be tried. The timeout is checked before trying the
4997 next host. This prevents very long delays if there are a large number of
4998 hosts and all are timing out (e.g. when the network connections are timing
4999 out). The value of the overall timeout can be changed by specifying an
5000 additional sub-option for "callout", called "maxwait". For example:
5002 verify = sender/callout=5s,maxwait=20s
5004 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5005 to the end before writing, but this should make it even safer).
5007 22. Exim was forgetting that it had advertised PIPELINING for the second and
5008 subsequent messages on an SMTP connection. It was also not resetting its
5009 memory on STARTTLS and an internal HELO.
5011 23. When Exim logs an SMTP synchronization error within a session, it now
5012 records whether PIPELINING has been advertised or not.
5014 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5015 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5016 rather than long int.
5018 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5020 26. Added the never_mail option to autoreply.
5026 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5027 crash if the getsockname() call failed; this can happen if a connection is
5028 closed very soon after it is established. The problem was simply in the
5029 order in which certain operations were done, causing Exim to try to write
5030 to the SMTP stream before it had set up the file descriptor. The bug has
5031 been fixed by making things happen in the correct order.
5037 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5038 before the connection was closed, thus losing the rejection response.
5040 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5041 some early Solaris releases, but causes trouble in current releases where
5042 socklen_t is defined.
5044 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5047 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5050 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5051 that releasing the top of it at the end releases what was used for sub-
5052 expansions (unless the block got too big). However, discard this block if
5053 the first thing is a variable or header, so that we can use its block when
5054 it is dynamic (useful for very large $message_headers, for example).
5056 6. Lookups now cache *every* query, not just the most recent. A new, separate
5057 store pool is used for this. It can be recovered when all lookup caches are
5058 flushed. Lookups now release memory at the end of their result strings.
5059 This has involved some general refactoring of the lookup sources.
5061 7. Some code has been added to the store_xxx() functions to reduce the amount
5062 of flapping under certain conditions.
5064 8. log_incoming_interface used to affect only the <= reception log lines. Now
5065 it causes the local interface and port to be added to several more SMTP log
5066 lines, for example "SMTP connection from", and rejection lines.
5068 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5070 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5072 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5073 could be overwritten at the end of the current message (or the start of a
5074 new message if it was set in a HELO ACL). The value is now preserved for
5075 the duration of the SMTP connection.
5077 12. If a transport had a headers_rewrite setting, and a matching header line
5078 contained an unqualified address, that address was qualified, even if it
5079 did not match any rewriting rules. The underlying bug was that the values
5080 of the flags that permit the existence of unqualified sender and recipient
5081 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5082 non-local messages, and by -bnq for local messages) were not being
5083 preserved with the message after it was received.
5085 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5086 "next input=" as part of the text comprising the host identity instead of
5087 the correct text. The code was using the same buffer for two different
5088 strings. However, depending on which order the printing function evaluated
5089 its arguments, the bug did not always show up. Under Linux, for example, my
5090 test suite worked just fine.
5092 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5093 doesn't work with some older versions of Perl. It has been changed to "my",
5094 which in any case is probably the better facility to use.
5096 15. A really picky compiler found some instances of statements for creating
5097 error messages that either had too many or two few arguments for the format
5100 16. The size of the buffer for calls to the DNS resolver has been increased
5101 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5102 for addresses that have a lot of PTR records. This alleviates a problem; it
5103 does not fully solve it.
5105 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5106 buffer now truncates the list and logs the incident, which is the same
5107 action as happens when Exim is looking up a host name and its aliases.
5108 Previously in this situation something unpredictable would happen;
5109 sometimes it was "internal error: store_reset failed".
5111 18. If a server dropped the connection unexpectedly when an Exim client was
5112 using GnuTLS and trying to read a response, the client delivery process
5113 crashed while trying to generate an error log message.
5115 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5116 string, for example:
5118 warn message = H1: something\nH2: something
5120 the text was added as a single header line from Exim's point of view
5121 though it ended up OK in the delivered message. However, searching for the
5122 second and subsequent header lines using $h_h2: did not work. This has been
5123 fixed. Similarly, if a system filter added multiple headers in this way,
5124 the routers could not see them.
5126 20. Expanded the error message when iplsearch is called with an invalid key to
5127 suggest using net-iplsearch in a host list.
5129 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5130 ACLs are no longer actually imposed (and a message to that effect is
5133 22. If a "gecos" field in a passwd entry contained escaped characters, in
5134 particular, if it contained a \" sequence, Exim got it wrong when building
5135 a From: or a Sender: header from that name. A second bug also caused
5136 incorrect handling when an unquoted " was present following a character
5137 that needed quoting.
5139 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5140 was not being matched caselessly.
5142 24. Arranged for all hyphens in the exim.8 source to be escaped with
5145 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5146 properly. Recipient callout cache records were still being keyed to include
5147 the sender, even when use_sender was set false. This led to far more
5148 callouts that were necessary. The sender is no longer included in the key
5149 when use_sender is false.
5151 26. Added "control = submission" modifier to ACLs.
5153 27. Added the ${base62d: operator to decode base 62 numbers.
5155 28. dnsdb lookups can now access SRV records.
5157 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5158 the configuration file.
5160 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5161 (-v) mode. This makes the output for a verbose queue run more intelligible.
5163 31. Added a use_postmaster feature to recipient callouts.
5165 32. Added the $body_zerocount variable, containing the number of binary zero
5166 bytes in the message body.
5168 33. The time of last modification of the "new" subdirectory is now used as the
5169 "mailbox time last read" when there is a quota error for a maildir
5172 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5174 35. Added +ignore_unknown as a special item in host lists.
5176 36. Code for decoding IPv6 addresses in host lists is now included, even if
5177 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5178 address was recognized as an IP address, but was then not correctly decoded
5179 into binary, causing unexpected and incorrect effects when compared with
5186 1. Very minor rewording of debugging text in manualroute to say "list of
5187 hosts" instead of "hostlist".
5189 2. If verify=header_syntax was set, and a header line with an unqualified
5190 address (no domain) and a large number of spaces between the end of the
5191 name and the colon was received, the reception process suffered a buffer
5192 overflow, and (when I tested it) crashed. This was caused by some obsolete
5193 code that should have been removed. The fix is to remove it!
5195 3. When running in the test harness, delay a bit after writing a bounce
5196 message to get a bit more predictability in the log output.
5198 4. Added a call to search_tidyup() just before forking a reception process. In
5199 theory, someone could use a lookup in the expansion of smtp_accept_max_
5200 per_host which, without the tidyup, could leave open a database connection.
5202 5. Added the variables $recipient_data and $sender_data which get set from a
5203 lookup success in an ACL "recipients" or "senders" condition, or a router
5204 "senders" option, similar to $domain_data and $local_part_data.
5206 6. Moved the writing of debug_print from before to after the "senders" test
5209 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5210 problems for message scanning, either using a data ACL, or using
5211 local_scan() because the Received: header was not generated till after they
5212 were called (in order to set the time as the time of reception completion).
5213 I have revised the way this works. The header is now generated after the
5214 body is received, but before the ACL or local_scan() are called. After they
5215 are run, the timestamp in the header is updated.
5221 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5222 before starting a queue runner without re-exec. This happened only when
5223 deliver_drop_privilege was set or when the Exim user was set to root. The
5224 effect of the bug was that timeouts during subsequent deliveries caused
5225 crashes instead of being properly handled. The handler is now left at its
5226 default (and expected) setting.
5228 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5229 message, again when deliver_drop_privilege is set or Exim is run as root.
5230 The bug described in (1) was not present in this case, but the tidying up
5231 of the other signals was missing. I have made the two cases consistent.
5233 3. The ignore_target_hosts setting on a manualroute router was being ignored
5234 for hosts that were looked up using the /MX notation.
5236 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5239 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5240 operated on the sender address. After changing the $sender_address to <>
5241 for the sender address verify, Exim was re-instated it as the original
5242 (before rewriting) address, but remembering that it had rewritten it, so it
5243 wasn't rewriting it again. This bug also had the effect of breaking the
5244 sender address verification caching when the sender address was rewritten.
5246 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5247 This has been changed so that if the ip literal address matches
5248 ignore_target_hosts, the router declines.
5250 7. Added expansion conditions match_domain, match_address, and match_local_
5251 part (NOT match_host).
5253 8. The placeholder for the Received: header didn't have a length field set.
5255 9. Added code to Exim itself and to exim_lock to test for a specific race
5256 condition that could lead to file corruption when using MBX delivery. The
5257 issue is with the lockfile that is created in /tmp. If this file is removed
5258 after a process has opened it but before that process has acquired a lock,
5259 there is the potential for a second process to recreate the file and also
5260 acquire a lock. This could lead to two Exim processes writing to the file
5261 at the same time. The added code performs the same test as UW imapd; it
5262 checks after acquiring the lock that its file descriptor still refers to
5263 the same named file.
5265 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5266 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5269 11. Added the smtp_active_hostname option. If used, this will typically be made
5270 to depend on the incoming interface address. Because $interface_address is
5271 not set up until the daemon has forked a reception process, error responses
5272 that can happen earlier (such as "too many connections") no longer contain
5275 12. If an expansion in a condition on a "warn" statement fails because a lookup
5276 defers, the "warn" statement is abandoned, and the next ACL statement is
5277 processed. Previously this caused the whole ACL to be aborted.
5279 13. Added the iplsearch lookup type.
5281 14. Added ident_timeout as a log selector.
5283 15. Added tls_certificate_verified as a log selector.
5285 16. Added a global option tls_require_ciphers (compare the smtp transport
5286 option of the same name). This controls incoming TLS connections.
5288 17. I finally figured out how to make tls_require_ciphers do a similar thing
5289 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5290 before starting the TLS session.
5292 18. Tabs are now shown as \t in -bP output.
5294 19. If the log selector return_path_on_delivery was set, Exim crashed when
5295 bouncing a message because it had too many Received: header lines.
5297 20. If two routers both had headers_remove settings, and the first one included
5298 a superfluous trailing colon, the final name in the first list and the
5299 first name in the second list were incorrectly joined into one item (with a
5300 colon in the middle).
5306 1. Added -C and -D options to the exinext utility, mainly to make it easier
5307 to include in the automated testing, but these could be helpful when
5308 multiple configurations are in use.
5310 2. The exinext utility was not formatting the output nicely when there was
5311 an alternate port involved in the retry record key, nor when there was a
5312 message id as well (for retries that were specific to a specific message
5313 and a specific host). It was also confused by IPv6 addresses, because of
5314 the additional colons they contain. I have fixed the IPv4 problem, and
5315 patched it up to do a reasonable job for IPv6.
5317 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5318 delivery, the log line now contains "pipelined" if PIPELINING was used.
5320 4. An SMTP transport process used to panic and die if the bind() call to set
5321 an explicit outgoing interface failed. This has been changed; it is now
5322 treated in the same way as a connect() failure.
5324 5. A reference to $sender_host_name in the part of a conditional expansion
5325 that was being skipped was still causing a DNS lookup. This no longer
5328 6. The def: expansion condition was not recognizing references to header lines
5329 that used bh_ and bheader_.
5331 7. Added the _cache feature to named lists.
5333 8. The code for checking quota_filecount in the appendfile transport was
5334 allowing one more file than it should have been.
5336 9. For compatibility with Sendmail, the command line option
5344 and sets the incoming protocol and host name (for trusted callers). The
5345 host name and its colon can be omitted when only the protocol is to be set.
5346 Note the Exim already has two private options, -pd and -ps, that refer to
5347 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5348 "s", but I don't think that's a major issue.
5350 10. A number of refactoring changes to the code, none of which should affect
5353 (a) The number of logging options was getting close to filling up the
5354 32-bit word that was used as a bit map. I have split them into two classes:
5355 those that are passed in the argument to log_write(), and those that are
5356 only ever tested independently outside of that function. These are now in
5357 separate 32-bit words, so there is plenty of room for expansion again.
5358 There is no change in the user interface or the logging behaviour.
5360 (b) When building, for example, log lines, the code previously used a
5361 macro that called string_cat() twice, in order to add two strings. This is
5362 not really sufficiently general. Furthermore, there was one instance where
5363 it was actually wrong because one of the argument was used twice, and in
5364 one call a function was used. (As it happened, calling the function twice
5365 did not affect the overall behaviour.) The macro has been replaced by a
5366 function that can join an arbitrary number of extra strings onto a growing
5369 (c) The code for expansion conditions now uses a table and a binary chop
5370 instead of a serial search (which was left over from when there were very
5371 few conditions). Also, it now recognizes conditions like "pam" even when
5372 the relevant support is not compiled in: a suitably worded error message is
5373 given if an attempt is made to use such a condition.
5375 11. Added ${time_interval:xxxxx}.
5377 12. A bug was causing one of the ddress fields not to be passed back correctly
5378 from remote delivery subprocesses. The field in question was not being
5379 subsequently used, so this caused to problems in practice.
5381 13. Added new log selectors queue_time and deliver_time.
5383 14. Might have fixed a bug in maildirsizefile handling that threw up
5384 "unexpected character" debug warnings, and recalculated the data
5385 unnecessarily. In any case, I expanded the warning message to give more
5388 15. Added the message "Restricted characters in address" to the statements in
5389 the default ACL that block characters like @ and % in local parts.
5391 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5392 Three changes have been made:
5394 (a) There was a serious bug; a negative response to MAIL caused the whole
5395 recipient domain to be cached as invalid, thereby blocking all messages
5396 to all local parts at the same domain, from all senders. This bug has
5397 been fixed. The domain is no longer cached after a negative response to
5398 MAIL if the sender used is not empty.
5400 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5403 (c) A new callout option, "use_sender" has been added for people who want
5404 the modified behaviour.
5410 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5413 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5414 indeed breaks things for older releases.
5416 3. Added additional logging to the case where there is a problem reading data
5417 from a filter that is running in a subprocess using a pipe, in order to
5418 try to track down a specific problem.
5420 4. Testing facility fudge: when running in the test harness and attempting
5421 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5422 getting "No route to host". Convert this to a timeout.
5424 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5427 6. Some OS don't have socklen_t but use size_t instead. This affects the
5428 fifth argument of getsockopt() amongst other things. This is now
5429 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5430 can be set for individual OS. I have set it for SunOS5, OSF1, and
5431 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5432 some earlier ones do not.
5434 7. Change 4.30/15 was not doing the test caselessly.
5436 8. The standard form for an IPv6 address literal was being rejected by address
5437 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5438 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5439 this, as well as the form without the "IPv6" on the front (but only when
5440 address literals are enabled, of course).
5442 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5444 10. Exim crashed if a message with an empty sender address specified by -f
5445 encountered a router with an errors_to setting. This could be provoked only
5446 by a command such as
5450 where an empty string was supplied; "<>" did not hit this bug.
5452 11. Installed PCRE release 4.5.
5454 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5455 remained set. It is now erased.
5457 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5458 times from message ids (which are base 36 rather than the normal 62).
5460 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5461 were being counted as actual protocol errors, and logged if the log
5462 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5463 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5464 and DATA following a set of rejected RCPTs do not count as protocol errors.
5465 In other words, Exim assumes they were pipelined, though this may not
5466 actually be the case. Of course, in all cases the client gets an
5467 appropriate error code.
5469 15. If a lookup fails in an ACL condition, a message about the failure may
5470 be available; it is used if testing the ACL cannot continue, because most
5471 such messages specify what the cause of the deferral is. However, some
5472 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5473 that caused an old message to be retained and used if a later statement
5474 caused a defer, replacing the real cause of the deferral.
5476 16. If an IP address had so many PTR records that the DNS lookup buffer
5477 was not large enough to hold them, Exim could crash while trying to process
5478 the truncated data. It now detects and logs this case.
5480 17. Further to 4.21/58, another change has been made: if (and only if) the
5481 first line of a message (the first header line) ends with CRLF, a bare LF
5482 in a subsequent header line has a space inserted after it, so as not to
5483 terminate the header.
5485 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5486 unnecessarily, used atoi() instead of strtol(), and didn't check the
5487 termination when getting file sizes from file names by regex.
5489 19. Completely re-implemented the support for maildirsize files, in the light
5490 of a number of problems with the previous contributed implementation
5491 (4.30/29). In particular:
5493 . If the quota is zero, the maildirsize file is maintained, but no quota is
5496 . If the maildir directory does not exist, it is created before any attempt
5497 to write a maildirsize file.
5499 . The quota value in the file is just a cache; if the quota is changed in
5500 the transport, the new value overrides.
5502 . A regular expression is available for excluding directories from the
5505 20. The autoreply transport checks the characters in options that define the
5506 message's headers; it allows continued headers, but it was checking with
5507 isspace() after an embedded newline instead of explicitly looking for a
5510 21. If all the "regular" hosts to which an address was routed had passed their
5511 expiry times, and had not reached their retry times, the address was
5512 bounced, even if fallback hosts were defined. Now Exim should go on to try
5515 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5516 equivalent code in the SMTP transport. Some hosts send humungous responses
5517 to HELO/EHLO, more than 1024 it seems.
5519 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5520 gives compiler warnings in some environments. I've now done it "properly",
5523 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5524 (because of problems with the built-in one) was declared to return uschar *
5525 instead of char *, causing compiler failure.
5527 25. Fixed a file descriptor leak when processing alias/forward files.
5529 26. Fixed a minor format string issue in dbfn.c.
5531 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5533 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5534 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5537 29. When a sender address is verified, it is cached, to save repeating the test
5538 when there is more than one recipient in a message. However, when the
5539 verification involves a callout, it is possible for different callout
5540 options to be set for different recipients. It is too complicated to keep
5541 track of this in the cache, so now Exim always runs a verification when a
5542 callout is required, relying on the callout cache for the optimization.
5543 The overhead is duplication of the address routing, but this should not be
5546 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5547 to be verified with callout=postmaster, and the main callout worked but the
5548 postmaster check failed, the verification correctly failed. However, if a
5549 subsequent RCPT command asked for sender verification *without* the
5550 postmaster check, incorrect caching caused this verification also to fail,
5553 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5554 it was not caching the DNS options (qualify_single, search_parents) that
5555 were used when the lookup failed. A subsequent lookup with different
5556 options therefore always gave the same answer, though there were cases
5557 where it should not have. (Example: a "domains = !$mx_any" option on a
5558 dnslookup router: the "domains" option is always processed without any
5559 widening, but the router might have qualify_single set.) Now Exim uses the
5560 cached value only when the same options are set.
5562 32. Added John Jetmore's "exipick" utility to the distribution.
5564 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5565 than a timeout (e.g. a certificate is required, and is not provided), an
5566 Exim server now closes the connection immediately. Previously it waited for
5567 the client to close - but if the client is SSL, it seems that they each
5568 wait for each other, leading to a delay before one of them times out.
5570 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5571 maintained 0.8.x compatibility because I don't think many are using it, and
5572 it is clearly obsolete.
5574 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5577 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5578 client certificate was expired. A simple patch fixes this, though I don't
5579 understand the full logic of why the verify callback is called multiple
5582 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5583 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5584 which causes problems with some clients (such as the Certicom SSL Plus
5585 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5586 disables the coutermeasure allowing Eudora to connect."
5588 38. Exim was not checking that a write() to a log file succeeded. This could
5589 lead to Bad Things if a log got too big, in particular if it hit a file
5590 size limit. Exim now panics and dies if it cannot write to a log file, just
5591 as it does if it cannot open a log file.
5593 39. Modified OS/Makefile-Linux so that it now contains
5595 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5597 The two -D definitions ensure that Exim is compiled with large file
5598 support, which makes it possible to handle log files that are bigger than
5601 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5602 instance) a domain was checked against a named list that involved a lookup,
5603 causing $domain_data to be set, then another domain was checked against the
5604 same list, then the first domain was re-checked, the value of $domain_data
5605 after the final check could be wrong. In particular, if the second check
5606 failed, it could be set empty. This bug probably also applied to
5609 41. The strip_trailing_dot option was not being applied to the address given
5610 with the -f command-line option.
5612 42. The code for reading a message's header from the spool was incrementing
5613 $received_count, but never initializing it. This meant that the value was
5614 incorrect (doubled) while delivering a message in the same process in which
5615 it was received. In the most common configuration of Exim, this never
5616 happens - a fresh exec is done - but it can happen when
5617 deliver_drop_privilege is set.
5619 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5620 it now includes up to 150 characters of the unexpected data in the log
5623 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5624 and building data strings. The size of both of these buffers was 10 000
5625 bytes - far larger than anybody would *ever* want, thought I. Needless to
5626 say, somebody hit the limit. I have increased the maximum line length to
5627 20 000 and the maximum data length of concatenated lines to 100 000. I have
5628 also fixed two bugs, because there was no checking on these buffers. Tsk,
5629 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5630 buffer is too small.
5632 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5633 lsearch lookups. Now it does.
5635 46. When parsing a route_list item in a manualroute router, a fixed-length
5636 buffer was used for the list of hosts. I made this 1024 bytes long,
5637 thinking that nobody would ever have a list of hosts that long. Wrong.
5638 Somebody had a whole pile of complicated expansion conditions, and the
5639 string was silently truncated, leading to an expansion error. It turns out
5640 that it is easier to change to an unlimited length (owing to other changes
5641 that have happened since this code was originally written) than to build
5642 structure for giving a limitation error. The length of the item that
5643 expands into the list of hosts is now unlimited.
5645 47. The lsearch lookup could not handle data where the length of text line was
5646 more than 4095 characters. Such lines were truncated, leading to shortened
5647 data being returned. It should now handle lines of any length.
5649 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5650 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5653 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5654 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5655 no longer happens in comment lines. A list of replacements is now placed
5656 at the head of all of the source files, except those whose only change is
5657 to replace PERL_COMMAND in the very first #! line.
5659 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5660 messages on the queue, with a bottom-up merge sort, using code contributed
5661 by Michael Haardt. This should make operations like -bp somewhat faster on
5662 large queues. It won't affect queue runners, except when queue_run_in_order
5665 51. Installed eximstats 1.31 in the distribution.
5667 52. Added support for SRV lookups to the dnslookup router.
5669 53. If an ACL referred to $message_body or $message_body_end, the value was not
5670 reset for any messages that followed in the same SMTP session.
5672 54. The store-handling optimization for building very long strings was not
5673 differentiating between the different store pools. I don't think this
5674 actually made any difference in practice, but I've tidied it.
5676 55. While running the routers to verify a sender address, $sender_address
5677 was still set to the sender address. This is wrong, because when routing to
5678 send a bounce to the sender, it would be empty. Therefore, I have changed
5679 it so that, while verifying a sender address, $sender_address is set to <>.
5680 (There is no change to what happens when verifying a recipient address.)
5682 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5683 target A or AAAA records (if not already returned) without resetting the
5684 qualify_single or search_parents options of the DNS resolver. These are
5685 inappropriate in this case because the targets of MX and SRV records must
5686 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5687 target that, when qualified, matched something in the local domain. These
5688 two options are now turned off when doing these lookups.
5690 57. It seems that at least some releases of Reiserfs (which does not have the
5691 concept of a fixed number of inodes) returns zero and not -1 for the
5692 number of available inodes. This interacted badly with check_spool_inodes,
5693 which assumed that -1 was the "no such thing" setting. What I have done is
5694 to check that the total number of inodes is greater than zero before doing
5695 the test of how many are available.
5697 58. When a "warn" ACL statement has a log_message modifier, the message is
5698 remembered, and not repeated. This is to avoid a lot of repetition when a
5699 message has many recipients that cause the same warning to be written.
5700 However, Exim was preserving the list of already written lines for an
5701 entire SMTP session, which doesn't seem right. The memory is now reset if a
5702 new message is started.
5704 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5705 debugging output unless log_rewrite was also set.
5707 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5708 of (int)(handle) when we know that handle contains (void *)(-1).
5710 61. The Exim daemon panic-logs an error return when it closes the incoming
5711 connection. However "connection reset by peer" seems to be common, and
5712 isn't really an error worthy of noting specially, so that particular error
5715 62. When Exim is trying to find all the local interfaces, it used to panic and
5716 die if the ioctl to get the interface flags failed. However, it seems that
5717 on at least one OS (Solaris 9) it is possible to have an interface that is
5718 included in the list of interfaces, but for which you get a failure error
5719 for this call. This happens when the interface is not "plumbed" into a
5720 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5721 failure of the "get flags" call assumes that the interface is down.
5723 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5724 makes life easier for people who are doing arithmetic on fields extracted
5725 from dates, where you often get leading zeros that should not be
5726 interpreted as octal.
5728 64. Added qualify_domain to the redirect router, to override the global
5731 65. If a pathologically long header line contained very many addresses (the
5732 report of this problem mentioned 10 000) and each of them was rewritten,
5733 Exim could use up a very large amount of memory. (It kept on making new
5734 copies of the header line as it rewrote, and never released the old ones.)
5735 At the expense of a bit more processing, the header rewriting function has
5736 been changed so that it no longer eats memory in this way.
5738 66. The generation of the Received: header has been moved from the time that a
5739 message starts to be received, to the time that it finishes. The timestamp
5740 in the Received: header should now be very close to that of the <= log
5741 line. There are two side-effects of this change:
5743 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5744 logged header lines no longer include the local Received: line, because
5745 it has not yet been created. The same applies to a copy of the message
5746 that is returned to a non-SMTP sender when a message is rejected.
5748 (b) When a filter file is tested using -bf, no additional Received: header
5749 is added to the test message. After some thought, I decided that this
5752 This change does not affect the value of $received_for. It is still set
5753 after address rewriting, but before local_scan() is called.
5755 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5757 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5758 gave an unhelpful panic error message, and a defer error. I have managed to
5759 change this behaviour so that it now rejects any supplied certificate,
5760 which seems right, as the list of acceptable certificates is empty.
5762 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5763 gave an unhelpful defer error. I have not managed to make this reject any
5764 supplied certificates, but the error message it gives is "no certificate
5765 supplied", which is not helpful.
5767 70. exigrep's output now also includes lines that are not associated with any
5768 message, but which match the given pattern. Implemented by a patch from
5769 Martin Sluka, which also tidied up the Perl a bit.
5771 71. Recipient callout verification, like sender verification, was using <> in
5772 the MAIL FROM command. This isn't really the right thing, since the actual
5773 sender may affect whether the remote host accepts the recipient or not. I
5774 have changed it to use the actual sender in the callout; this means that
5775 the cache record is now keyed on a recipient/sender pair, not just the
5776 recipient address. There doesn't seem to be a real danger of callout loops,
5777 since a callout by the remote host to check the sender would use <>.
5778 [SEE ABOVE: changed after hitting problems.]
5780 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5781 temporary errors. However, in the case of such a code being given after
5782 the end of a data transmission (i.e. after ".") Exim was failing to write
5783 a retry record for the message. (Yes, there was some broken host that was
5784 actually sending 8xx at this point.)
5786 73. An unknown lookup type in a host list could cause Exim to panic-die when
5787 the list was checked. (An example that provoked this was putting <; in the
5788 middle of a list instead of at the start.) If this happened during a DATA
5789 ACL check, a -D file could be left lying around. This kind of configuration
5790 error no longer causes Exim to die; instead it causes a defer error. The
5791 incident is still logged to the main and panic logs.
5793 74. Buglet left over from Exim 3 conversion. The message "too many messages
5794 in one connection" was written to the rejectlog but not the mainlog, except
5795 when address rewriting (yes!) was being logged.
5797 75. Added write_rejectlog option.
5799 76. When a system filter was run not as root (that is, when system_filter_user
5800 was set), the values of the $n variables were not being returned to the
5801 main process; thus, they were not subsequently available in the $sn
5804 77. Added +return_path_on_delivery log selector.
5806 78. A connection timeout was being treated differently from recipients deferred
5807 when testing hosts_max_try with a message that was older than the host's
5808 retry timeout. (The host should not be counted, thus allowing all hosts to
5809 be tried at least once before bouncing.) This may have been the cause of an
5810 occasionally reported bug whereby a message would remain on the queue
5811 longer than the retry timeout, but would be bounced if a delivery was
5812 forced. I say "may" because I never totally pinned down the problem;
5813 setting up timeout/retry tests is difficult. See also the next item.
5815 79. The ultimate address timeout was not being applied to errors that involved
5816 a combination of host plus message (for example, a timeout on a MAIL
5817 command). When an address resolved to a number of possible hosts, and they
5818 were not all tried for each delivery (e.g. because of hosts_max_try), a
5819 message could remain on the queue longer than the retry timeout.
5821 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5824 81. Fixed an obscure SMTP outgoing bug which required at least the following
5825 conditions: (a) there was another message waiting for the same server;
5826 (b) the server returned 5xx to all RCPT commands in the first message so
5827 that the message was not completed; (c) the server dropped the connection
5828 or gave a negative response to the RSET that Exim sends to abort the
5829 transaction. The observed case was a dropped connection after DATA that had
5830 been sent in pipelining mode. That is, the server had advertised PIPELINING
5831 but was not implementing it correctly. The effect of the bug was incorrect
5832 behaviour, such as trying another host, and this could lead to a crash.
5838 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5839 and daemon.c were passed as pointers to ints; they should have been
5840 pointers to socklen_t variables (which are typically unsigned ints).
5842 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5845 3. Fixed a really odd bug that affected only the testing scheme; patching a
5846 certain fixed string in the binary changed the value of another string that
5847 happened to be identical to the end of the original first string.
5849 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5850 name", it returns that address as the IP address. On some operating
5851 systems (e.g. Solaris), it also passes back the IP address string as the
5852 "host name". However, on others (e.g. Linux), it passes back an empty
5853 string. Exim wasn't checking for this, and was changing the host name to an
5854 empty string, assuming it had been canonicalized.
5856 5. Although rare, it is permitted to have more than one PTR record for a given
5857 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5858 all the names associated with an address, because they do in Solaris.
5859 However, it seems that they do not in Linux for data that comes from the
5860 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5861 I found this out when I moved to a new Linux workstation and tried to run
5862 the Exim test suite.
5864 To get round this problem I have changed the code so that it now does its
5865 own call to the DNS to look up PTR records when searching for a host name.
5866 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5867 addresses that are only in /etc/hosts are still found.
5869 This behaviour is, however, controlled by an option called host_lookup_
5870 order, which defaults to "bydns:byaddr". If people want to use the other
5871 order, or indeed, just use one or the other means of lookup, they can
5872 specify it in this variable.
5874 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5875 some operating systems, this comes back from gethostbyaddr() as an empty
5876 string, and this is what Exim used to test for. However, it seems that in
5877 other systems, "." is yielded. Exim now tests for this case too.
5879 7. The values of check_spool_space and check_log_space are now held internally
5880 as a number of kilobytes instead of an absolute number of bytes. If a
5881 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5882 kilobyte. This means that much larger values can be stored.
5884 8. Exim monitor: an attempt to get the action menu when not actually pointing
5885 at a message produces an empty menu entitled "No message selected". This
5886 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5887 no entries in it ("Shell widget menu has zero width and/or height"). So I
5888 have added a single, blank menu entry in this case.
5890 9. Added ${quote_local_part.
5892 10. MIME decoding is now applied to the contents of Subject: header lines when
5895 11. Now that a reference to $sender_host_address automatically causes a reverse
5896 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5897 host lookup before query-style lookups in lists that might use this
5898 variable. This has therefore been abolished, and the "net-" prefix is no
5899 longer necessary for query-style lookups.
5901 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5902 have been a typo for LFLAGS, so it has been changed.
5904 13. The install script calls Exim with "-C /dev/null" in order to find the
5905 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5906 to be output. However, since Exim outputs its version number before the
5907 error, it didn't break the script. It just looked ugly. I fixed this by
5908 always allowing "-C /dev/null" if the caller is root.
5910 14. Ignore overlarge ACL variable number when reading spool file - insurance
5911 against a later release with more variables having written the file.
5913 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5914 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5915 this, as well as the form without the "IPv6" on the front.
5917 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5918 OS/Makefile-Darwin file.
5920 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5921 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5923 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5924 variable. However, some OS use ints for inodes. I've added cast to long int
5925 to get rid of the compiler warning.
5927 19. I had forgotten to lock out "/../" in configuration file names when
5928 ALT_CONFIG_PREFIX was set.
5930 20. Routers used for verification do not need to specify transports. However,
5931 if such a router generated a host list, and callout was configured, Exim
5932 crashed, because it could not find a port number from the (non-existent)
5933 transport. It now assumes port 25 in this circumstance.
5935 21. Added the -t option to exigrep.
5937 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5938 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5939 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5940 them is set, LOOKUP_LSEARCH is forced.
5942 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5944 24. Added sender and host information to the "rejected by local_scan()" log
5945 line; previously there was no indication of these.
5947 25. Added .include_if_exists.
5949 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5950 Linux. It turns out that not all file systems support this. Apparently some
5951 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5952 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5953 not supported on the file descriptor, is now ignored when Exim is trying to
5954 sync a directory. This applies only to Linux.
5956 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5958 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5959 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5962 29. Added support for maildirsize files from supplied patch (modified a bit).
5964 30. The use of :fail: followed by an empty string could lead Exim to respond to
5965 sender verification failures with (e.g.):
5967 550 Verification failed for <xxx>
5968 550 Sender verify failed
5970 where the first response line was missing the '-' that indicates it is not
5971 the final line of the response.
5973 31. The loop for finding the name of the user that called Exim had a hardwired
5974 limit of 10; it now uses the value of finduser_retries, which is used for
5975 all other user lookups.
5977 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5980 33. Exim was neglecting to zero errno before one call of strtol() when
5981 expanding a string and expecting an integer value. On some systems this
5982 resulted in spurious "integer overflow" errors. Also, it was casting the
5983 result into an int without checking.
5985 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5986 did not work. The code looks as if it has *never* worked, though it appears
5987 to have been documented since at least release 1.62. I have made it work.
5989 35. The "timeout_DNS" error in retry rules, also documented since at least
5990 1.62, also never worked. As it isn't clear exactly what this means, and
5991 clearly it isn't a major issue, I have abolished the feature by treating it
5992 as "timeout", and writing a warning to the main and panic logs.
5994 36. The display of retry rules for -brt wasn't always showing the error code
5997 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5998 timeout_connect_A, timeout_connect_MX.
6000 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6001 to the empty sender.
6003 39. The daemon was not analysing the content of -oX till after it had closed
6004 stderr and disconnected from the controlling terminal. This meant that any
6005 syntax errors were only noted on the panic log, and the return code from
6006 the command was 0. By re-arranging the code a little, I've made the
6007 decoding happen first, so such errors now appear on stderr, and the return
6008 code is 1. However, the actual setting up of the sockets still happens in
6009 the disconnected process, so errors there are still only recorded on the
6012 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6013 connections (as happens on some IP stacks) was logged at start up time as
6014 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6015 it from "IPv6 and IPv4", which means that two separate sockets are being
6018 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6019 says whether AF_INET or AF_INET6 was passed as an argument.
6021 42. Exiwhat output was messed up when time zones were included in log
6024 43. Exiwhat now gives more information about the daemon's listening ports,
6025 and whether -tls-on-connect was used.
6027 44. The "port" option of the smtp transport is now expanded.
6029 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6030 silently ignored. Now an error message is written to the main and panic
6033 46. There's a new ACL modifier called "logwrite" which writes to a log file
6034 as soon as it is encountered.
6036 47. Added $local_user_uid and $local_user_gid at routing time.
6038 48. Exim crashed when trying to verify a sender address that was being
6041 49. Exim was recognizing only a space character after ".include". It now also
6042 recognizes a tab character.
6044 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6045 extracting the relevant information from the specification. The man page no
6046 longer contains scrambled data for the -d option, and I've added a section
6047 at the front about calling Exim under different names.
6049 51. Added "extra_headers" argument to the "mail" command in filter files.
6051 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6054 53. Installed eximstats 1.29.
6056 54. Added transport_filter_timeout as a generic transport option.
6058 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6059 Cc: header lines. This was required by RFC 822, but it not required by RFC
6062 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6063 incoming messages that did not have them. Now it does so only if the
6064 message originates locally, that is, if there is no associated remote host
6065 address. When Resent- header lines are present, this applies to the Resent-
6066 lines rather than the non-Resent- lines.
6068 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6069 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6071 58. Messages for configuration errors now include the name of the main
6072 configuration file - useful now that there may be more than one file in a
6073 list (.included file names were always shown).
6075 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6076 for those rare installations that do not start the daemon as root or run it
6077 setuid root. I've cut out the call to initgroups() if the daemon is not
6080 60. The Exim user and group can now be bound into the binary as text strings
6081 that are looked up at the start of Exim's processing.
6083 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6085 62. Added $mailstore_basename variable.
6087 63. Installed patch to sieve.c from Michael Haardt.
6089 64. When Exim failed to open the panic log after failing to open the main log,
6090 the original message it was trying to log was written to stderr and debug
6091 output, but if they were not available (the usual case in production), it
6092 was lost. Now it is written to syslog before the two lines that record the
6093 failures to open the logs.
6095 65. Users' Exim filters run in subprocesses under the user's uid. It is
6096 possible for a "deliver" command or an alias in a "personal" command to
6097 provoke an address rewrite. If logging of address rewriting is configured,
6098 this fails because the process is not running as root or exim. There may be
6099 a better way of dealing with this, but for the moment (because 4.30 needs
6100 to be released), I have disabled address rewrite logging when running a
6101 filter in a non-root, non-exim process.
6107 1. The buildconfig auxiliary program wasn't quoting the value set for
6108 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6109 not defined. This bug was masked in 4.22 by the effect that was fixed in
6112 2. Some messages that were rejected after a message id was allocated were
6113 shown as "incomplete" by exigrep. It no longer does this for messages that
6114 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6116 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6117 have allow_domain_literals set, the ID did not get logged in the <= line.
6118 Domain literals are now always recognized in Message-ID: header lines.
6120 4. The first argument for a ${extract expansion item is the key name or field
6121 number. Leading and trailing spaces in this item were not being ignored,
6122 causing some misleading effects.
6124 5. When deliver_drop_privilege was set, single queue runner processes started
6125 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6126 same command in the process it spins off) were not dropping privilege.
6128 6. When the daemon running as "exim" started a queue runner, it always
6129 re-executed Exim in the spun-off process. This is a waste of effort when
6130 deliver_drop_privilege is set. The new process now just calls the
6131 queue-runner function directly.
6137 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6140 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6141 ignored. Though the use of -oP was forcing the writing of a pid file, it
6142 was always written to the default place.
6144 3. If the message "no IP address found for host xxxx" is generated during
6145 incoming verification, it is now followed by identification of the incoming
6146 connection (so you can more easily find what provoked it).
6148 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6150 5. Added some features to "harden" Exim a bit more against certain attacks:
6152 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6153 be put in Local/Makefile. This is like the never_users runtime option,
6154 but it cannot be overridden. The default setting is "root".
6156 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6157 prefix string with which any file named in a -C command line option
6160 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6161 is retained for -C and -D only if the caller of Exim is root. Without
6162 it, the exim user may also use -C and -D and retain privilege.
6164 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6165 command line option is disabled.
6167 6. Macro names set by the -D option must start with an upper case letter, just
6168 like macro names defined in the configuration file.
6170 7. Added "dereference=" facility to LDAP.
6172 8. Two instances of the typo "uknown" in the source files are fixed.
6174 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6175 the Configure-Makefile script screwed up while processing it.
6177 10. Incorporated PCRE 4.4.
6179 11. The SMTP synchronization check was not operating right at the start of an
6180 SMTP session. For example, it could not catch a HELO sent before the client
6181 waited for the greeting. There is now a check for outstanding input at the
6182 point when the greeting is written. Because of the duplex, asynchronous
6183 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6184 way, but not yet received, when the check is performed.
6186 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6187 on TCP/IP sockets, because this apparently causes some broken clients to
6190 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6191 unchanged) from the Cygwin maintainer.
6193 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6194 was supported instead of testing for mbx. Effectively a typo.
6196 15. The spa authenticator server code was not checking that the input it
6197 received was valid base64.
6199 16. The debug output line for the "set" modifier in ACLs was not showing the
6200 name of the variable that was being set.
6202 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6204 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6205 lookup on its own. Something else was needed to trigger the lookup. For
6206 example, a match in host_lookup or the need for a host name in a host list.
6207 Now, if $sender_host_name is referenced and the host name has not yet been
6208 looked up, a lookup is performed. If the lookup fails, the variable remains
6209 empty, and $host_lookup_failed is set to "1".
6211 19. Added "eqi" as a case-independent comparison operator.
6213 20. The saslauthd authentication condition could segfault if neither service
6214 nor realm was specified.
6216 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6217 error message that was logged was misleading, and incoming SMTP
6218 connections were dropped. The message is now more accurate, and temporary
6219 errors are given to SMTP connections.
6221 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6222 (except RSET) with a 421 error, until QUIT is received. However, it was
6223 failing to send a response to QUIT.
6225 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6226 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6227 ACL is now run after the helo_try_verify_hosts code.
6229 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6230 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6231 case-independent, so other case variants are also recognized). Apparently
6232 some systems use these upper case variants.
6234 25. If more than two messages were waiting for the same host, and a transport
6235 filter was specified for the transport, Exim sent two messages over the
6236 same TCP/IP connection, and then failed with "socket operation on non-
6237 socket" when it tried to send the third.
6239 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6241 27. The extern definition of crypt16() in expand.c was not being excluded when
6242 the OS had its own crypt16() function.
6244 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6245 as a preferred synonym for return_size_limit, both as an option and as an
6248 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6250 30. Changed the default configuration ACL to relax the local part checking rule
6251 for addresses that are not in any local domains. For these addresses,
6252 slashes and pipe symbols are allowed within local parts, but the sequence
6253 /../ is explicitly forbidden.
6255 31. SPA server authentication was not clearing the challenge buffer before
6258 32. log_message in a "warn" ACL statement was writing to the reject log as
6259 well as to the main log, which contradicts the documentation and doesn't
6260 seem right (because no rejection is happening). So I have stopped it.
6262 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6263 However, I am unable to do any testing of this.
6265 34. Fixed an infelicity in the appendfile transport. When checking directories
6266 for a mailbox, to see if any needed to be created, it was accidentally
6267 using path names with one or more superfluous leading slashes; tracing
6268 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6270 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6271 contents are added to the log line that is written for every discarded
6272 recipient. (Previously a log_message setting was ignored.)
6274 36. The ${quote: operator now quotes the string if it is empty.
6276 37. The install script runs exim in order to find its version number. If for
6277 some reason other than non-existence or emptiness, which it checks, it
6278 could not run './exim', it was installing it with an empty version number,
6279 i.e. as "exim-". This error state is now caught, and the installation is
6282 38. An argument was missing from the function that creates an error message
6283 when Exim fails to connect to the socket for saslauthd authentication.
6284 This could cause Exim to crash, or give a corrupted message.
6286 39. Added isip, isip4, and isip6 to ${if conditions.
6288 40. The ACL variables $acl_xx are now saved with the message, and can be
6289 accessed later in routers, transports, and filters.
6291 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6292 strings in the file are not string-expanded.
6294 42. If a MAIL command specified a SIZE value that was too large to fit into an
6295 int variable, the check against message_size_limit failed. Such values are
6296 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6297 day this will have to be increased, but I don't think I want to be around
6298 when emails are that large.
6305 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6306 iconv() is not standard in FreeBSD.
6308 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6309 IPv6 enabled. The observed symptom was a segmentation fault on return from
6310 the function os_common_find_running_interfaces() in src/os.c.
6312 3. In the check_special_case() function in daemon.c I had used "errno" as an
6313 argument name, which causes warnings on some systems. This was basically a
6314 typo, since it was named "eno" in the comments!
6316 4. The code that waits for the clock to tick (at a resolution of some fraction
6317 of a second) so as to ensure message-id uniqueness was always waiting for
6318 at least one whole tick, when it could have waited for less. [This is
6319 almost certainly not relevant at current processor speeds, where it is
6320 unlikely to ever wait at all. But we try to future-proof.]
6322 5. The function that sleeps for a time interval that includes fractions of a
6323 second contained a race. It did not block SIGALRM between setting the
6324 timer, and suspending (a couple of lines later). If the interval was short
6325 and the sigsuspend() was delayed until after it had expired, the suspension
6326 never ended. On busy systems this could lead to processes getting stuck for
6329 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6330 process, before it forks any delivery processes. The open lookup caching
6331 mechanism meant that the open file or database connection was passed into
6332 the delivery process. The problem was that delivery processes always tidy
6333 up cached lookup data. This could cause a problem for the next delivery
6334 process started by the queue runner, because the external queue runner
6335 process does not know about the closure. So the next delivery process
6336 still has data in the lookup cache. In the case of a file lookup, there was
6337 no problem because closing a file descriptor in a subprocess doesn't affect
6338 the parent. However, if the lookup was caching a connection to a database,
6339 the connection was closed, and the second delivery process was likely to
6340 see errors such as "PGSQL: query failed: server closed the connection
6341 unexpectedly". The problem has been fixed by closing all cached lookups
6342 in a queue runner before running a delivery process.
6344 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6345 seem to have the "const" qualifier which it has on other OS. I've
6348 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6349 *of the same type* (client or server) with the same public name that an
6350 error should be diagnosed.
6352 9. When Exim looked up a host name for an IP address, but failed to find the
6353 original IP address when looking up the host name (a safety check), it
6354 output the message "<ip address> does not match any IP for NULL", which was
6355 confusing, to say the least. The bug was that the host name should have
6356 appeared instead of "NULL".
6358 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6359 user that is built into the binary, and the -C or -D options is used, root
6360 privilege is dropped before the configuration file is read. In addition,
6361 logging is switched to stderr instead of the normal log files. If the
6362 configuration then re-defines the Exim user, the unprivileged environment
6363 is probably not what is expected, so Exim logs a panic warning message (but
6366 However, if deliver_drop_privilege is set, the unprivileged state may well
6367 be exactly what is intended, so the warning has been cut out in that case,
6368 and Exim is allowed to try to write to its normal log files.
6374 1. smtp_return_error_details was not giving details for temporary sender
6375 or receiver verification errors.
6377 2. Diagnose a configuration error if two authenticators have the same public
6380 3. Exim used not to create the message log file for a message until the first
6381 delivery attempt. This could be confusing when incoming messages were held
6382 for policy or load reasons. The message log file is now created at the time
6383 the message is received, and an initial "Received" line is written to it.
6385 4. The automatically generated man page for command line options had a minor
6386 bug that caused no ill effects; however, a more serious problem was that
6387 the procedure for building the man page automatically didn't always
6388 operate. Consequently, release 4.20 contains an out-of-date version. This
6389 shouldn't happen again.
6391 5. When building Exim with embedded Perl support, the script that builds the
6392 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6393 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6395 6. The freeze_tell option was not being used for messages that were frozen on
6396 arrival, either by an ACL or by local_scan().
6398 7. Added the smtp_incomplete_transaction log selector.
6400 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6401 was accepting AUTH without a new EHLO.
6403 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6404 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6407 10. Logging of TCP/IP connections (when configured) now happens in the main
6408 daemon process instead of the child process, so that the TCP/IP connection
6409 count is more accurate (but it can never be perfect).
6411 11. The use of "drop" in a nested ACL was not being handled correctly in the
6412 outer ACL. Now, if condition failure induced by the nested "drop" causes
6413 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6414 or "require"), the connection is dropped.
6416 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6417 that yield "discard" can now be used with an "accept" or a "discard" verb,
6418 but an error is generated for any others (because I can't see a useful way
6419 to define what should happen).
6421 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6422 are now processed in the same way as lines in the Exim configuration file.
6423 In particular, continuation lines are supported.
6425 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6427 15. Added -ti meaning -t -i.
6429 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6430 domains, and warn by logging if others are found.
6432 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6433 returned from GIFCONF when Exim is trying to find the list of interfaces on
6434 a host. The code in os.c has been modified to copy each ifreq to an aligned
6435 structure in all cases.
6437 Also, in some cases, the returned ifreq's were being copied to a 'struct
6438 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6439 means the last couple of bytes of an IPv6 address could be chopped if the
6440 ifreq contained only a normal sockaddr (14 bytes storage).
6442 18. Named domain lists were not supported in the hosts_treat_as_local option.
6443 An entry such as +xxxx was not recognized, and was treated as a literal
6446 19. Ensure that header lines added by a DATA ACL are included in the reject log
6447 if the ACL subsequently rejects the message.
6449 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6450 MD5 (which is deprecated).
6452 21. When testing a filter file using -bf, Exim was writing a message when it
6453 took the sender from a "From " line in the message, but it was not doing so
6454 when it took $return_path from a Return-Path: header line. It now does.
6456 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6457 with a valid header line field name (a series of printing characters
6458 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6460 23. Changed "disc" in the source to "disk" to conform to the documentation and
6461 the book and for uniformity.
6463 24. Ignore Sendmail's -Ooption=value command line item.
6465 25. When execve() failed while trying to run a command in a pipe transport,
6466 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6467 could be confused with a return value of 69 from the command itself. This
6468 has been changed to 127, the value the shell returns if it is asked to run
6469 a non-existent command. The wording for the related log line suggests a
6470 non-existent command as the problem.
6472 26. If received_header_text expands to an empty string, do not add a Received:
6473 header line to the message. (Well, it adds a token one on the spool, but
6474 marks it "old" so that it doesn't get used or transmitted.)
6476 27. Installed eximstats 1.28 (addition of -nt option).
6478 28. There was no check for failure on the call to getsockname() in the daemon
6479 code. This can fail if there is a shortage of resources on the system, with
6480 ENOMEM, for example. A temporary error is now given on failure.
6482 29. Contrary to the C standard, it seems that in some environments, the
6483 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6484 program. Exim now does this explicitly; it affects the formatting of
6485 timestamps using strftime().
6487 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6488 complaints. I've now initialized all the variables, to avoid this.
6490 32. Header lines added by a system filter were not being "seen" during
6491 transport-time rewrites.
6493 33. The info_callback() function passed to OpenSSL is set up with type void
6494 (*)(SSL *, int, int), as described somewhere. However, when calling the
6495 function (actually a macro) that sets it up, the type void(*)() is
6496 expected. I've put in a cast to prevent warnings from picky compilers.
6498 34. If a DNS black list lookup found a CNAME record, but there were no A
6499 records associated with the domain it pointed at, Exim crashed.
6501 35. If a DNS black list lookup returned more than one A record, Exim ignored
6502 all but the first. It now scans all returned addresses if a particular IP
6503 value is being sought. In this situation, the contents of the
6504 $dnslist_value variable are a list of all the addresses, separated by a
6507 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6508 to accept a host name and all its aliases if the forward lookup for any of
6509 them yielded the IP address of the incoming connection. Now it accepts only
6510 those names whose forward lookup yields the correct IP address. Any other
6511 names are discarded. This closes a loophole whereby a rogue DNS
6512 administrator could create reverse DNS records to break through a
6513 wildcarded host restriction in an ACL.
6515 37. If a user filter or a system filter that ran in a subprocess used any of
6516 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6517 the wrong values were passed to the pipe command ($thisaddress had the
6518 value of $0, $0 had the value of $1, etc). This bug was introduced by
6519 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6521 38. Improved the line breaking for long SMTP error messages from ACLs.
6522 Previously, if there was no break point between 40 and 75 characters, Exim
6523 left the rest of the message alone. Two changes have been made: (a) I've
6524 reduced the minimum length to 35 characters; (b) if it can't find a break
6525 point between 35 and 75 characters, it looks ahead and uses the first one
6526 that it finds. This may give the occasional overlong line, but at least the
6527 remaining text gets split now.
6529 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6530 file descriptors might be low, and that setting 1000 would always raise it.
6531 It turns out that in some environments, the limit is already over 1000 and
6532 that lowering it causes trouble. So now Exim takes care not to decrease it.
6534 40. When delivering a message, the value of $return_path is set to $sender_
6535 address at the start of routing (routers may change the value). By an
6536 oversight, this default was not being set up when an address was tested by
6537 -bt or -bv, which affected the outcome if any router or filter referred to
6540 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6541 log only when "message" or "log_message" are set. However, if one of the
6542 conditions was an address verification, or a call to a nested ACL, the
6543 messages generated by the underlying test were being passed through. This
6544 no longer happens. The underlying message is available in $acl_verify_
6545 message for both "message" and "log_message" expansions, so it can be
6546 passed through if needed.
6548 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6549 new expansion $bh_ to give the encoded byte string without charset
6550 translation. Translation happens only if iconv() is available; HAVE_ICONV
6551 indicates this at build time. HEADERS_CHARSET gives the charset to
6552 translate to; headers_charset can change it in the configuration, and
6553 "headers charset" can change it in an individual filter file.
6555 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6556 that creates RFC 2047 encoded "words" labels them as that charset instead
6557 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6558 expansion operator; (ii) when Exim creates a From: line for a local
6559 message; (iii) when a header line is rewritten to include a "phrase" part.
6561 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6562 buggy, causing it to skip the first lines of messages whose message ID
6563 ended in 'D'. This would not have bitten before Exim release 4.14, because
6564 message IDs were unlikely to end in 'D' before then. The effect was to have
6565 incorrect size information for certain domains.
6567 45. #include "config.h" was missing at the start of the crypt16.c module. This
6568 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6571 46. If there was a timeout during a "random" callout check, Exim treated it as
6572 a failure of the random address, and carried on sending RSET and the real
6573 address. If the delay was just some slowness somewhere, the response to the
6574 original RCPT would be taken as a response to RSET and so on, causing
6575 mayhem of various kinds.
6577 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6578 when I implemented it. It didn't allow for the fact that some option values
6579 may legitimately be negative (e.g. size_addition), and it didn't even do
6580 the right test for positive values.
6582 48. Domain names in DNS records are case-independent. Exim always looks them up
6583 in lower case. Some resolvers return domain names in exactly the case they
6584 appear in the zone file, that is, they may contain uppercase letters. Not
6585 all resolvers do this - some return always lower case. Exim was treating a
6586 change of case by a resolver as a change of domain, similar to a widening
6587 of a domain abbreviation. This triggered its re-routing code and so it was
6588 trying to route what was effectively the same domain again. This normally
6589 caused routing to fail (because the router wouldn't handle the domain
6590 twice). Now Exim checks for this case specially, and just changes the
6591 casing of the domain that it ultimately uses when it transmits the message
6594 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6597 50. If a filter generated a file delivery with a non-absolute name (possible if
6598 no home directory exists for the router), the forbid_file option was not
6601 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6602 the existing equality matching.
6604 52. Exim was using ints instead of ino_t variables in some places where it was
6605 dealing with inode numbers.
6607 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6608 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6609 called TMPDIR, and if it finds it is different, it changes its value.
6611 54. The smtp_printf() function is now made available to local_scan() so
6612 additional output lines can be written before returning. There is also an
6613 smtp_fflush() function to enable the detection of a dropped connection.
6614 The variables smtp_input and smtp_batched_input are exported to
6617 55. Changed the default runtime configuration: the message "Unknown user"
6618 has been removed from the ACL, and instead placed on the localuser router,
6619 using the cannot_route_message feature. This means that any verification
6620 failures that generate their own messages won't get overridden. Similarly,
6621 the "Unrouteable address" message that was in the ACL for unverifiable
6622 relay addresses has also been removed.
6624 56. Added hosts_avoid_esmtp to the smtp transport.
6626 57. The exicyclog script was not checking for the esoteric option
6627 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6628 will work only if exicyclog is run under the appropriate euid.
6630 58. Following a discussion on the list, the rules by which Exim recognises line
6631 endings on incoming messages have been changed. The -dropcr and drop_cr
6632 options are now no-ops, retained only for backwards compatibility. The
6633 following line terminators are recognized: LF CRLF CR. However, special
6634 processing applies to CR:
6636 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6637 nor a local message in the state where . is a terminator.
6639 (ii) If a bare CR is encountered in a header line, an extra space is added
6640 after the line terminator so as not to end the header. The reasoning
6641 behind this is that bare CRs in header lines are most likely either
6642 to be mistakes, or people trying to play silly games.
6644 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6645 was being incorrectly given as 18 bytes larger than it should have been.
6646 This is a VOB (very old bug).
6648 60. This may never have affected anything current, but just in case it has:
6649 When the local host is found other than at the start of a list of hosts,
6650 the local host, those with the same MX, and any that follow, are discarded.
6651 When the list in question was part of a longer list of hosts, the following
6652 hosts (not currently being processed) were also being discarded. This no
6653 longer happens. I'm not sure if this situation could ever has previously
6656 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6659 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6660 header. This is something that is recommended in a new Internet Draft, and
6661 is something that is documented as being done by Sendmail. There are two
6662 possible values. For messages generated by the autoreply transport, Exim
6665 Auto-Submitted: auto-replied
6667 whereas for all other generated messages (e.g. bounces) it adds
6669 Auto-Submitted: auto-generated
6671 63. The "personal" condition in filters now includes a test for the
6672 Auto-Submitted: header. If it contains the string "auto-" the message it
6673 not considered personal.
6675 64. Added rcpt_include_affixes as a generic transport option.
6677 65. Added queue_only_override (default true).
6679 66. Added the syslog_duplication option.
6681 67. If what should have been the first header line of a message consisted of
6682 a space followed by a colon, Exim was mis-interpreting it as a header line.
6683 It isn't of course - it is syntactically invalid and should therefore be
6684 treated as the start of the message body. The misbehaviour could have
6685 caused a number of strange effects, including loss of data in subsequent
6686 header lines, and spool format errors.
6688 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6689 client host had authenticated. This control can now be exercised by an ACL
6690 for more flexibility.
6692 69. By default, callouts do not happen when testing with -bh. There is now a
6693 variant, -bhc, which does actually run the callout code, including
6694 consulting and updating the callout cache.
6696 70. Added support for saslauthd authentication, courtesy of Alexander
6699 71. If statvfs() failed on the spool or log directories while checking their
6700 size for availability, Exim confusingly gave the error "space shortage".
6701 Furthermore, in debugging mode it crashed with a floating point exception.
6702 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6703 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6704 serious problem, Exim now writes to the main and panic logs when this
6705 happens, with details of the failure. It then refuses to accept the
6706 incoming message, giving the message "spool directory problem" or "log
6707 directory problem" with a 421 code for SMTP messages.
6709 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6710 0, 1, and 2 exist, because some OS complain for execs without them (see
6711 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6712 descriptors. However, the code omitted to check that the open succeeded,
6713 causing mysterious errors if for some reason the permissions on /dev/null
6714 got screwed. Now Exim writes a message to the main and panic logs, and
6715 bombs out if it can't open /dev/null.
6717 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6718 interact so that it is all more flexible. It is supposed to remain
6719 backwards compatible. Also added extra_local_interfaces.
6721 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6722 to bomb out with an assertion failure - to the client this appears as a
6723 connection drop. This problem occurs in the part of the code that was taken
6724 from the Samba project. Fortunately, the assertion is in a very simple
6725 function, so I have fixed this by reproducing the function inline in the
6726 one place where it is called, and arranging for authentication to fail
6727 instead of killing the process with assert().
6729 75. The SPA client code was not working when the server requested OEM rather
6730 than Unicode encoding.
6732 76. Added code to make require_files with a specific uid setting more usable in
6733 the case where statting the file as root fails - usually a non-root-mounted
6734 NFS file system. When this happens and the failure is EACCES, Exim now
6735 forks a subprocess and does the per-uid checking as the relevant uid.
6737 77. Added process_log_path.
6739 78. If log_file_path was not explicitly set, a setting of check_log_space or
6740 check_log_inodes was ignored.
6742 79. If a space check for the spool or log partitions fails, the incident is now
6743 logged. Of course, in the latter case the data may get lost...
6745 80. Added the %p formatting code to string_format() so that it can be used to
6746 print addresses in debug_print(). Adjusted all the address printing in the
6747 debugging in store.c to use %p rather than %d.
6749 81. There was a concern that a line of code in smtp_in.c could overflow a
6750 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6751 initially expressed, the concern was not well-founded, because trailing
6752 spaces are removed early. However, if the trailing spaces were followed by
6753 a NULL, they did not get removed, so the overflow was possible. Two fixes
6756 (a) I re-wrote the offending code in a cleaner fashion.
6757 (b) If an incoming SMTP command contains a NULL character, it is rejected
6760 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6761 runs initgroups(), so that if the Exim user is in any additional groups,
6762 they will be used during message reception.
6768 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob