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.
19 JH/02 Rework error string handling in DKIM to pass more info back to callers.
20 This permits better logging.
22 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
23 do not close it down and have the child transport start it up again on
24 the passed-on TCP connection. Instead, proxy the child (and any
25 subsequent ones) for TLS via a unix-domain socket channel. Logging is
26 affected: the continued delivery log lines do not have any DNSSEC, TLS
27 Certificate or OCSP information. TLS cipher information is still logged.
29 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
30 identical IP addresses on different listening ports. Will also affect
33 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
34 add noisy ifdef guards to special-case this sillyness.
35 Patch from Bernd Kuhls.
37 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
38 than 255 are no longer allowed.
40 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
41 Disable the session-cache too, which might reduce our load. Since we
42 currrectly use a new context for every connection, both as server and
43 client, there is no benefit for these.
44 GnuTLS appears to not support tickets server-side by default (we don't
45 call gnutls_session_ticket_enable_server()) but client side is enabled
46 by default on recent versions (3.1.3 +) unless the PFS priority string
49 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
50 <https://reproducible-builds.org/specs/source-date-epoch/>.
52 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
53 the check for any unsuccessful recipients did not notice the limit, and
54 erroneously found still-pending ones.
56 JH/08 Pipeline CHUNKING command and data together, on kernels that support
57 MSG_MORE. Only in-clear (not on TLS connections).
59 JH/09 Avoid using a temporary file during transport using dkim. Unless a
60 transport-filter is involved we can buffer the headers in memory for
61 creating the signature, and read the spool data file once for the
62 signature and again for transmission.
64 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
65 4.77 as the kernel support then wasn't solid, having issues in 64bit
66 mode. Now, it's been long enough. Add support for FreeBSD also.
68 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
69 case where the routing stage had gathered several addresses to send to
70 a host before calling the transport for the first, we previously failed
71 to close down TLS in the old transport process before passing the TCP
72 connection to the new process. The new one sent a STARTTLS command
73 which naturally failed, giving a failed delivery and bloating the retry
74 database. Investigation and fix prototype from Wolfgang Breyha.
76 JH/12 Fix check on SMTP command input synchronisation. Previously there were
77 false-negatives in the check that the sender had not preempted a response
78 or prompt from Exim (running as a server), due to that code's lack of
79 awareness of the SMTP input buffering.
81 PP/04 Add commandline_checks_require_admin option.
82 Exim drops privileges sanely, various checks such as -be aren't a
83 security problem, as long as you trust local users with access to their
84 own account. When invoked by services which pass untrusted data to
85 Exim, this might be an issue. Set this option in main configuration
86 AND make fixes to the calling application, such as using `--` to stop
89 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
90 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
91 responses to those, into a single TLS record each way (this usually means
92 a single packet). As a side issue, smtp_enforce_sync now works on TLS
95 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
96 affects you only if you're dancing at the edge of the param size limits.
97 If you are, and this message makes sense to you, then: raise the
98 configured limit or use OpenSSL 1.1. Nothing we can do for older
101 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
102 cmdline element to get the documented default one. Previously it was
105 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
106 are unsuable for non-message-specific reasons - eg. connection timeout,
107 banner-time rejection.
109 JH/16 Fix logging of delivery remote port, when specified by router, under
112 PP/06 Repair manualroute's ability to take options in any order, even if one
113 is the name of a transport.
116 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
118 JH/17 Change the list-building routines interface to use the expanding-string
119 triplet model, for better allocation and copying behaviour.
121 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
122 Previously it was constructed the first time a possibly-matching string
123 was met in the configuration file input during startup; now it is done
126 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
127 compatible one, to avoid the (poorly documented) possibility of a config
128 file in the working directory redirecting the DB files, possibly correpting
129 some existing file. CVE-2017-10140 assigned for BDB.
131 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
132 cache-hot. Previously, although the result was properly cached, the
133 initial verify call returned a defer.
135 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
136 the main verify for receipient in uncached-mode.
138 JH/22 Retire historical build files to an "unsupported" subdir. These are
139 defined as "ones for which we have no current evidence of testing".
141 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
142 if present. Previously it was ignored.
144 JH/24 Start using specified-initialisers in C structure init coding. This is
145 a C99 feature (it's 2017, so now considered safe).
147 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
148 if was a fixed-sized field and bitmask ops via macros; it is now more
151 PP/07 GitHub PR 56: Apply MariaDB build fix.
152 Patch provided by Jaroslav Škarvada.
154 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
155 during Coverity cleanups [4.87 JH/47]
156 Diagnosis and fix provided by Michael Fischer v. Mollard.
158 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
159 the right size to place the terminating semicolon on its own folded
160 line, the header hash was calculated to an incorrect value thanks to
161 the (relaxed) space the fold became.
163 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
164 and confused the parent.
166 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
167 which could crash as a result. This could lead to undeliverable messages.
169 JH/28 Logging: "next input sent too soon" now shows where input was truncated
172 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
173 matters on fast-turnover and PID-randomising systems, which were getting
174 out-of-order delivery.
176 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
177 a possibly-overlapping copy. The symptom was that "Remote host closed
178 connection in response to HELO" was logged instead of the actual 4xx
181 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
182 Previously only that bufferd was discarded, resulting in SYMTP command
185 JH/32 DKIM: when a message has multiple signatures matching an identity given
186 in dkim_verify_signers, run the dkim acl once for each. Previously only
187 one run was done. Bug 2189.
189 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
190 "panic the current process" to "deliberately defer". The panic log is
191 still written with the problem list name; the mail and reject logs now
192 get a temp-reject line for the message that was being handled, saying
193 something like "domains check lookup or other defer". The SMTP 451
194 message is still "Temporary local problem".
196 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
197 A crafted sequence of BDAT commands could result in in-use memory beeing
204 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
205 than -2003 did; needs libidn2 in addition to libidn.
207 JH/02 The path option on a pipe transport is now expanded before use.
209 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
210 Patch provided by "Björn", documentation fix added too.
212 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
213 missing a wire-to-host endian conversion.
215 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
216 close after a BDAT command line could be taken as a following command,
217 giving a synch failure. Fix by only checking for synch immediately
218 before acknowledging the chunk.
220 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
221 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
222 macro. Patches provided by Josh Soref.
224 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
225 Previously we did not; the RFC seems ambiguous and VRFY is not listed
226 by IANA as a service extension. However, John Klensin suggests that we
229 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
230 the dkim code may be unix-mode line endings rather than smtp wire-format
231 CRLF, so prepend a CR to any bare LF.
233 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
234 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
236 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
237 the first were themselves being wrongly included in the feed into dkim
238 processing; with most chunk sizes in use this resulted in an incorrect
239 body hash calculated value.
241 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
242 DKIM signature block, for verification. Although advised against by
243 standards it is specifically not ruled illegal.
245 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
247 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
248 missing a body hash (the bh= tag).
250 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
251 It seems that HAProxy sends the Proxy Protocol information in clear and
252 only then does a TLS startup, so do the same.
254 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
255 TCP connections (such as for Spamd) unless the daemon successfully set
256 Fast Open mode on its listening sockets. This fixes breakage seen on
257 too-old kernels or those not configured for Fast Open, at the cost of
258 requiring both directions being enabled for TFO, and TFO never being used
259 by non-daemon-related Exim processes.
261 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
262 endings, at least on the first header line. Try to canonify any that get
263 past that check, despite the cost.
265 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
266 now limited to an arbitrary five deep, while parsing addresses with the
267 strip_excess_angle_brackets option enabled.
269 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
270 instead leave the unprompted TLS handshake in socket buffer for the
271 TLS library to consume.
273 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
275 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
277 JH/16 Drop variables when they go out of scope. Memory management drops a whole
278 region in one operation, for speed, and this leaves assigned pointers
279 dangling. Add checks run only under the testsuite which checks all
280 variables at a store-reset and panics on a dangling pointer; add code
281 explicitly nulling out all the variables discovered. Fixes one known
282 bug: a transport crash, where a dangling pointer for $sending_ip_address
283 originally assigned in a verify callout, is re-used.
285 PP/06 Drop '.' from @INC in various Perl scripts.
287 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
289 PP/08 Reduce a number of compilation warnings under clang; building with
290 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
291 should be warning-free.
293 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
295 HS/01 Fix portability problems introduced by PP/08 for platforms where
296 realloc(NULL) is not equivalent to malloc() [SunOS et al].
298 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
299 chunk. This allows us to accept broken chunked messages. We need a more
300 general solution here.
302 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
303 already-broken messages in the queue.
305 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
307 JH/19 Fix reference counting bug in routing-generated-address tracking.
313 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
314 supports it and a size is available (ie. the sending peer gave us one).
316 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
317 years of being deprecated). The replacements are the ACLs
318 acl_smtp_mime and acl_not_smtp_mime.
320 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
321 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
322 or even an in-clear connection were permitted. Now, if the host lookup
323 was dnssec and dane was requested then the host is only used if the
324 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
325 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
326 if one fails this test.
327 This means that a poorly-configured remote DNS will make it incommunicado;
328 but it protects against a DNS-interception attack on it.
330 JH/04 Bug 1810: make continued-use of an open smtp transport connection
331 non-noisy when a race steals the message being considered.
333 JH/05 If main configuration option tls_certificate is unset, generate a
334 self-signed certificate for inbound TLS connections.
336 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
337 in rewrites and routers.
339 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
340 and logged a warning sing 4.83; now they are a configuration file error.
342 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
343 (lacking @domain). Apply the same qualification processing as RCPT.
345 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
347 JH/10 Support ${sha256:} applied to a string (as well as the previous
350 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
351 a cutthrough deliver is pending, as we always want to make a connection.
352 This also avoids re-routing the message when later placing the cutthrough
353 connection after a verify cache hit.
354 Do not update it with the verify result either.
356 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
357 when routing results in more than one destination address.
359 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
360 signing (which inhibits the cutthrough capability). Previously only
361 the presence of an option was tested; now an expansion evaluating as
362 empty is permissible (obviously it should depend only on data available
363 when the cutthrough connection is made).
365 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
366 the relevant preceding SMTP command did not note the pipelining mode.
368 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
369 Previously they were not counted.
371 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
372 as one having no matching records. Previously we deferred the message
373 that needed the lookup.
375 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
376 distinguished as "(=".
378 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
379 for missing MX records. Previously it only worked for missing A records.
381 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
383 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
384 after the data-go-ahead and data-ack. Patch from Jason Betts.
386 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
387 even for a "none" policy. Patch from Tony Meyer.
389 JH/22 Fix continued use of a connection for further deliveries. If a port was
390 specified by a router, it must also match for the delivery to be
393 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
394 When one of the recipients of a message was unsuitable for the connection
395 (has no matching addresses), we lost track of needing to mark it
396 deferred. As a result mail would be lost.
398 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
400 JH/25 Decoding ACL controls is now done using a binary search; the source code
401 takes up less space and should be simpler to maintain. Merge the ACL
402 condition decode tables also, with similar effect.
404 JH/26 Fix problem with one_time used on a redirect router which returned the
405 parent address unchanged. A retry would see the parent address marked as
406 delivered, so not attempt the (identical) child. As a result mail would
409 JH/27 Fix a possible security hole, wherein a process operating with the Exim
410 UID can gain a root shell. Credit to http://www.halfdog.net/ for
411 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
414 JH/28 Enable {spool,log} filesystem space and inode checks as default.
415 Main config options check_{log,spool}_{inodes,space} are now
416 100 inodes, 10MB unless set otherwise in the configuration.
418 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
419 Previously it only applied to the main-section connection policy
422 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
424 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
425 by me. Added RFC7919 DH primes as an alternative.
427 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
428 are not in the system include path.
430 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
431 GnuTLS, when a session startup failed (eg because the client disconnected)
432 Exim did stdio operations after fclose. This was exposed by a recent
433 change which nulled out the file handle after the fclose.
435 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
436 signed directly by the cert-signing cert, rather than an intermediate
437 OCSP-signing cert. This is the model used by LetsEncrypt.
439 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
441 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
442 an incoming connection.
444 HS/02 Bug 1802: Do not half-close the connection after sending a request
447 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
448 fallback to "prime256v1".
450 JH/34 SECURITY: Use proper copy of DATA command in error message.
451 Could leak key material. Remotely exploitable. CVE-2016-9963.
457 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
458 and 3.4.4 - once the server is enabled to respond to an OCSP request
459 it does even when not requested, resulting in a stapling non-aware
460 client dropping the TLS connection.
462 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
463 support variable-length bit vectors. No functional change.
465 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
466 The I= interface field on outgoing lines is now after the H= remote
467 host field, same as incoming lines. There is a separate
468 outgoing_interface log selector which allows you to disable the
471 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
472 If not running log_selector +smtp_connection the mainlog would be held
473 open indefinitely after a "too many connections" event, including to a
474 deleted file after a log rotate. Leave the per net connection logging
475 leaving it open for efficiency as that will be quickly detected by the
476 check on the next write.
478 HS/01 Bug 1671: Fix post transport crash.
479 Processing the wait-<transport> messages could crash the delivery
480 process if the message IDs didn't exist for some reason. When
481 using 'split_spool_directory=yes' the construction of the spool
482 file name failed already, exposing the same netto behaviour.
484 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
485 mime_regex ACL conditions.
487 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
488 to DSN fail messages (bounces): remote IP, remote greeting, remote response
489 to HELO, local diagnostic string.
491 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
492 log line to debug. Even when configured with a tls authenticator many
493 client connections are expected to not authenticate in this way, so
494 an authenticate fail is not an error.
496 HS/02 Add the Exim version string to the process info. This way exiwhat
497 gives some more detail about the running daemon.
499 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
500 matter for fast-change records such as DNSBLs.
502 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
503 retry record, even if constant. There may be multiple transports with
504 different interface settings and the retry behaviour needs to be kept
507 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
509 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
511 JH/10 Bug 840: fix log_defer_output option of pipe transport
513 JH/11 Bug 830: use same host for all RCPTS of a message, even under
514 hosts_randomize. This matters a lot when combined with mua_wrapper.
516 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
517 ${quote_pgsql:<string>} operator.
519 JH/13 Bug 1708: avoid misaligned access in cached lookup.
521 JH/14 Change header file name for freeradius-client. Relevant if compiling
522 with Radius support; from the Gentoo tree and checked under Fedora.
524 JH/15 Bug 1712: Introduce $prdr_requested flag variable
526 JH/16 Bug 1714: Permit an empty string as expansion result for transport
527 option transport_filter, meaning no filtering.
529 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
531 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
532 defaults to "*" (all hosts). The variable is now available when not built
533 with TLS, default unset, mainly to enable keeping the testsuite sane.
534 If a server certificate is not supplied (via tls_certificate) an error is
535 logged, and clients will find TLS connections fail on startup. Presumably
536 they will retry in-clear.
537 Packagers of Exim are strongly encouraged to create a server certificate
538 at installation time.
540 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
541 with the $config_file variable.
543 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
544 in transport context, after the attempt, and per-recipient. The latter type
545 is per host attempted. The event data is the error message, and the errno
546 information encodes the lookup type (A vs. MX) used for the (first) host,
547 and the trailing two digits of the smtp 4xx response.
549 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
550 to write to mainlog (or rejectlog, paniclog) in the window between file
551 creation and permissions/ownership being changed. Particularly affects
552 installations where exicyclog is run as root, rather than exim user;
553 result is that the running daemon panics and dies.
555 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
557 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
558 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
559 "pri" and "weight". Note that the previous implicit priority given by the
560 list order is no longer honoured.
562 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
565 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
566 by defining SUPPORT_SOCKS.
568 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
569 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
570 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
571 variables are renamed to proxy_{local,external}_{address,port}.
573 JH/27 Move Internationalisation support from Experimental to mainline, enabled
574 for a build by defining SUPPORT_I18N
576 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
577 of the query string, and make ${quote_redis:} do that quoting.
579 JH/29 Move Events support from Experimental to mainline, enabled by default
580 and removable for a build by defining DISABLE_EVENT.
582 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
584 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
585 cached by the daemon.
587 JH/32 Move Redis support from Experimental to mainline, enabled for a build
588 by defining LOOKUP_REDIS. The libhiredis library is required.
590 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
591 keys are given for lookup.
593 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
594 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
595 only supported when built with TLS support. The PolarSSL SHA routines
596 are still used when the TLS library is too old for convenient support.
598 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
599 openssl_options), for security. OpenSSL forces this from version 1.1.0
600 server-side so match that on older versions.
602 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
603 allocation for $value could be released as the expansion processing
604 concluded, but leaving the global pointer active for it.
606 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
607 and to use the domains and local_parts ACL conditions.
609 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
610 incorrectly not doubled on cutthrough transmission, hence seen as a
611 body-termination at the receiving system - resulting in truncated mails.
612 Commonly the sender saw a TCP-level error, and retransmitted the message
613 via the normal store-and-forward channel. This could result in duplicates
614 received - but deduplicating mailstores were liable to retain only the
615 initial truncated version.
617 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
619 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
621 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
622 we're in there, support oversigning also; bug 1309.
624 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
626 HS/04 Add support for keep_environment and add_environment options.
628 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
629 either intentional arithmetic overflow during PRNG, or testing config-
632 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
633 delivery resulted in actual delivery. Cancel cutthrough before DATA
636 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
637 rejects a recipient: pass the reject to the originator.
639 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
640 Many were false-positives and ignorable, but it's worth fixing the
643 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
644 for the new environment-manipulation done at startup. Move the routines
645 from being local to tls.c to being global via the os.c file.
647 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
648 an extract embedded as result-arg for a map, the first arg for extract
649 is unavailable so we cannot tell if this is a numbered or keyed
650 extraction. Accept either.
656 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
659 JH/02 The smtp transport option "multi_domain" is now expanded.
661 JH/03 The smtp transport now requests PRDR by default, if the server offers
664 JH/04 Certificate name checking on server certificates, when exim is a client,
665 is now done by default. The transport option tls_verify_cert_hostnames
666 can be used to disable this per-host. The build option
667 EXPERIMENTAL_CERTNAMES is withdrawn.
669 JH/05 The value of the tls_verify_certificates smtp transport and main options
670 default to the word "system" to access the system default CA bundle.
671 For GnuTLS, only version 3.0.20 or later.
673 JH/06 Verification of the server certificate for a TLS connection is now tried
674 (but not required) by default. The verification status is now logged by
675 default, for both outbound TLS and client-certificate supplying inbound
678 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
681 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
682 Status Notification (bounce) messages are now MIME format per RFC 3464.
683 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
684 under the control of the dsn_advertise_hosts option, and routers may
685 have a dsn_lasthop option.
687 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
688 default, modifiable by a malware= option. The list separator for
689 the options can now be changed in the usual way. Bug 68.
691 JH/10 The smtp_receive_timeout main option is now expanded before use.
693 JH/11 The incoming_interface log option now also enables logging of the
694 local interface on delivery outgoing connections.
696 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
697 if the interface and destination host and port all match.
699 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
702 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
703 Patch from Andrew Lewis.
705 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
706 now supports optional time-restrictions, weighting, and priority
707 modifiers per server. Patch originally by <rommer@active.by>.
709 JH/16 The spamd_address main option now supports a mixed list of local
710 and remote servers. Remote servers can be IPv6 addresses, and
711 specify a port-range.
713 JH/17 Bug 68: The spamd_address main option now supports an optional
714 timeout value per server.
716 JH/18 Bug 1581: Router and transport options headers_add/remove can
717 now have the list separator specified.
719 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
722 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
725 JH/21 Support for the A6 type of dns record is withdrawn.
727 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
728 rather than the verbs used.
730 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
731 from 255 to 1024 chars.
733 JH/24 Verification callouts now attempt to use TLS by default.
735 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
736 are generic router options now. The defaults didn't change.
738 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
739 Original patch from Alexander Shikoff, worked over by JH.
741 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
742 files if ZCAT_COMMAND is not executable.
744 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
746 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
748 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
749 Normally benign, it bites when the pair was led to by a CNAME;
750 modern usage is to not canonicalize the domain to a CNAME target
751 (and we were inconsistent anyway for A-only vs AAAA+A).
753 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
755 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
756 when evaluating $sender_host_dnssec.
758 JH/31 Check the HELO verification lookup for DNSSEC, adding new
759 $sender_helo_dnssec variable.
761 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
763 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
765 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
767 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
768 documented as working, but never had. Support all but $spam_report.
770 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
771 added for tls authenticator.
773 HS/03 Add perl_taintmode main config option
779 TL/01 When running the test suite, the README says that variables such as
780 no_msglog_check are global and can be placed anywhere in a specific
781 test's script, however it was observed that placement needed to be near
782 the beginning for it to behave that way. Changed the runtest perl
783 script to read through the entire script once to detect and set these
784 variables, reset to the beginning of the script, and then run through
785 the script parsing/test process like normal.
787 TL/02 The BSD's have an arc4random API. One of the functions to induce
788 adding randomness was arc4random_stir(), but it has been removed in
789 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
790 function when detected.
792 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
793 cause callback expansion.
795 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
796 syntax errors in an expansion can be treated as a string instead of
797 logging or causing an error, due to the internal use of bool_lax
798 instead of bool when processing it.
800 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
801 server certificates when making smtp deliveries.
803 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
805 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
807 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
809 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
810 Merged patch from Sebastian Wiedenroth.
812 JH/05 Fix results-pipe from transport process. Several recipients, combined
813 with certificate use, exposed issues where response data items split
814 over buffer boundaries were not parsed properly. This eventually
815 resulted in duplicates being sent. This issue only became common enough
816 to notice due to the introduction of connection certificate information,
817 the item size being so much larger. Found and fixed by Wolfgang Breyha.
819 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
820 size buffer was used, resulting in syntax errors when an expansion
823 JH/07 Add support for directories of certificates when compiled with a GnuTLS
824 version 3.3.6 or later.
826 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
827 is EXPERIMENTAL_EVENT, the main-configuration and transport options
828 both become "event_action", the variables become $event_name, $event_data
829 and $event_defer_errno. There is a new variable $verify_mode, usable in
830 routers, transports and related events. The tls:cert event is now also
831 raised for inbound connections, if the main configuration event_action
834 TL/06 In test suite, disable OCSP for old versions of openssl which contained
835 early OCSP support, but no stapling (appears to be less than 1.0.0).
837 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
838 server certificate names available under the smtp transport option
839 "tls_verify_cert_hostname" now do not permit multi-component wildcard
842 JH/10 Time-related extraction expansions from certificates now use the main
843 option "timezone" setting for output formatting, and are consistent
844 between OpenSSL and GnuTLS compilations. Bug 1541.
846 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
847 encoded parameter in the incoming message. Bug 1558.
849 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
850 include certificate info, eximon was claiming there were spoolfile
853 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
855 JH/14 Log delivery-related information more consistently, using the sequence
856 "H=<name> [<ip>]" wherever possible.
858 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
859 are problematic for Debian distribution, omit them from the release
862 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
864 JH/16 Fix string representation of time values on 64bit time_t architectures.
867 JH/17 Fix a null-indirection in certextract expansions when a nondefault
868 output list separator was used.
873 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
874 checkers that were complaining about end of non-void function with no
877 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
878 This was a regression introduced in 4.83 by another bugfix.
880 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
882 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
883 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
889 TF/01 Correctly close the server side of TLS when forking for delivery.
891 When a message was received over SMTP with TLS, Exim failed to clear up
892 the incoming connection properly after forking off the child process to
893 deliver the message. In some situations the subsequent outgoing
894 delivery connection happened to have the same fd number as the incoming
895 connection previously had. Exim would try to use TLS and fail, logging
896 a "Bad file descriptor" error.
898 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
899 utilities have not been installed.
901 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
902 temporary space as the ACL may create new global variables.
904 TL/01 LDAP support uses per connection or global context settings, depending
905 upon the detected version of the libraries at build time.
907 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
908 to extract and use the src ip:port in logging and expansions as if it
909 were a direct connection from the outside internet. PPv2 support was
910 updated based on HAProxy spec change in May 2014.
912 JH/02 Add ${listextract {number}{list}{success}{fail}}.
914 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
915 Properly escape header and check for NULL return.
917 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
920 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
922 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
923 characters in header names, implemented as a verify condition.
924 Contributed by Michael Fischer v. Mollard.
926 TL/05 Rename SPF condition results err_perm and err_temp to standardized
927 results permerror and temperror. Previous values are deprecated but
928 still accepted. In a future release, err_perm and err_temp will be
929 completely removed, which will be a backward incompatibility if the
930 ACL tests for either of these two old results. Patch contributed by
931 user bes-internal on the mailing list.
933 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
935 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
936 selectors, in both main and reject logs.
938 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
941 JH/07 Add malware type "sock" for talking to simple daemon.
943 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
945 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
946 routers/transports under cutthrough routing.
948 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
949 numbers. Touch up "bool" conditional to keep the same definition.
951 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
953 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
955 JH/12 Expand items in router/transport headers_add or headers_remove lists
956 individually rather than the list as a whole. Bug 1452.
958 Required for reasonable handling of multiple headers_ options when
959 they may be empty; requires that headers_remove items with embedded
960 colons must have them doubled (or the list-separator changed).
962 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
963 view the policy declared in the DMARC record. Currently, $dmarc_status
964 is a combined value of both the record presence and the result of the
967 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
969 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
970 dnslookup router and the smtp transport (applying to the forward
973 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
974 of ldap servers used for a specific lookup. Patch provided by Heiko
977 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
978 New variable $lookup_dnssec_authenticated for observability.
980 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
981 Patch submitted by Lars Timman.
983 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
985 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
986 Requires trusted mode and valid format message id, aborts otherwise.
987 Patch contributed by Heiko Schlichting.
989 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
990 certextract with support for various fields. Bug 1358.
992 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
993 is requested by default, modifiable by smtp transport option
996 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
997 operate on certificate variables to give certificate fingerprints
998 Also new ${sha256:cert_variable}.
1000 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1002 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1003 Christian Aistleitner.
1005 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1007 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1008 file. Patch from Wolfgang Breyha.
1010 JH/25 Expand the coverage of the delivery $host and $host_address to
1011 client authenticators run in verify callout. Bug 1476.
1013 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1014 align with daemon_smtp_ports. Bug 72.
1016 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1017 support and error reporting did not work properly.
1019 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1020 and is readable. Patch from Andrew Colin Kissa.
1022 TL/14 Enhance documentation of ${run expansion and how it parses the
1023 commandline after expansion, particularly in the case when an
1024 unquoted variable expansion results in an empty value.
1026 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1028 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1029 by renaming away from it. Addresses GH issue 15, reported by
1032 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1034 TL/15 SECURITY: prevent double expansion in math comparison functions
1035 (can expand unsanitized data). Not remotely exploitable.
1042 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1044 PP/02 Make -n do something, by making it not do something.
1045 When combined with -bP, the name of an option is not output.
1047 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1050 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1051 $sender_host_name and config options to manage this, and basic check
1054 PP/05 DSCP support for outbound connections and control modifier for inbound.
1056 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1057 (Only plugin which currently uses this is kerberos4, which nobody should
1058 be using, but we should make it available and other future plugins might
1059 conceivably use it, even though it would break NAT; stuff *should* be
1060 using channel bindings instead).
1062 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1063 name; added for Sendmail compatibility; requires admin caller.
1064 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1065 just ignore it); requires trusted caller.
1066 Also parse but ignore: -Ac -Am -X<logfile>
1069 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1071 TL/02 Add +smtp_confirmation as a default logging option.
1073 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1074 Patch by Magnus Holmgren from 2007-02-20.
1076 TL/04 Bugzilla 1281 - Spec typo.
1077 Bugzilla 1283 - Spec typo.
1078 Bugzilla 1290 - Spec grammar fixes.
1080 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1082 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1084 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1085 mailing list by by Dmitry Isaikin.
1087 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1089 JH/02 Support "G" suffix to numbers in ${if comparisons.
1091 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1093 NM/01 Bugzilla 1197 - Spec typo
1094 Bugzilla 1196 - Spec examples corrections
1096 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1098 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1099 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1102 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1103 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1105 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1106 "acl {{name}{arg}...}", and optional args on acl condition
1109 JH/05 Permit multiple router/transport headers_add/remove lines.
1111 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1113 JH/07 Avoid using a waiting database for a single-message-only transport.
1114 Performance patch from Paul Fisher. Bugzilla 1262.
1116 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1119 JH/09 Add $headers_added variable, with content from use of ACL modifier
1120 add_header (but not yet added to the message). Bugzilla 199.
1122 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1123 Pulled from Bugzilla 817 by Wolfgang Breyha.
1125 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1127 (nb: this is the same fix as in Exim 4.80.1)
1129 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1132 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1135 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1137 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1138 advertise SMTP AUTH mechanism to us, instead of a generic
1139 protocol violation error. Also, make Exim more robust to bad
1140 data from the Dovecot auth socket.
1142 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1144 When a queue runner is handling a message, Exim first routes the
1145 recipient addresses, during which it prunes them based on the retry
1146 hints database. After that it attempts to deliver the message to
1147 any remaining recipients. It then updates the hints database using
1150 So if a recipient address works intermittently, it can get repeatedly
1151 deferred at routing time. The retry hints record remains fresh so the
1152 address never reaches the final cutoff time.
1154 This is a fairly common occurrence when a user is bumping up against
1155 their storage quota. Exim had some logic in its local delivery code
1156 to deal with this. However it did not apply to per-recipient defers
1157 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1159 This change adds a proper retry rule check during routing so that the
1160 final cutoff time is checked against the message's age. We only do
1161 this check if there is an address retry record and there is not a
1162 domain retry record; this implies that previous attempts to handle
1163 the address had the retry_use_local_parts option turned on. We use
1164 this as an approximation for the destination being like a local
1165 delivery, as in LMTP.
1167 I suspect this new check makes the old local delivery cutoff check
1168 redundant, but I have not verified this so I left the code in place.
1170 TF/02 Correct gecos expansion when From: is a prefix of the username.
1172 Test 0254 submits a message to Exim with the header
1176 When I ran the test suite under the user fanf2, Exim expanded
1177 the header to contain my full name, whereas it should have added
1178 a Resent-Sender: header. It erroneously treats any prefix of the
1179 username as equal to the username.
1181 This change corrects that bug.
1183 GF/01 DCC debug and logging tidyup
1184 Error conditions log to paniclog rather than rejectlog.
1185 Debug lines prefixed by "DCC: " to remove any ambiguity.
1187 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1189 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1190 Bug spotted by Jeremy Harris; was flawed since initial commit.
1191 Would have resulted in OCSP responses post-SNI triggering an Exim
1192 NULL dereference and crash.
1194 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1196 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1197 Bug detection, analysis and fix by Samuel Thibault.
1198 Bugzilla 1331, Debian bug #698092.
1200 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1202 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1203 Server implementation by Todd Lyons, client by JH.
1204 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1205 config variable "prdr_enable" controls whether the server
1206 advertises the facility. If the client requests PRDR a new
1207 acl_data_smtp_prdr ACL is called once for each recipient, after
1208 the body content is received and before the acl_smtp_data ACL.
1209 The client is controlled by both of: a hosts_try_prdr option
1210 on the smtp transport, and the server advertisement.
1211 Default client logging of deliveries and rejections involving
1212 PRDR are flagged with the string "PRDR".
1214 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1215 fclose(). Diagnosis by Todd Lyons.
1217 PP/17 Update configure.default to handle IPv6 localhost better.
1218 Patch by Alain Williams (plus minor tweaks).
1221 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1222 This is now consistent with GnuTLS, and is now documented: the
1223 previous undocumented portable approach to treating the option as
1224 unset was to force an expansion failure. That still works, and
1225 an empty string is now equivalent.
1227 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1228 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1229 not performing validation itself.
1231 PP/20 Added force_command boolean option to pipe transport.
1232 Patch from Nick Koston, of cPanel Inc.
1234 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1237 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1239 PP/21 Fix eximon continuous updating with timestamped log-files.
1240 Broken in a format-string cleanup in 4.80, missed when I repaired the
1241 other false fix of the same issue.
1242 Report and fix from Heiko Schlichting.
1245 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1246 Report from Prashanth Katuri.
1248 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1249 It's SecureTransport, so affects any MacOS clients which use the
1250 system-integrated TLS libraries, including email clients.
1252 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1253 using a MIME ACL for non-SMTP local injection.
1254 Report and assistance in diagnosis by Warren Baker.
1256 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1258 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1260 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1261 last id that failed so it may be referenced in subsequent ACL's.
1263 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1266 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1267 ldap library initialization, allowing self-signed CA's to be
1268 used. Also properly sets require_cert option later in code by
1269 using NULL (global ldap config) instead of ldap handle (per
1270 session). Bug diagnosis and testing by alxgomz.
1272 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1273 the src/util/ subdirectory.
1275 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1276 renamed to Transport Post Delivery Action by Jeremy Harris, as
1279 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1280 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1281 redis_servers = needs to be configured which will be used by the redis
1282 lookup. Patch from Warren Baker, of The Packet Hub.
1284 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1286 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1287 hostname or reverse DNS when processing a host list. Used suggestions
1288 from multiple comments on this bug.
1290 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1292 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1293 Missed a few lines, added it to make the runtest require no keyboard
1296 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1297 contains upper case chars. Make router use caseful_local_part.
1299 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1300 support when GnuTLS has been built with p11-kit.
1306 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1308 This, or similar/improved, will also be change PP/11 of 4.82.
1314 PP/01 Handle short writes when writing local log-files.
1315 In practice, only affects FreeBSD (8 onwards).
1316 Bugzilla 1053, with thanks to Dmitry Isaikin.
1318 NM/01 Bugzilla 949 - Documentation tweak
1320 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1323 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1325 PP/02 Implemented gsasl authenticator.
1327 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1329 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1330 `pkg-config foo` for cflags/libs.
1332 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1333 with rest of GSASL and with heimdal_gssapi.
1335 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1336 `pkg-config foo` for cflags/libs for the TLS implementation.
1338 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1339 properties get this fed in as external SSF. A number of robustness
1340 and debugging improvements to the cyrus_sasl authenticator.
1342 PP/08 cyrus_sasl server now expands the server_realm option.
1344 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1345 Patch by Jeremy Harris.
1347 PP/10 Added dbmjz lookup type.
1349 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1351 PP/12 MAIL args handles TAB as well as SP, for better interop with
1352 non-compliant senders.
1353 Analysis and variant patch by Todd Lyons.
1355 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1356 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1357 Patch from Dirk Mueller <dmueller@suse.com>
1359 PP/13 tls_peerdn now print-escaped for spool files.
1360 Observed some $tls_peerdn in wild which contained \n, which resulted
1361 in spool file corruption.
1363 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1364 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1365 or write after TLS renegotiation, which otherwise led to messages
1368 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1369 as a tracking header (ie: a signed header comes before the signature).
1370 Patch from Wolfgang Breyha.
1372 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1373 comma-sep list; embedded commas doubled.
1375 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1377 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1379 Report and patch from Dmitry Banschikov.
1381 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1382 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1383 failures. We appear to now support TLS1.1+ with Exim.
1385 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1386 lets Exim select keys and certificates based upon TLS SNI from client.
1387 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1388 before an outbound SMTP session. New log_selector, +tls_sni.
1390 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1391 NULL dereference. Report and patch from Alun Jones.
1393 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1394 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1395 Not seeing resolver debug output on NetBSD, but suspect this is a
1396 resolver implementation change.
1398 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1399 Left warnings. Added "eximon gdb" invocation mode.
1401 PP/21 Defaulting "accept_8bitmime" to true, not false.
1403 PP/22 Added -bw for inetd wait mode support.
1405 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1406 locate the relevant includes and libraries. Made this the default.
1408 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1409 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1411 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1412 This may cause build issues on older platforms.
1414 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1415 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1416 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1417 Added SNI support via GnuTLS too.
1418 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1420 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1422 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1423 Applied second patch from Janne, implementing suggestion to default
1424 multiple-strings-in-record handling to match SPF spec.
1426 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1428 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1429 read-only, out of scope).
1430 Patch from Wolfgang Breyha, report from Stuart Northfield.
1432 PP/29 Fix three issues highlighted by clang analyser static analysis.
1433 Only crash-plausible issue would require the Cambridge-specific
1434 iplookup router and a misconfiguration.
1435 Report from Marcin Mirosław.
1437 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1439 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1440 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1441 As part of this, removing so much warning spew let me fix some minor
1442 real issues in debug logging.
1444 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1445 assignment on my part. Fixed.
1447 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1448 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1449 Janne Snabb (who went above and beyond: thank you).
1451 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1452 string otherwise requires a connection and a bunch more work and it's
1453 relatively easy to get wrong. Should also expose TLS library linkage
1456 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1457 64-bit ${eval} (JH/03).
1459 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1460 GNU libc to support some of the 64-bit stuff, should not lead to
1461 conflicts. Defined before os.h is pulled in, so if a given platform
1462 needs to override this, it can.
1464 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1465 protection layer was required, which is not implemented.
1466 Bugzilla 1254, patch from Wolfgang Breyha.
1468 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1469 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1470 tls_dhparam take prime identifiers. Also unbreak combination of
1471 OpenSSL+DH_params+TLSSNI.
1473 PP/39 Disable SSLv2 by default in OpenSSL support.
1479 PP/01 Solaris build fix for Oracle's LDAP libraries.
1480 Bugzilla 1109, patch from Stephen Usher.
1482 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1484 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1487 TF/02 Fix a couple more cases where we did not log the error message
1488 when unlink() failed. See also change 4.74-TF/03.
1490 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1491 lock up or crash if it happened to be inside a call to libc when it
1492 got a SIGUSR1 from exiwhat.
1494 The SIGUSR1 handler appends the current process status to the process
1495 log which is later printed by exiwhat. It used to use the general
1496 purpose logging code to do this, but several functions it calls are
1497 not safe for signals.
1499 The new output code in the SIGUSR1 handler is specific to the process
1500 log, and simple enough that it's easy to inspect for signal safety.
1501 Removing some special cases also simplifies the general logging code.
1502 Removing the spurious timestamps from the process log simplifies
1505 TF/04 Improved ratelimit ACL condition.
1507 The /noupdate option has been deprecated in favour of /readonly which
1508 has clearer semantics. The /leaky, /strict, and /readonly update modes
1509 are mutually exclusive. The update mode is no longer included in the
1510 database key; it just determines when the database is updated. (This
1511 means that when you upgrade Exim will forget old rate measurements.)
1513 Exim now checks that the per_* options are used with an update mode that
1514 makes sense for the current ACL. For example, when Exim is processing a
1515 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1516 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1517 must specify per_mail/readonly. If you omit the update mode it defaults to
1518 /leaky where that makes sense (as before) or /readonly where required.
1520 The /noupdate option is now undocumented but still supported for
1521 backwards compatibility. It is equivalent to /readonly except that in
1522 ACLs where /readonly is required you may specify /leaky/noupdate or
1523 /strict/noupdate which are treated the same as /readonly.
1525 A useful new feature is the /count= option. This is a generalization
1526 of the per_byte option, so that you can measure the throughput of other
1527 aggregate values. For example, the per_byte option is now equivalent
1528 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1530 The per_rcpt option has been generalized using the /count= mechanism
1531 (though it's more complicated than the per_byte equivalence). When it is
1532 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1533 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1534 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1535 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1536 Note that using per_rcpt with a non-readonly update mode in more than
1537 one ACL will cause the recipients to be double-counted. (The per_mail
1538 and per_byte options don't have this problem.)
1540 The handling of very low rates has changed slightly. If the computed rate
1541 is less than the event's count (usually one) then this event is the first
1542 after a long gap. In this case the rate is set to the same as this event's
1543 count, so that the first message of a spam run is counted properly.
1545 The major new feature is a mechanism for counting the rate of unique
1546 events. The new per_addr option counts the number of different
1547 recipients that someone has sent messages to in the last time period. It
1548 behaves like per_rcpt if all the recipient addresses are different, but
1549 duplicate recipient addresses do not increase the measured rate. Like
1550 the /count= option this is a general mechanism, so the per_addr option
1551 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1552 example, measure the rate that a client uses different sender addresses
1553 with the options per_mail/unique=$sender_address. There are further
1554 details in the main documentation.
1556 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1558 TF/06 Removed a few PCRE remnants.
1560 TF/07 Automatically extract Exim's version number from tags in the git
1561 repository when doing development or release builds.
1563 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1564 Bugzilla 879. Patch from Paul Fisher.
1566 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1567 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1570 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1572 PP/05 Variable $av_failed, true if the AV scanner deferred.
1573 Bugzilla 1078. Patch from John Horne.
1575 PP/06 Stop make process more reliably on build failure.
1576 Bugzilla 1087. Patch from Heiko Schlittermann.
1578 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1579 Bugzilla 1089. Patch from Heiko Schlittermann.
1581 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1582 Bugzilla 1131. Patch from Holger Weiß.
1584 PP/09 Handle IPv6 addresses with SPF.
1585 Bugzilla 860. Patch from Wolfgang Breyha.
1587 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1589 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1592 PP/11 match_* no longer expand right-hand-side by default.
1593 New compile-time build option, EXPAND_LISTMATCH_RHS.
1594 New expansion conditions, "inlist", "inlisti".
1596 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1598 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1600 PP/14 fix log_write() format string regression from TF/03.
1601 Bugzilla 1152. Patch from Dmitry Isaikin.
1607 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1609 PP/02 Harmonised TLS library version reporting; only show if debugging.
1610 Layout now matches that introduced for other libraries in 4.74 PP/03.
1612 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1614 PP/04 New "dns_use_edns0" global option.
1616 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1619 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1620 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1622 TK/01 Updated PolarSSL code to 0.14.2.
1623 Bugzilla 1097. Patch from Andreas Metzler.
1625 PP/07 Catch divide-by-zero in ${eval:...}.
1626 Fixes bugzilla 1102.
1628 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1631 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1632 format-string attack -- SECURITY: remote arbitrary code execution.
1634 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1635 time unintentionally subject to list matching rules, letting the header
1636 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1637 arbitrary string expansion). This allowed for information disclosure.
1639 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1640 INT_MIN/-1 -- value coerced to INT_MAX.
1646 NM/01 Workaround for PCRE version dependency in version reporting
1649 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1650 This fixes portability to compilers other than gcc, notably
1651 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1653 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1654 makefiles for portability to HP-UX and POSIX correctness.
1656 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1657 Also via indented variable definition in the Makefile.
1658 (Debugging by Oliver Heesakkers).
1660 PP/02 Restore caching of spamd results with expanded spamd_address.
1661 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1663 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1664 Improves build reliability. Fix from: Frank Elsner
1666 NM/02 Fix wide character breakage in the rfc2047 coding
1667 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1669 NM/03 Allow underscore in dnslist lookups
1670 Fixes bug 1026. Patch from Graeme Fowler
1672 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1673 Code patches from Adam Ciarcinski of NetBSD.
1675 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1678 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1679 is logged, to avoid truncation. Patch from John Horne.
1681 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1682 Patch from Jakob Hirsch.
1684 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1685 SQL string expansion failure details.
1686 Patch from Andrey Oktyabrski.
1688 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1689 Patch from Simon Arlott.
1691 PP/09 New lookups functionality failed to compile on old gcc which rejects
1692 extern declarations in function scope.
1693 Patch from Oliver Fleischmann
1695 PP/10 Use sig_atomic_t for flags set from signal handlers.
1696 Check getgroups() return and improve debugging.
1697 Fixed developed for diagnosis in bug 927 (which turned out to be
1700 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1701 Patch from Mark Zealey.
1703 PP/12 Bugzilla 1056: Improved spamd server selection.
1704 Patch from Mark Zealey.
1706 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1707 Based on patch from Heiko Schlittermann.
1709 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1710 Patch from Uwe Doering, sign-off by Michael Haardt.
1712 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1713 variable declaration deep within a block. Bug and patch from
1716 PP/15 lookups-Makefile IRIX compatibility coercion.
1718 PP/16 Make DISABLE_DKIM build knob functional.
1720 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1721 Patch by Simon Arlott
1723 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1724 variable argument macros. Our copy now differs from upstream.
1730 TF/01 Failure to get a lock on a hints database can have serious
1731 consequences so log it to the panic log.
1733 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1734 controlled using the smtp_confirmation log selector.
1736 TF/03 Include the error message when we fail to unlink a spool file.
1738 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1739 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1740 for maintaining out-of-tree patches for some time.
1742 PP/01 Bugzilla 139: Documentation and portability issues.
1743 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1744 Handle per-OS dynamic-module compilation flags.
1746 PP/02 Let /dev/null have normal permissions.
1747 The 4.73 fixes were a little too stringent and complained about the
1748 permissions on /dev/null. Exempt it from some checks.
1749 Reported by Andreas M. Kirchwitz.
1751 PP/03 Report version information for many libraries, including
1752 Exim version information for dynamically loaded libraries. Created
1753 version.h, now support a version extension string for distributors
1754 who patch heavily. Dynamic module ABI change.
1756 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1757 privilege escalation vulnerability whereby the Exim run-time user
1758 can cause root to append content of the attacker's choosing to
1761 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1764 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1765 If dropping privileges for untrusted macros, we disabled normal logging
1766 on the basis that it would fail; for the Exim run-time user, this is not
1767 the case, and it resulted in successful deliveries going unlogged.
1768 Fixed. Reported by Andreas Metzler.
1774 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1775 only prepend for the Resent-* case. Fixes regression introduced in
1776 Exim 4.70 by NM/22 for Bugzilla 607.
1778 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1779 increasing numbers of administrators be bitten by this.
1781 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1783 PP/03 Bugzilla 994: added openssl_options main configuration option.
1785 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1787 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1789 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1791 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1792 without a peer certificate, leading to a segfault because of an
1793 assumption that peers always have certificates. Be a little more
1794 paranoid. Problem reported by Martin Tscholak.
1796 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1797 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1798 NB: ClamAV planning to remove STREAM in "middle of 2010".
1799 CL also introduces -bmalware, various -d+acl logging additions and
1800 more caution in buffer sizes.
1802 PP/09 Implemented reverse_ip expansion operator.
1804 PP/10 Bugzilla 937: provide a "debug" ACL control.
1806 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1808 PP/12 Bugzilla 973: Implement --version.
1810 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1812 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1814 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1816 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1817 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1818 ignore trailing whitespace.
1820 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1822 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1825 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1826 Notification from Dr Andrew Aitchison.
1828 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1829 ExtendedDetectionInfo response format.
1830 Notification from John Horne.
1832 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1835 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1836 XSL and documented dependency on system catalogs, with examples of how
1839 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1842 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1843 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1844 configuration file which is writeable by the Exim user or group.
1846 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1847 of configuration files to cover files specified with the -C option if
1848 they are going to be used with root privileges, not just the default
1851 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1852 option (effectively making it always true).
1854 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1855 files to be used while preserving root privileges.
1857 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1858 that rogue child processes cannot use them.
1860 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1861 run-time user, instead of root.
1863 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1864 Exim run-time user without dropping privileges.
1866 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1867 result string, instead of calling string_vformat() twice with the same
1870 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1871 for other users. Others should always drop root privileges if they use
1872 -C on the command line, even for a whitelisted configure file.
1874 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1876 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1882 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1883 $data_path, and $header_path variables; fixed documentation bugs and
1886 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1887 exipick to access non-standard spools, including the "frozen" queue
1890 NM/01 Bugzilla 965: Support mysql stored procedures.
1891 Patch from Alain Williams
1893 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1895 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1896 Patch from Andreas Metzler
1898 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1899 Patch from Kirill Miazine
1901 NM/05 Bugzilla 671: Added umask to procmail example.
1903 JJ/03 installed exipick 20100323.0, fixing doc bug
1905 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1906 directory. Notification and patch from Dan Rosenberg.
1908 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1910 TK/02 Improve log output when DKIM signing operation fails.
1912 MH/01 Treat the transport option dkim_domain as a colon separated
1913 list, not as a single string, and sign the message with each element,
1914 omitting multiple occurences of the same signer.
1916 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1917 Bugzilla 985, 986. Patch by Simon Arlott
1919 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1920 Patch by Simon Arlott
1922 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1923 MBX locking. Notification from Dan Rosenberg.
1929 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1931 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1933 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1935 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1937 NM/04 Bugzilla 915: Flag broken perl installation during build.
1943 TK/01 Added patch by Johannes Berg that expands the main option
1944 "spamd_address" if it starts with a dollar sign.
1946 TK/02 Write list of recipients to X-Envelope-Sender header when building
1947 the mbox-format spool file for content scanning (suggested by Jakob
1950 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1951 (http://www.dcc-servers.net/) support via dccifd. Activated by
1952 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1954 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1955 by Mark Daniel Reidel <mr@df.eu>.
1957 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1958 When building exim an external PCRE library is now needed -
1959 PCRE is a system library on the majority of modern systems.
1960 See entry on PCRE_LIBS in EDITME file.
1962 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1963 conversation. Added nologin parameter to request.
1964 Patch contributed by Kirill Miazine.
1966 TF/01 Do not log submission mode rewrites if they do not change the address.
1968 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1970 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1971 log files in place. Contributed by Roberto Lima.
1973 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1975 TF/03 Bugzilla 615: When checking the local_parts router precondition
1976 after a local_part_suffix or local_part_prefix option, Exim now
1977 does not use the address's named list lookup cache, since this
1978 contains cached lookups for the whole local part.
1980 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1981 Robert Millan. Documentation is in experimental-spec.txt.
1983 TF/04 Bugzilla 668: Fix parallel build (make -j).
1985 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1987 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1988 Patch provided by Jan Srzednicki.
1990 TF/05 Leading white space used to be stripped from $spam_report which
1991 wrecked the formatting. Now it is preserved.
1993 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1994 that they are available at delivery time.
1996 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1998 TF/08 TLS error reporting now respects the incoming_interface and
1999 incoming_port log selectors.
2001 TF/09 Produce a more useful error message if an SMTP transport's hosts
2002 setting expands to an empty string.
2004 NM/06 Bugzilla 744: EXPN did not work under TLS.
2005 Patch provided by Phil Pennock.
2007 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2008 Patch provided by Richard Godbee.
2010 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2011 acl_smtp_notquit, added index entry.
2013 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2014 Patch provided by Eugene Bujak.
2016 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2017 accept(). Patch provided by Maxim Dounin.
2019 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2020 Patch provided by Phil Pennock.
2022 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2024 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2025 Patch provided by Brad "anomie" Jorsch.
2027 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2028 Patch provided by Dean Brooks.
2030 TK/05 Add native DKIM support (does not depend on external libraries).
2032 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2033 Patch provided by Graeme Fowler.
2035 NM/16 Bugzilla 851: Documentation example syntax fix.
2037 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2039 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2042 NM/19 Bugzilla 745: TLS version reporting.
2043 Patch provided by Phil Pennock.
2045 NM/20 Bugzilla 167: bool: condition support.
2046 Patch provided by Phil Pennock.
2048 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2049 clients. Patch provided by Phil Pennock.
2051 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2052 Patch provided by Brad "anomie" Jorsch.
2054 NM/23 Bugzilla 687: Fix misparses in eximstats.
2055 Patch provided by Heiko Schlittermann.
2057 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2058 Patch provided by Heiko Schlittermann.
2060 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2061 plus update to original patch.
2063 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2065 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2066 Patch provided by David Brownlee.
2068 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2070 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2072 NM/30 Bugzilla 888: TLS documentation bugfixes.
2074 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2076 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2077 Unlike the original bugzilla I have changed all shell scripts in src tree.
2079 NM/33 Bugzilla 898: Transport filter timeout fix.
2080 Patch by Todd Rinaldo.
2082 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2083 Patch by Serge Demonchaux.
2085 NM/35 Bugzilla 39: Base64 decode bug fixes.
2086 Patch by Jakob Hirsch.
2088 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2090 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2092 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2094 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2100 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2101 ALT-N's libdkim that I have put here:
2102 http://duncanthrax.net/exim-experimental/
2104 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2105 were called 'true' and it seems that C99 defines that as a reserved
2106 keyword to be used with 'bool' variable types. That means you could
2107 not include C99-style headers which use bools without triggering
2108 build errors in sieve.c.
2110 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2111 as mailq or other aliases. Changed the --help handling significantly
2112 to do whats expected. exim_usage() emits usage/help information.
2114 SC/01 Added the -bylocaldomain option to eximstats.
2116 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2118 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2120 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2126 PH/01 Another patch from the Sieve maintainer.
2128 PH/02 When an IPv6 address is converted to a string for single-key lookup
2129 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2130 dots are used instead of colons so that keys in lsearch files need not
2131 contain colons. This was done some time before quoting was made available
2132 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2133 (notated using the quote facility) so as to distinguish them from IPv4
2134 keys. This meant that lookups for IP addresses in host lists did not work
2135 for iplsearch lookups.
2137 This has been fixed by arranging for IPv6 addresses to be expressed with
2138 colons if the lookup type is iplsearch. This is not incompatible, because
2139 previously such lookups could never work.
2141 The situation is now rather anomalous, since one *can* have colons in
2142 ordinary lsearch keys. However, making the change in all cases is
2143 incompatible and would probably break a number of configurations.
2145 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2148 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2149 conversion specification without a maximum field width, thereby enabling
2150 a rogue spamd server to cause a buffer overflow. While nobody in their
2151 right mind would setup Exim to query an untrusted spamd server, an
2152 attacker that gains access to a server running spamd could potentially
2153 exploit this vulnerability to run arbitrary code as the Exim user.
2155 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2156 $primary_hostname instead of what libspf2 thinks the hosts name is.
2158 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2159 a directory entry by the name of the lookup key. Previously, if a
2160 symlink pointed to a non-existing file or a file in a directory that
2161 Exim lacked permissions to read, a lookup for a key matching that
2162 symlink would fail. Now it is enough that a matching directory entry
2163 exists, symlink or not. (Bugzilla 503.)
2165 PH/03 The body_linecount and body_zerocount variables are now exported in the
2168 PH/04 Added the $dnslist_matched variable.
2170 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2171 This means they are set thereafter only if the connection becomes
2174 PH/06 Added the client_condition to authenticators so that some can be skipped
2175 by clients under certain conditions.
2177 PH/07 The error message for a badly-placed control=no_multiline_responses left
2178 "_responses" off the end of the name.
2180 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2182 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2183 (without spaces) instead of just copying the configuration text.
2185 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2187 PH/11 Added $max_received_linelength.
2189 PH/12 Added +ignore_defer and +include_defer to host lists.
2191 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2192 way in which PCRE > 7.0 is built.
2194 PH/14 Implemented queue_only_load_latch.
2196 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2197 MAIL command. The effect was to mangle the value on 64-bit systems.
2199 PH/16 Another patch from the Sieve maintainer.
2201 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2203 PH/18 If a system quota error occurred while trying to create the file for
2204 a maildir delivery, the message "Mailbox is full" was not appended to the
2205 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2206 only to a quota excession during the actual writing of the file.
2208 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2209 characters?) which causes problems in log lines. The DN values are now
2210 passed through string_printing() before being added to log lines.
2212 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2213 and InterBase are left for another time.)
2215 PH/21 Added message_body_newlines option.
2217 PH/22 Guard against possible overflow in moan_check_errorcopy().
2219 PH/23 POSIX allows open() to be a macro; guard against that.
2221 PH/24 If the recipient of an error message contained an @ in the local part
2222 (suitably quoted, of course), incorrect values were put in $domain and
2223 $local_part during the evaluation of errors_copy.
2229 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2230 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2233 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2234 issue a MAIL command.
2236 PH/02 In an ACL statement such as
2238 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2240 if a client was not listed at all, or was listed with a value other than
2241 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2242 the condition was not true (as it should be), so access was not denied.
2243 The bug was that the ! inversion was incorrectly passed on to the second
2244 item. This has been fixed.
2246 PH/03 Added additional dnslists conditions == and =& which are different from
2247 = and & when the dns lookup returns more than one IP address.
2249 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2250 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2252 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2253 FSYNC, which compiles an option called disable_fsync that allows for
2254 bypassing fsync(). The documentation is heavily laced with warnings.
2256 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2258 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2259 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2260 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2261 including adding "make clean"; (3) Added -fPIC when compiling the test
2262 dynamically loaded module, to get rid of a warning.
2264 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2265 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2266 = 0s. The bug is otherwise harmless.
2268 PH/07 There was a bug in the dovecot authenticator such that the value of
2269 $auth1 could be overwritten, and so not correctly preserved, after a
2270 successful authentication. This usually meant that the value preserved by
2271 the server_setid option was incorrect.
2273 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2275 PH/09 Installed PCRE release 7.0.
2277 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2278 run for batched SMTP input. It is now run at the start of every message
2279 in the batch. While fixing this I discovered that the process information
2280 (output by running exiwhat) was not always getting set for -bs and -bS
2281 input. This is fixed, and it now also says "batched" for BSMTP.
2283 PH/11 Added control=no_pipelining.
2285 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2286 patch, slightly modified), and move the expansion of helo_data till after
2287 the connection is made in the smtp transport (so it can use these
2290 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2292 PH/14 Added log_selector = +pid.
2294 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2296 PH/16 Add ${if forany and ${if forall.
2298 PH/17 Added dsn_from option to vary the From: line in DSNs.
2300 PH/18 Flush SMTP output before performing a callout, unless control =
2301 no_callout_flush is set.
2303 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2304 was true (the default) a successful delivery failed to delete the retry
2305 item, thus causing premature timeout of the address. The bug is now
2308 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2310 PH/21 Long custom messages for fakedefer and fakereject are now split up
2311 into multiline responses in the same way that messages for "deny" and
2312 other ACL rejections are.
2314 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2315 with slight modification.
2317 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2318 draft, changing the syntax and factoring some duplicate code".
2320 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2321 for deliveries of the second and subsequent messages over the same SMTP
2324 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2325 ${reduce, with only minor "tidies".
2327 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2329 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2330 expansion side effects.
2332 PH/27 When a message times out after an over-quota error from an Exim-imposed
2333 quota, the bounce message says "mailbox is full". This message was not
2334 being given when it was a system quota that was exceeded. It now should
2337 MH/03 Made $recipients available in local_scan(). local_scan() already has
2338 better access to the recipient list through recipients_list[], but
2339 $recipients can be useful in postmaster-provided expansion strings.
2341 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2342 in the case of a MAIL command with additional options following the
2343 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2344 were accidentally chopped off.
2346 PH/29 SMTP synchronization checks are implemented when a command is read -
2347 there is a check that no more input is waiting when there shouldn't be
2348 any. However, for some commands, a delay in an ACL can mean that it is
2349 some time before the response is written. In this time, more input might
2350 arrive, invalidly. So now there are extra checks after an ACL has run for
2351 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2352 pipelining has not been advertised.
2354 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2356 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2357 returned in the SMTP response when smtp_return_error_details was false.
2358 This has been fixed.
2360 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2361 instead of the C I/O that was originally supplied, because problems were
2362 reported on Solaris.
2364 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2365 Exim which did not show up earlier: it was assuming that a call to
2366 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2367 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2368 was a macro that became an assignment, so it seemed to work. This has
2369 changed to a proper function call with a void return, hence the compile
2370 error. Exim's code has been fixed.
2372 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2375 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2377 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2379 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2380 because it turns out to be too big or there is a timeout) there is no
2381 "Completed" line in the log. When some messages of this type were
2382 selected by exigrep, they were listed as "not completed". Others were
2383 picked up by some special patterns. I have improved the selection
2384 criteria to be more general.
2386 PH/38 The host_find_failed option in the manualroute router can now be set
2387 to "ignore", to completely ignore a host whose IP address cannot be
2388 found. If all hosts are ignored, the behaviour is controlled by the new
2389 host_all_ignored option.
2391 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2392 homing or because of multiple MX records with /mx) generated more than
2393 one IP address, and the following item turned out to be the local host,
2394 all the secondary addresses of the first item were incorrectly removed
2395 from the list, along with the local host and any following hosts (which
2396 is what is supposed to happen).
2398 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2399 whoever called Exim into the -H file. In the case of the daemon it was
2400 behaving confusingly. When first started, it used values for whoever
2401 started the daemon, but after a SIGHUP it used the Exim user (because it
2402 calls itself on a restart). I have changed the code so that it now always
2405 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2406 message are rejected with the same error (e.g. no authentication or bad
2407 sender address), and a DATA command is nevertheless sent (as can happen
2408 with PIPELINING or a stupid MUA), the error message that was given to the
2409 RCPT commands is included in the rejection of the DATA command. This is
2410 intended to be helpful for MUAs that show only the final error to their
2413 PH/42 Another patch from the Sieve maintainer.
2415 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2416 Eximstats - Fixed some broken HTML links and added missing column headers
2418 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2419 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2421 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2423 PH/43 Yet another patch from the Sieve maintainer.
2425 PH/44 I found a way to check for a TCP/IP connection going away before sending
2426 the response to the final '.' that terminates a message, but only in the
2427 case where the client has not sent further data following the '.'
2428 (unfortunately, this is allowed). However, in many cases there won't be
2429 any further data because there won't be any more messages to send. A call
2430 to select() can be used: if it shows that the input is "ready", there is
2431 either input waiting, or the socket has been closed. An attempt to read
2432 the next input character can distinguish the two cases. Previously, Exim
2433 would have sent an OK response which the client would never have see.
2434 This could lead to message repetition. This fix should cure that, at
2435 least in a lot of common cases.
2437 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2438 advertised in response to EHLO.
2444 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2445 fixed by 4.65/MH/01 (is this a record?) are fixed:
2447 (i) An empty string was always treated as zero by the numeric comparison
2448 operators. This behaviour has been restored.
2450 (ii) It is documented that the numeric comparison operators always treat
2451 their arguments as decimal numbers. This was broken in that numbers
2452 starting with 0 were being interpreted as octal.
2454 While fixing these problems I realized that there was another issue that
2455 hadn't been noticed. Values of message_size_limit (both the global option
2456 and the transport option) were treated as octal if they started with 0.
2457 The documentation was vague. These values are now always treated as
2458 decimal, and I will make that clear in the documentation.
2464 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2465 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2468 MH/01 Don't check that the operands of numeric comparison operators are
2469 integers when their expansion is in "skipping" mode (fixes bug
2470 introduced by 4.64-PH/07).
2472 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2473 child addresses, Exim now panics and dies. Previously, because the count
2474 is held in a short int, deliveries were likely to be lost. As such a
2475 large number of recipients for a single message is ridiculous
2476 (performance will be very, very poor), I have chosen to impose a limit
2477 rather than extend the field.
2483 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2484 leftover -K file (the existence of which was triggered by #402).
2485 While we were at it, introduced process PID as part of the -K
2486 filename. This should rule out race conditions when creating
2489 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2490 processing considerably. Previous code took too long for large mails,
2491 triggering a timeout which in turn triggers #401.
2493 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2494 in the DK code in transports.c. sendfile() is not really portable,
2495 hence the _LINUX specificness.
2497 TF/01 In the add_headers option to the mail command in an Exim filter,
2498 there was a bug that Exim would claim a syntax error in any
2499 header after the first one which had an odd number of characters
2502 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2503 callout verification, Exim cached a "reject" for the entire domain. This
2504 is correct for most verifications, but it is not correct for a recipient
2505 verification with use_sender or use_postmaster set, because in that case
2506 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2507 case of MAIL FROM:<> rejection from other early rejections (e.g.
2508 rejection of HELO). When verifying a recipient using a non-null MAIL
2509 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2510 Whatever the result of the callout, the value of the domain cache is
2511 left unchanged (for any other kind of callout, getting as far as trying
2512 RCPT means that the domain itself is ok).
2514 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2517 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2518 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2519 session, the daemon ignores ECONNECT errors and logs others; it now
2520 ignores EPIPE as well.
2522 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2523 (quoted-printable decoding).
2525 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2526 later the small subsequent patch to fix an introduced bug.
2528 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2530 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2532 PH/08 An error is now given if message_size_limit is specified negative.
2534 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2535 to be given (somewhat) arbitrary names.
2537 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2540 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2541 miscellaneous code fixes
2543 PH/10 Added the log_reject_target ACL modifier to specify where to log
2546 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2547 hostname. This is wrong, because it relates to the incoming message (and
2548 probably the interface on which it is arriving) and not to the outgoing
2549 callout (which could be using a different interface). This has been
2550 changed to use the value of the helo_data option from the smtp transport
2551 instead - this is what is used when a message is actually being sent. If
2552 there is no remote transport (possible with a router that sets up host
2553 addresses), $smtp_active_hostname is used.
2555 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2556 tweaks were necessary in order to get it to work (see also 21 below):
2557 (a) The code assumed that strncpy() returns a negative number on buffer
2558 overflow, which isn't the case. Replaced with Exim's string_format()
2560 (b) There were several signed/unsigned issues. I just did the minimum
2561 hacking in of casts. There is scope for a larger refactoring.
2562 (c) The code used strcasecmp() which is not a standard C function.
2563 Replaced with Exim's strcmpic() function.
2564 (d) The code set only $1; it now sets $auth1 as well.
2565 (e) A simple test gave the error "authentication client didn't specify
2566 service in request". It would seem that Dovecot has changed its
2567 interface. Fortunately there's a specification; I followed it and
2568 changed what the client sends and it appears to be working now.
2570 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2573 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2574 address A is aliased to B and C, where B exists and C does not. Without
2575 -v the output is "A verified" because verification stops after a
2576 successful redirection if more than one address is generated. However,
2577 with -v the child addresses are also verified. Exim was outputting "A
2578 failed to verify" and then showing the successful verification for C,
2579 with its parentage. It now outputs "B failed to verify", showing B's
2580 parentage before showing the successful verification of C.
2582 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2583 look up a TXT record in a specific list after matching in a combined
2586 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2587 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2588 they consult the DNS. I had assumed they would set it the way they
2589 wanted; and indeed my experiments on Linux seem to show that in some
2590 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2591 To be on the safe side, however, I have now made the interface to
2592 host_find_byname() similar to host_find_bydns(), with an argument
2593 containing the DNS resolver options. The host_find_byname() function now
2594 sets these options at its start, just as host_find_bydns() does. The smtp
2595 transport options dns_qualify_single and dns_search_parents are passed to
2596 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2597 of host_find_byname() use the default settings of RES_DEFNAMES
2598 (qualify_single) but not RES_DNSRCH (search_parents).
2600 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2601 spool_read_header() do less string testing, by means of a preliminary
2602 switch on the second character of optional "-foo" lines. (This is
2603 overdue, caused by the large number of possibilities that now exist.
2604 Originally there were few.) While I was there, I also converted the
2605 str(n)cmp tests so they don't re-test the leading "-" and the first
2606 character, in the hope this might squeeze out yet more improvement.
2608 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2609 flag allowing group syntax was set by the header_syntax check but not
2610 turned off, possible causing trouble later; (2) The flag was not being
2611 set at all for the header_verify test, causing "group"-style headers to
2612 be rejected. I have now set it in this case, and also caused header_
2613 verify to ignore an empty address taken from a group. While doing this, I
2614 came across some other cases where the code for allowing group syntax
2615 while scanning a header line wasn't quite right (mostly, not resetting
2616 the flag correctly in the right place). These bugs could have caused
2617 trouble for malformed header lines. I hope it is now all correct.
2619 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2620 with the "reply" argument non-NULL. The code, however (which originally
2621 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2622 but it didn't always do it. This confused somebody who was copying the
2623 code for some other use. I have removed all the tests.
2625 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2626 feature that was used to support insecure browsers during the U.S. crypto
2627 embargo. It requires special client support, and Exim is probably the
2628 only MTA that supported it -- and would never use it because real RSA is
2629 always available. This code has been removed, because it had the bad
2630 effect of slowing Exim down by computing (never used) parameters for the
2631 RSA_EXPORT functionality.
2633 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2634 authenticator to fail if there's a tab character in the incoming data
2635 (there should never be unless someone is messing about, as it's supposed
2636 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2637 if the connection is using TLS or if the remote IP is the same as the
2638 local IP, and the "valid-client-cert option" if a client certificate has
2641 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2642 authenticators. This can be used for authorization after authentication
2643 succeeds. (In the case of plaintext, it servers for both authentication
2646 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2647 if any retry times were supplied.
2649 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2650 connection, where there is no client IP address to check. In this
2651 situation, the verify now always succeeds.
2653 PH/25 Applied John Jetmore's -Mset patch.
2655 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2657 PH/27 In a string expansion for a processed (not raw) header when multiple
2658 headers of the same name were present, leading whitespace was being
2659 removed from all of them, but trailing whitespace was being removed only
2660 from the last one. Now trailing whitespace is removed from each header
2661 before concatenation. Completely empty headers in a concatenation (as
2662 before) are ignored.
2664 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2665 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2667 PH/29 [Removed. This was a change that I later backed out, and forgot to
2668 correct the ChangeLog entry (that I had efficiently created) before
2669 committing the later change.]
2671 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2672 address errors (4xx response to RCPT) over the same connection as other
2673 messages routed to the same hosts. Such deliveries are always "forced",
2674 so retry times are not inspected. This resulted in far too many retries
2675 for the affected addresses. The effect occurred only when there were more
2676 hosts than the hosts_max_try setting in the smtp transport when it had
2677 the 4xx errors. Those hosts that it had tried were not added to the list
2678 of hosts for which the message was waiting, so if all were tried, there
2679 was no problem. Two fixes have been applied:
2681 (i) If there are any address or message errors in an SMTP delivery, none
2682 of the hosts (tried or untried) are now added to the list of hosts
2683 for which the message is waiting, so the message should not be a
2684 candidate for sending over the same connection that was used for a
2685 successful delivery of some other message. This seems entirely
2686 reasonable: after all the message is NOT "waiting for some host".
2687 This is so "obvious" that I'm not sure why it wasn't done
2688 previously. Hope I haven't missed anything, but it can't do any
2689 harm, as the worst effect is to miss an optimization.
2691 (ii) If, despite (i), such a delivery is accidentally attempted, the
2692 routing retry time is respected, so at least it doesn't keep
2693 hammering the server.
2695 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2696 in ${readsocket because some servers need this prod.
2698 PH/32 Added some extra debug output when updating a wait-xxx database.
2700 PH/33 The hint "could be header name not terminated by colon", which has been
2701 given for certain expansion errors for a long time, was not being given
2702 for the ${if def:h_colon_omitted{... case.
2704 PH/34 The spec says: "With one important exception, whenever a domain list is
2705 being scanned, $domain contains the subject domain." There was at least
2706 one case where this was not true.
2708 PH/35 The error "getsockname() failed: connection reset by peer" was being
2709 written to the panic log as well as the main log, but it isn't really
2710 panic-worthy as it just means the connection died rather early on. I have
2711 removed the panic log writing for the ECONNRESET error when getsockname()
2714 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2715 runs only) independently of the message's sender address. This meant
2716 that, if the 4xx error was in fact related to the sender, a different
2717 message to the same recipient with a different sender could confuse
2718 things. In particular, this can happen when sending to a greylisting
2719 server, but other circumstances could also provoke similar problems.
2720 I have changed the default so that the retry time for these errors is now
2721 based a combination of the sender and recipient addresses. This change
2722 can be overridden by setting address_retry_include_sender=false in the
2725 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2726 remote server are returned as part of bounce messages. This was not
2727 happening for LMTP over a pipe (the lmtp transport), but now it is the
2728 same for both kinds of LMTP.
2730 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2731 in header lines that were in fact CNAMEs. This is no longer the case.
2733 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2734 and queue runs started by the daemon processed all messages. This has
2735 been fixed so that -R and -S can now usefully be given with -q<time>.
2737 PH/40 Import PCRE release 6.7 (fixes some bugs).
2739 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2741 PH/42 Give an error if -q is specified more than once.
2743 PH/43 Renamed the variables $interface_address and $interface_port as
2744 $received_ip_address and $received_port, to make it clear that these
2745 values apply to message reception, and not to the outgoing interface when
2746 a message is delivered. (The old names remain recognized, of course.)
2748 PH/44 There was no timeout on the connect() call when using a Unix domain
2749 socket in the ${readsocket expansion. There now is.
2751 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2752 be meaningful with "accept".
2754 SC/01 Eximstats V1.43
2755 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2757 SC/02 Eximstats V1.44
2758 Use a glob alias rather than an array ref in the generated
2759 parser. This improves both readability and performance.
2761 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2762 Collect SpamAssassin and rejection statistics.
2763 Don't display local sender or destination tables unless
2764 there is data to show.
2765 Added average volumes into the top table text output.
2767 SC/04 Eximstats V1.46
2768 Collect data on the number of addresses (recipients)
2769 as well as the number of messages.
2771 SC/05 Eximstats V1.47
2772 Added 'Message too big' to the list of mail rejection
2773 reasons (thanks to Marco Gaiarin).
2775 SC/06 Eximstats V1.48
2776 Mainlog lines which have GMT offsets and are too short to
2777 have a flag are now skipped.
2779 SC/07 Eximstats V1.49 (Alain Williams)
2780 Added the -emptyok flag.
2782 SC/08 Eximstats V1.50
2783 Fixes for obtaining the IP address from reject messages.
2785 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2786 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2787 whitespace changes from 4.64-PH/27
2789 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2792 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2793 are found, allow negative numbers in numeric criteria)
2795 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2797 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2798 to match changes made in 4.64-PH/43
2800 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2802 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2803 "domain" is actually an IP address, and force a failure. This locks out
2804 those revolvers/nameservers that support "A-for-A" lookups, in
2805 contravention of the specifications.
2807 PH/48 When a host name was looked up from an IP address, and the subsequent
2808 forward lookup of the name timed out, the host name was left in
2809 $sender_host_name, contrary to the specification.
2811 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2812 restricted to single-key lookups, Exim was not diagnosing an error if
2813 * or *@ was used with a query-style lookup.
2815 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2817 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2818 long ago, but noone interested enough thought of it. Let's just say that
2819 the "1.1" means that there are some new functions that weren't there at
2820 some point in the past.
2822 PH/51 Error processing for expansion failure of helo_data from an smtp
2823 transport during callout processing was broken.
2825 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2826 tested/used via the -bh/-bhc/-bs options.
2828 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2829 bug, fixed in subsequent PCRE releases).
2831 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2832 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2834 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2840 SC/01 Use a glob alias rather than an array ref in eximstats generated
2841 parser. This improves both readability and performance.
2843 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2844 Don't display local sender or destination tables in eximstats unless
2845 there is data to show.
2846 Added average volumes into the eximstats top table text output.
2848 SC/03 Collect data on the number of addresses (recipients) as well
2849 as the number of messages in eximstats.
2851 TF/01 Correct an error in the documentation for the redirect router. Exim
2852 does (usually) call initgroups() when daemonizing.
2854 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2855 with consistent privilege compared to when running as a daemon.
2857 TF/03 Note in the spec that $authenticated_id is not set for local
2858 submissions from trusted users.
2860 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2861 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2863 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2864 by adding some example configuration directives to the default
2865 configuration file. A little bit of work is required to uncomment the
2866 directives and define how usernames and passwords are checked, but
2867 there is now a framework to start from.
2869 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2870 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2871 without this. I don't know how relevant this is to other LDAP libraries.
2873 PH/02 Add the verb name to the "unknown ACL verb" error.
2875 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2877 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2879 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2880 directory not expanded when it should be if an expanded home directory
2881 was set for the address (which is overridden by the transport).
2883 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2886 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2887 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2888 because it is too late at that time, and has no effect.
2890 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2891 security issue with \' (bugzilla #107). I could not use the
2892 PQescapeStringConn() function, because it needs a PGconn value as one of
2895 PH/08 When testing addresses using -bt, indicate those final addresses that
2896 are duplicates that would not cause an additional delivery. At least one
2897 person was confused, thinking that -bt output corresponded to deliveries.
2898 (Suppressing duplicates isn't a good idea as you lose the information
2899 about possibly different redirections that led to the duplicates.)
2901 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2902 systems where poll() doesn't work, in particular OS X.
2904 PH/10 Added more information to debugging output for retry time not reached.
2906 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2907 operations in malware.c.
2909 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2912 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2913 syslog_duplication set false, log lines that would normally be written
2914 both the the main log and to the reject log were not written to syslog at
2917 PH/14 In the default configuration, change the use of "message" in ACL warn
2918 statements to "add_header".
2920 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2921 not followed by a command (e.g. "seen endif").
2923 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2924 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2927 PH/17 Added extra conditions to the default value of delay_warning_condition
2931 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2932 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2933 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2936 The Auto-Submitted: and various List- headers are standardised, whereas I
2937 don't think Precedence: ever was.
2939 PH/18 Refactored debugging code in route_finduser() to show more information,
2940 in particular, the error code if getpwnam() issues one.
2942 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2943 This is apparently needed in addition to the PH/07 change above to avoid
2944 any possible encoding problems.
2946 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2947 but not after initializing Perl.
2949 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2950 output them only if debugging. By default they are written stderr,
2951 apparently, which is not desirable.
2953 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2956 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2959 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2961 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2962 authenticated or an ident call has been made. Suppress the default
2963 values for $authenticated_id and $authenticated_sender (but permit -oMai
2964 and -oMas) when testing with -bh.
2966 PH/24 Re-jigged the order of the tests in the default configuration so that the
2967 tests for valid domains and recipients precede the DNS black list and CSA
2968 tests, on the grounds that those ones are more expensive.
2970 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2971 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2972 This bug exists in every version of Exim that I still have, right back to
2975 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2976 However, an attempt to turn on case-sensitivity in a regex key by
2977 including (?-i) didn't work because the subject string was already
2978 lowercased, and the effects were non-intuitive. It turns out that a
2979 one-line patch can be used to allow (?-i) to work as expected.
2985 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2986 other effects) broke the use of negated acl sub-conditions.
2988 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2991 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2992 "Deny" causes Exim to reject the incoming connection with a 554 error.
2993 Unfortunately, if there is a major crisis, such as a disk failure,
2994 tcp-wrappers gives "deny", whereas what one would like would be some
2995 kind of temporary error. A kludge has been added to help with this.
2996 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2997 554 error is used if errno is still zero or contains ENOENT (which occurs
2998 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3001 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3003 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3004 errors. Otherwise a message that provokes a temporary error (when other
3005 messages do not) can cause a whole host to time out.
3007 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3008 addresses were routed directly to files or pipes from a redirect router.
3009 File deliveries just didn't batch; pipe deliveries might have suffered
3012 PH/06 A failure to get a lock for a hints database would erroneously always say
3013 "Failed to get write lock", even when it was really a read lock.
3015 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3016 of 0600. This has been changed to use the value of the lockfile_mode
3017 option (which defaults to 0600).
3019 PH/08 Applied small patch from the Sieve maintainer.
3021 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3022 folder from quota calculations, a direct delivery into this folder messed
3023 up the contents of the maildirsize file. This was because the regex was
3024 used only to exclude .Trash (or whatever) when the size of the mailbox
3025 was calculated. There was no check that a delivery was happening into an
3026 excluded directory. This bug has been fixed by ignoring all quota
3027 processing for deliveries into excluded directories.
3029 PH/10 Added the maildirfolder_create_regex option to appendfile.
3035 PH/01 The code for finding all the local interface addresses on a FreeBSD
3036 system running IPv6 was broken. This may well have applied to all BSD
3037 systems, as well as to others that have similar system calls. The broken
3038 code found IPv4 interfaces correctly, but gave incorrect values for the
3039 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3040 that it would not match correctly against @[] and not recognize the IPv6
3043 PH/02 The ipliteral router was not recognizing addresses of the form user@
3044 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3046 PH/03 Added disable_ipv6.
3048 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3049 decoded form, because it is most often used to construct To: headers
3050 lines in autoreplies, and the decoded form may well be syntactically
3051 invalid. However, $reply_address has leading white space removed, and all
3052 newlines turned into spaces so that the autoreply transport does not
3055 PH/05 If group was specified without a user on a router, and no group or user
3056 was specified on a transport, the group from the router was ignored.
3058 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3059 for visible compile-time settings that can be used to change these
3060 numbers, for those that want even more. Backwards compatibility with old
3061 spool files has been maintained. However, going back to a previous Exim
3062 release will lost any variables that are in spool files.
3064 PH/07 Two small changes when running in the test harness: increase delay when
3065 passing a TCP/IP connection to a new process, in case the original
3066 process has to generate a bounce, and remove special handling of
3067 127.0.0.2 (sic), which is no longer necessary.
3069 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3070 be the same on different OS.
3072 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3075 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3076 whether --show-vars was specified or not
3078 JJ/02 exipick: Added support for new ACL variable spool format introduced
3081 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3082 syntactically invalid From: or Reply-to: line, and a filter used this to
3083 generate an autoreply, and therefore failed to obtain an address for the
3084 autoreply, Exim could try to deliver to a non-existent relative file
3085 name, causing unrelated and misleading errors. What now happens is that
3086 it logs this as a hard delivery error, but does not attempt to create a
3089 PH/11 The exinext utility has a -C option for testing purposes, but although
3090 the given file was scanned by exinext itself; it wasn't being passed on
3091 when Exim was called.
3093 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3094 an end-of-file indication when reading a command response.
3096 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3097 compiled. In many other places in Exim, IPv6 addresses are always
3098 recognized, so I have changed this. It also means that IPv4 domain
3099 literals of the form [IPV4:n.n.n.n] are now always recognized.
3101 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3102 used if the router is not running as root, for example, when verifying at
3103 ACL time, or when using -bh. The debugging output from this situation was
3104 non-existent - all you got was a failure to exec. I have made two
3107 (a) Failures to set uid/gid, the current directory, or a process leader
3108 in a subprocess such as that created by queryprogram now generate
3109 suitable debugging output when -d is set.
3111 (b) The queryprogram router detects when it is not running as root,
3112 outputs suitable debugging information if -d is set, and then runs
3113 the subprocess without attempting to change uid/gid.
3115 PH/15 Minor change to Makefile for building test_host (undocumented testing
3118 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3119 additional section of a DNS packet that returns MX or SRV records.
3120 Instead, it always explicitly searches for A/AAAA records. This avoids
3121 major problems that occur when a DNS server includes only records of one
3122 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3123 fixed another bug: if SRV records were looked up and the corresponding
3124 address records were *not* found in the additional section, the port
3125 values from the SRV records were lost.
3127 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3128 using the correct key (the original address) when searching the retry
3129 rules in order to find which one to use for generating the retry hint.
3131 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3132 adding the default one. Similarly, if it contains a Reply-To: header, the
3133 errors_reply_to option, if set, is not used.
3135 PH/19 When calculating a retry time, Exim used to measure the "time since
3136 failure" by looking at the "first failed" field in the retry record. Now
3137 it does not use this if it is later than than the arrival time of the
3138 message. Instead it uses the arrival time. This makes for better
3139 behaviour in cases where some deliveries succeed, thus re-setting the
3140 "first failed" field. An example is a quota failure for a huge message
3141 when small messages continue to be delivered. Without this change, the
3142 "time since failure" will always be short, possible causing more frequent
3143 delivery attempts for the huge message than are intended.
3144 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3146 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3147 $1, $2, $3) because the numerical variables can be reset during some
3148 expansion items (e.g. "match"), thereby losing the authentication data.
3150 PH/21 Make -bV show the size of off_t variables so that the test suite can
3151 decide whether to run tests for quotas > 2G.
3153 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3154 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3155 filecount value is greater than 2G or if a quota value is greater than 2G
3156 on a system where the size of off_t is not greater than 4, a panic error
3159 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3160 never match. The debug and -bh output now contains an explicit error
3161 message indicating a malformed IPv4 address or mask.
3163 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3164 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3165 PH/23 above applies.
3167 PH/25 Do not write to syslog when running in the test harness. The only
3168 occasion when this arises is a failure to open the main or panic logs
3169 (for which there is an explicit test).
3171 PH/26 Added the /no_tell option to "control=freeze".
3173 PH/27 If a host name lookup failed very early in a connection, for example, if
3174 the IP address matched host_lookup and the reverse lookup yielded a name
3175 that did not have a forward lookup, an error message of the form "no IP
3176 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3177 could be logged. Now it outputs the IP address instead of "NULL".
3179 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3180 allows the sender and the authenticated sender to be set when
3181 submitting a message from within Exim. Since child_open_exim() is
3182 documented for local_scan(), the new function should be too.
3184 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3185 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3186 results in an empty string is now treated as unset.
3188 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3190 PH/31 Added sender_verify_fail logging option.
3192 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3193 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3194 tidied the source and removed it altogether.
3196 PH/33 When a queue run was abandoned because the load average was too high, a
3197 log line was always written; now it is written only if the queue_run log
3198 selector is set. In addition, the log line for abandonment now contains
3199 information about the queue run such as the pid. This is always present
3200 in "start" and "stop" lines but was omitted from the "abandon" line.
3202 PH/34 Omit spaces between a header name and the colon in the error message that
3203 is given when verify = headers_syntax fails (if there are lots of them,
3204 the message gets confusing).
3206 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3207 names, as there are now some PTR records that contain slashes. This check
3208 is only to protect against broken name servers that fall over on strange
3209 characters, so the fact that it applies to all lookups doesn't matter.
3211 PH/36 Now that the new test suite is complete, we can remove some of the
3212 special code in Exim that was needed for the old test suite. For example,
3213 sorting DNS records because real resolvers return them in an arbitrary
3214 order. The new test suite's fake resolver always returns records in the
3217 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3218 bounces) except when queue_only is set, to avoid logging races between
3219 the different processes.
3221 PH/38 Panic-die if .include specifies a non-absolute path.
3223 PH/39 A tweak to the "H" retry rule from its user.
3225 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3226 a label. They prevented compilation on older perls.
3228 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3229 a warning to be raised on newish perls.
3231 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3232 on queue. Changes to match documented behaviour of showing count of
3233 messages matching specified criteria.
3235 PH/40 Changed the default ident timeout from 30s to 5s.
3237 PH/41 Added support for the use of login_cap features, on those BSD systems
3238 that have them, for controlling the resources used by pipe deliveries.
3240 PH/42 The content-scanning code uses fopen() to create files in which to put
3241 message data. Previously it was not paying any attention to the mode of
3242 the files. Exim runs with umask(0) because the rest of the code creates
3243 files with open(), and sets the required mode explicitly. Thus, these
3244 files were ending up world-writeable. This was not a big issue, because,
3245 being within the spool directory, they were not world-accessible. I have
3246 created a function called modefopen, which takes an additional mode
3247 argument. It sets umask(777), creates the file, chmods it to the required
3248 mode, then resets the umask. All the relevant calls to fopen() in the
3249 content scanning code have been changed to use this function.
3251 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3252 to 24 hours. This avoids potential overflow problems when processing G
3253 and H retry rules. I suspect nobody ever tinkers with this value.
3255 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3257 PH/45 When the plaintext authenticator is running as a client, the server's
3258 challenges are checked to ensure they are valid base64 strings. By
3259 default, the authentication attempt is cancelled if an invalid string is
3260 received. Setting client_ignore_invalid_base64 true ignores these errors.
3261 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3262 they are received. Thus, the responses can be made to depend on the
3263 challenges. If an invalid string is ignored, an empty string is placed in
3266 PH/46 Messages that are created by the autoreply transport now contains a
3267 References: header, in accordance with RFCs 2822 and 3834.
3269 PH/47 Added authenticated_sender_force to the smtp transport.
3271 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3273 PH/49 Installed latest patch from the Sieve maintainer.
3275 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3276 was also set, the appendfile transport was unnecessarily scanning a
3277 directory of message files (e.g. for maildir delivery) to find the count
3278 of files (along with the size), even though it did not need this
3279 information. It now does the scan only if it needs to find either the
3280 size of the count of files.
3282 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3284 PH/52 Two bugs concerned with error handling when the smtp transport is
3287 (i) Exim was not creating retry information for temporary errors given
3288 for individual recipients after the DATA command when the smtp transport
3289 was used in LMTP mode. This meant that they could be retried too
3290 frequently, and not timed out correctly.
3292 (ii) Exim was setting the flag that allows error details to be returned
3293 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3294 recipients that were returned after the DATA command.
3296 PH/53 This is related to PH/52, but is more general: for any failing address,
3297 when detailed error information was permitted to be returned to the
3298 sender, but the error was temporary, then after the final timeout, only
3299 "retry timeout exceeded" was returned. Now it returns the full error as
3300 well as "retry timeout exceeded".
3302 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3303 do this, and (what is worse) MTAs that accept it.
3305 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3306 will now be deprecated.
3308 PH/56 New os.c-cygwin from the Cygwin maintainer.
3310 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3311 formats (previously only available in exim formats via -bpr, -bpru,
3312 and -bpra. Now also available in native and exiqgrep formats)
3314 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3315 with very large, slow to parse queues
3317 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3319 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3321 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3322 responses to authentication challenges, though it was showing the
3323 challenges; (ii) I've removed the CR characters from the debug output for
3326 PH/58 Allow for the insertion of a newline as well as a space when a string
3327 is turned into more than one encoded-word during RFC 2047 encoding. The
3328 Sieve code now uses this.
3330 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3331 data_4xx, lost_connection, tls_required.
3333 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3334 message was being sent as an SMTP response.
3336 PH/61 Add -l and -k options to exicyclog.
3338 PH/62 When verifying, if an address was redirected to one new address, so that
3339 verification continued, and the new address failed or deferred after
3340 having set something in $address_data, the value of $address_data was not
3341 passed back to the ACL. This was different to the case when no
3342 redirection occurred. The value is now passed back in both cases.
3344 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3345 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3346 use login_cap.h, so on its own it isn't the distinguishing feature. The
3347 new name refers directly to the setclassresources() function.
3349 PH/65 Added configuration files for NetBSD3.
3351 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3353 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3354 is preferred over IPv4.
3356 PH/68 The bounce_return_message and bounce_return_body options were not being
3357 honoured for bounces generated during the reception of non-SMTP messages.
3358 In particular, this applied to messages rejected by the ACL. This bug has
3359 been fixed. However, if bounce_return_message is true and bounce_return_
3360 body is false, the headers that are returned for a non-SMTP message
3361 include only those that have been read before the error was detected.
3362 (In the case of an ACL rejection, they have all been read.)
3364 PH/69 The HTML version of the specification is now built in a directory called
3365 spec_html instead of spec.html, because the latter looks like a path with
3366 a MIME-type, and this confuses some software.
3368 PH/70 Catch two compiler warnings in sieve.c.
3370 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3371 function verify_get_ident() calls ip_connect() to connect a socket, but
3372 if the "connect()" function timed out, ip_connect() used to close the
3373 socket. However, verify_get_ident() also closes the socket later, and in
3374 between Exim writes to the log, which may get opened at this point. When
3375 the socket was closed in ip_connect(), the log could get the same file
3376 descriptor number as the socket. This naturally causes chaos. The fix is
3377 not to close the socket in ip_connect(); the socket should be closed by
3378 the function that creates it. There was only one place in the code where
3379 this was missing, in the iplookup router, which I don't think anybody now
3380 uses, but I've fixed it anyway.
3382 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3383 well as to direct DNS lookups. Otherwise the handling of names in host
3384 lists is inconsistent and therefore confusing.
3390 PH/01 Two changes to the default runtime configuration:
3392 (1) Move the checks for relay_from_hosts and authenticated clients from
3393 after to before the (commented out) DNS black list checks.
3395 (2) Add control=submission to the relay_from_hosts and authenticated
3396 clients checks, on the grounds that messages accepted by these
3397 statements are most likely to be submissions.
3399 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3401 (1) Generate an error if the third argument for the ${prvs expansion is
3404 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3407 (3) Reset the variables that are obtained from the first argument of
3408 ${prvscheck and used in the second argument before leaving the code,
3409 because their memory is reclaimed, so using them afterwards may do
3412 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3413 one (it's much easier than Tom thought :-).
3415 (5) Because of (4), we can now allow for the use of $prvscheck_result
3416 inside the third argument.
3418 PH/03 For some reason, the default setting of PATH when running a command from
3419 a pipe transport was just "/usr/bin". I have changed it to
3422 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3423 anything to be listed in the output from -bV.
3425 PH/05 When a filter generated an autoreply, the entire To: header line was
3426 quoted in the delivery log line, like this:
3428 => >A.N.Other <ano@some.domain> <original@ddress> ...
3430 This has been changed so that it extracts the operative address. There
3431 may be more than one such address. If so, they are comma-separated, like
3434 => >ano@some.domain,ona@other.domain <original@ddress> ...
3436 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3437 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3438 not being looked up in the rDNS to get a host name, Exim was showing the
3439 IP address twice in Received: lines, even though the IP addresses were
3440 identical. For example:
3442 Received: from [1.2.3.4] (helo=[1.2.3.4])
3444 However, if the real host name was known, it was omitting the HELO data
3445 if it matched the actual IP address. This has been tidied up so that it
3446 doesn't show the same IP address twice.
3448 PH/07 When both +timestamp and +memory debugging was on, the value given by
3449 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3450 called by the debug printing, thereby overwriting the timestamp buffer.
3451 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3453 PH/08 When the original message was included in an autoreply transport, it
3454 always said "this is a copy of the message, including all the headers",
3455 even if body_only or headers_only was set. It now gives an appropriate
3458 PH/09 Applied a patch from the Sieve maintainer which:
3460 o fixes some comments
3461 o adds the (disabled) notify extension core
3462 o adds some debug output for the result of if/elsif tests
3463 o points to the current vacation draft in the documentation
3464 and documents the missing references header update
3468 o fixes a bug in processing the envelope test (when testing
3469 multiple envelope elements, the last element determined the
3472 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3473 Electronic Mail") by including:
3475 Auto-submitted: auto-generated
3477 in the messages that it generates (bounce messages and others, such as
3478 warnings). In the case of bounce messages for non-SMTP messages, there was
3479 also a typo: it was using "Auto_submitted" (underscore instead of
3480 hyphen). Since every message generated by Exim is necessarily in response
3481 to another message, thes have all been changed to:
3483 Auto-Submitted: auto-replied
3485 in accordance with these statements in the RFC:
3487 The auto-replied keyword:
3489 - SHOULD be used on messages sent in direct response to another
3490 message by an automatic process,
3492 - MUST NOT be used on manually-generated messages,
3494 - MAY be used on Delivery Status Notifications (DSNs) and Message
3495 Disposition Notifications (MDNs),
3497 - MUST NOT be used on messages generated by automatic or periodic
3498 processes, except for messages which are automatic responses to
3501 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3502 to the default Received: header definition.
3504 PH/12 Added log selector acl_warn_skipped (default on).
3506 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3507 variables because (a) they are in the wrong storage pool and (b) even if
3508 they were copied, it wouldn't work properly because of the caching.
3510 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3511 checking when decoding. Apparently there are clients that generate
3512 overlong encoded strings. Why am I not surprised?
3514 PH/15 If the first argument of "${if match_address" was not empty, but did not
3515 contain an "@" character, Exim crashed. Now it writes a panic log message
3516 and treats the condition as false.
3518 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3520 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3521 extension "envelope-auth". The code is finished and in agreement with
3522 other implementations, but there is no documentation so far and in fact,
3523 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3524 not changing the active code.
3526 Print executed "if" and "elsif" statements when debugging is used. This
3527 helps a great deal to understand what a filter does.
3529 Document more things not specified clearly in RFC3028. I had all this
3530 sorted out, when out of a sudden new issues came to my mind. Oops."
3532 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3535 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3536 canonical form (as documented). However, after a host name lookup from
3537 the IP address, check_host() was doing a simple string comparison with
3538 addresses acquired from the DNS when checking that the found name did
3539 have the original IP as one of its addresses. Since any found IPv6
3540 addresses are likely to be in abbreviated form, the comparison could
3541 fail. Luckily, there already exists a function for doing the comparison
3542 by converting both addresses to binary, so now that is used instead of
3543 the text comparison.
3545 PH/20 There was another similar case to PH/19, when a complete host name was
3546 given in a host list; looking up its IP address could give an abbreviated
3547 form, whereas the current host's name might or might not be abbreviated.
3548 The same fix has been applied.
3554 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3555 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3558 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3559 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3561 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3563 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3564 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3565 Also, for "Auto-Submitted", treat anything other than "no" as
3566 non-personal, in accordance with RFC 3834. (Previously it treated
3567 anything starting "auto-" as non-personal.)
3569 TF/01 The control=submission/name=... option had a problem with syntax
3570 errors if the name included a slash character. The /name= option
3571 now slurps the rest of the string, so it can include any characters
3572 but it must come last in the list of options (after /sender_retain
3575 PH/05 Some modifications to the interface to the fake nameserver for the new
3583 TK/01 Added the "success_on_redirect" address verification option. See
3584 NewStuff for rationale and an example.
3586 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3588 PH/02 Patch to exigrep to allow it to work on syslog lines.
3590 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3591 fread() to skip over the body file's header line, because in Cygwin the
3592 header line is locked and is inaccessible.
3594 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3595 co-exist for some time) to make it clear that it is the Exim ID that is
3596 referenced, not the Message-ID: header line.
3598 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3599 string_format() function, because snprintf() does not exist on all
3602 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3603 ${stat: expansion item.
3605 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3606 protocol synchronization error", to keep the pedants happy.
3608 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3609 well as for IRIX systems, when gcc is being used. See the host.c source
3612 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3614 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3617 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3618 transport and to the smtp transport in LMTP mode.
3620 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3622 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3623 run a filter in a subprocess. This could lead to confusion in subsequent
3624 lookups in the parent process. There should also be a search_tidyup() at
3625 the end of the subprocess.
3627 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3628 only if the host matched helo_try_verify_hosts, which caused the
3629 verification to occur when the EHLO/HELO command was issued. The ACL just
3630 tested the remembered result. Now, if a previous verification attempt has
3631 not happened, "verify = helo" does it there and then.
3633 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3635 TK/03 Fix log output including CR from clamd.
3637 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3638 exist provoked a memory error which could cause a segfault.
3640 PH/15 Installed PCRE 6.2
3642 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3644 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3645 of the problem. Specifically, suggested +O2 rather than +O1 for the
3648 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3650 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3651 record, the port was not correctly logged when the outgoing_port log
3652 selector was set (it logged the transort's default port).
3654 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3655 fallback_hosts, and "hosts" in the smtp transport.
3657 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3658 host errors such as "Connection refused".
3660 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3661 authentication with radiusclient 0.4.9:
3663 - Error returned from rc_read_config was caught wrongly
3664 - Username/password not passed on to radius server due to wrong length.
3666 The presumption is that some radiusclient API changes for 4.51/PH/17
3667 were not taken care of correctly. The code is still untested by me (my
3668 Linux distribution still has 0.3.2 of radiusclient), but it was
3669 contributed by a Radius user.
3671 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3672 expanding the "port" option of the smtp transport.
3674 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3675 while reading a MIME header. Thanks to Tom Hughes for a patch.
3677 PH/24 Include config.h inside local_scan.h so that configuration settings are
3680 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3681 that in an ACL for RCPT (for example), you can examine exactly what was
3684 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3685 commands, but it was not correctly comparing the address with the actual
3686 client host address. Thus, it would show the EHLO address in Received:
3687 header lines when this was not necessary.
3689 PH/27 Added the % operator to ${eval:}.
3691 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3692 it should be ignoring failures (because with -C, for example, it has lost
3693 privilege). It wasn't ignoring creation failures other than "already
3696 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3699 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3700 because an input line was too long, either on its own, or by virtue of
3701 too many continuations, the temporary file was not being removed, and the
3702 return code was incorrect.
3704 PH/31 Missing "BOOL" in function definition in filtertest.c.
3706 PH/32 Applied Sieve patches from the maintainer.
3708 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3710 PH/33 Added "verify = not_blind".
3712 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3713 Local/Makefile (with some defaults set). These are used in built scripts
3714 such as exicyclog, but they have never been used in the exim_install
3715 script (though there are many overriding facilities there). I have
3716 arranged that the exim_install script now takes note of these two
3719 PH/35 Installed configuration files for Dragonfly.
3721 PH/36 When a locally submitted message by a trusted user did not contain a
3722 From: header, and the sender address was obtained from -f or from an SMTP
3723 MAIL command, and the trusted user did not use -F to supply a sender
3724 name, $originator_name was incorrectly used when constructing a From:
3725 header. Furthermore, $originator_name was used for submission mode
3726 messages from external hosts without From: headers in a similar way,
3727 which is clearly wrong.
3729 PH/37 Added control=suppress_local_fixups.
3731 PH/38 When log_selector = +received_sender was set, and the addition of the
3732 sender made the log line's construction buffer exactly full, or one byte
3733 less than full, an overflow happened when the terminating "\n" was
3736 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3737 when the result of a list match is failure because a DNS lookup failed.
3739 PH/40 RM_COMMAND is now used in the building process.
3741 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3742 the "build-* directories that it finds.
3744 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3745 address in a domain literal was a prefix of an interface address.
3747 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3748 when verifying a sender address, unless rewrite_headers is false.
3750 PH/44 Wrote a long comment about why errors_to addresses are verified as
3751 recipients, not senders.
3753 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3754 the ratelimit ACL was added.
3756 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3758 PH/46 Added extra information about PostgreSQL errors to the error string.
3760 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3761 avoiding the need to install special zones in a real server. This is
3762 backwards compatible; if it can't find the fake resolver, it drops back.
3763 Thus, both old and new test suites can be run.
3765 TF/02 Added util/ratelimit.pl
3767 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3768 clock is set back in time.
3770 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3771 Candler <B.Candler@pobox.com>.
3773 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3774 OK for addresses that are the result of redirections.
3776 PH/48 A number of further additions for the benefit of the new test suite,
3777 including a fake gethostbyname() that interfaces to the fake DNS resolver
3780 TF/06 The fix for widen_domains has also been applied to qualify_single and
3781 search_parents which are the other dnslookup options that can cause
3784 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3787 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3789 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3790 service" if it encountered an MX record with an empty target hostname.
3791 The message is now "an MX or SRV record indicated no SMTP service".
3793 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3794 if the DNS of the sending site is misconfigured. This is quite a
3795 common situation. This change restores the behaviour of treating a
3796 helo verification defer as a failure.
3798 PH/51 If self=fail was set on a router, the bounce message did not include the
3799 actual error message.
3805 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3807 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3808 command itself ended in error, the underlying message about the transport
3809 filter timeout was being overwritten with the pipe command error. Now the
3810 underlying error message should be appended to the second error message.
3812 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3814 PH/02 Reduce the amount of output that "make" produces by default. Full output
3815 can still be requested.
3817 PH/03 The warning log line about a condition test deferring for a "warn" verb
3818 was being output only once per connection, rather than after each
3819 occurrence (because it was using the same function as for successful
3820 "warn" verbs). This seems wrong, so I have changed it.
3822 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3823 it should not have, which might have caused a crash in the right
3824 circumstances, but probably never did.
3826 PH/04 Installed a modified version of Tony Finch's patch to make submission
3827 mode fix the return path as well as the Sender: header line, and to
3828 add a /name= option so that you can make the user's friendly name appear
3831 TF/03 Added the control = fakedefer ACL modifier.
3833 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3834 Mark Lowes for thorough testing.
3836 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3838 TK/03 Merged latest SRS patch from Miles Wilton.
3840 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3841 with the definition in sysexits.h (which is #included earlier).
3842 Fortunately, Exim does not actually use EX_OK. The code used to try to
3843 preserve the sysexits.h value, by assuming that macro definitions were
3844 scanned for macro replacements. I have been disabused of this notion,
3845 so now the code just undefines EX_OK before #including unistd.h.
3847 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3848 in the smtp transport. When a block could not be written in a single
3849 write() function, the timeout was being re-applied to each part-write.
3850 This seems wrong - if the receiver was accepting one byte at a time it
3851 would take for ever. The timeout is now adjusted when this happens. It
3852 doesn't have to be particularly precise.
3854 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3855 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3857 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3858 without a domain if the check to <postmaster@domain> fails.
3860 SC/01 Eximstats: added -xls and the ability to specify output files
3861 (patch written by Frank Heydlauf).
3863 SC/02 Eximstats: use FileHandles for outputting results.
3865 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3867 SC/04 Eximstats: fixed display of large numbers with -nvr option
3869 SC/05 Eximstats: fixed merging of reports with empty tables.
3871 SC/06 Eximstats: added the -include_original_destination flag
3873 SC/07 Eximstats: removed tabs and trailing whitespace.
3875 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3877 TK/06 MBOX spool code: Add real "From " MBOX separator line
3878 so the .eml file is really in mbox format (even though
3879 most programs do not really care). Patch from Alex Miller.
3881 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3882 The latter is generated from $received_to and is only set if the
3883 message has one envelope recipient. SA can use these headers,
3884 obviously out-of-the-box. Patch from Alex Miller.
3886 PH/08 The ${def test on a variable was returning false if the variable's
3887 value was "0", contrary to what the specification has always said!
3888 The result should be true unless the variable is empty.
3890 PH/09 The syntax error of a character other than { following "${if
3891 def:variable_name" (after optional whitespace) was not being diagnosed.
3892 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3893 accidental colon was present, for example, could give incorrect results.
3895 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3896 result is used (not including appendfile, where other changes are about
3899 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3900 This involved changing a lot of size variables from int to off_t. It
3901 should work with maildirs and everything.
3903 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3904 spamd dying while we are connected to it.
3906 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3909 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3910 function for BDB 4.3.
3912 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3914 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3915 the sources for PCRE have been re-arranged and more files are now
3918 PH/15 The code I had for printing potentially long long variables in PH/11
3919 above was not the best (it lost precision). The length of off_t variables
3920 is now inspected at build time, and an appropriate printing format (%ld
3921 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3922 to be "long long int" or "long int". This is needed for the internal
3923 formatting function string_vformat().
3925 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3926 the configuration file to be ":syslog", then the script "guesses" where
3927 the logs files are, rather than using the compiled in default. In our
3928 case the guess is not the same as the compiled default, so the script
3929 suddenly stopped working when I started to use syslog. The patch checks
3930 to see if log_file_path is "". If so, it attempts to read it from exim
3931 with no configuration file to get the compiled in version, before it
3932 falls back to the previous guessing code."
3934 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3935 implementing BATV in an Exim configuration. See NewStuff for the gory
3938 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3939 Makefile that are specific to HP-UX.
3941 PH/18 If the "use_postmaster" option was set for a recipient callout together
3942 with the "random" option, the postmaster address was used as the MAIL
3943 FROM address for the random test, but not for the subsequent recipient
3944 test. It is now used for both.
3946 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3947 patch removes a few documentation additions to RFC 3028, because the
3948 latest draft now contains them. It adds the new en;ascii-case comparator
3949 and a new error check for 8bit text in MIME parts. Comparator and
3950 require names are now matched exactly. I enabled the subaddress
3951 extension, but it is not well tested yet (read: it works for me)."
3953 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3954 rework some of the code of TK/09 above to avoid the hardwired use of
3955 "%lld" and "long long". Replaced the call to snprintf() with a call to
3958 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3959 records point to non-existent hosts", "retry timeout exceeded", and
3960 "retry time not reached for any host after a long failure period".
3962 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3963 experimental DomainKeys support:
3965 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3966 (2) On an error such as an illegally used "control", the wrong name for
3967 the control was given.
3969 These problems did NOT occur unless DomainKeys support was compiled.
3971 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3973 PH/24 Added ${if match_ip condition.
3975 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3976 databases so that it will be absolutely obvious if a crash occurs in the
3977 DB library. This is a regular occurrence (often caused by mis-matched
3980 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3981 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3982 picked up on a user's system that detects such things. There doesn't seem
3983 to be a gcc warning option for this - only an attribute that has to be
3984 put on the function's prototype. It seems that in Fedora Core 4 they have
3985 set this on a number of new functions. No doubt there will be more in due
3988 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3989 specify a transport. However, if an address that was verified by such a
3990 router was the subject of a callout, Exim crashed because it tried to
3991 read the rcpt_include_affixes from the non-existent transport. Now it
3992 just assumes that the setting of that option is false. This bug was
3993 introduced by 4.51/PH/31.
3995 PH/28 Changed -d+all to exclude +memory, because that information is very
3996 rarely of interest, but it makes the output a lot bigger. People tend to
3997 do -d+all out of habit.
3999 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4000 code in os-type was giving problems when libc.so lives in lib64, like on
4003 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4004 aren't the modern standard, and it seems that some systems' include files
4005 don't always have them. Exim was already checking for some of the newer
4006 ones like T_AAAA, and defining it itself. I've added checks for all the
4007 record types that Exim uses.
4009 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4010 not automatically generating a new one, as it is supposed to. This
4011 prevented TLS from working. If the file did exist, but contained invalid
4012 data, a new version was generated, as expected. It was only the case of a
4013 non-existent file that was broken.
4015 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4016 with a change in libdomainkeys > 0.64.
4018 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4019 from DNS. If the selector record carries the flag, it now has
4020 precedence over the domain-wide flag.
4022 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4024 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4025 the use of a port name that isn't defined in /etc/services) occurred, the
4026 message was deferred as in a normal delivery, and thus remained on the
4027 spool, instead of being failed because of the mua_wrapper setting. This
4028 is now fixed, and I tidied up some of the mua_wrapper messages at the
4031 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4032 the messages in a hash of arrays rather than using individual hashes.
4033 This is a bit cleaner and results in dramatic memory savings, albeit
4034 at a slight CPU cost.
4036 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4037 as requested by Marc Sherman.
4039 SC/10 Eximstats: added histograms for user specified patterns as requested
4042 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4044 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4045 fopen() in the content-scanning modules that did not already have it.
4051 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4052 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4054 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4056 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4058 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4059 to test Sieve filters that use "vacation".
4061 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4062 that changes the way the GnuTLS parameters are stored in the cache file.
4063 The new format can be generated externally. For backward compatibility,
4064 if the data in the cache doesn't make sense, Exim assumes it has read an
4065 old-format file, and it generates new data and writes a new file. This
4066 means that you can't go back to an older release without removing the
4069 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4070 work if there are any delivery delays because "one_time" forces the
4071 parent to be marked "delivered", so its unseen clone is never tried
4072 again. For this reason, Exim now forbids the simultaneous setting of
4075 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4076 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4077 entry if you want to know the details. The fix, however, neglected to
4078 consider the case when local delivery batching is involved. The test for
4079 "previously delivered" was not happening when checking to see if an
4080 address could be batched with a previous (undelivered) one; under
4081 certain circumstances this could lead to multiple deliveries to the same
4084 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4085 in its include files, and this causes problems building Exim.
4087 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4088 header_syntax) but Exim was just ignoring anything given after a slash.
4089 In particular, this caused confusion with an attempt to use "verify =
4090 reverse_host_lookup/defer_ok". An error is now given when options are
4091 supplied for verify items that do not have them. (Maybe reverse_host_
4092 lookup should have a defer_ok option, but that's a different point.)
4094 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4095 defined by RFC 821) to 2048, because there were problems with some AUTH
4096 commands, and RFC 1869 says the size should be increased for extended
4097 SMTP commands that take arguments.
4099 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4102 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4103 "unknown" error; now it says that the functionality isn't in the binary.
4105 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4106 an address' error message when a string expansion fails (syntax or
4107 whatever). Otherwise the password may appear in the log. Following change
4108 PH/42 below, there is no longer a chance of it appearing in a bounce
4111 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4113 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4114 of its addresses was ever tried. (Bugzilla bug #2.)
4116 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4117 the result incorrectly in the debug output. (It correctly added a newline
4118 to what was transported.)
4120 TF/01 Added $received_time.
4122 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4123 commented out examples of how to interface to a virus scanner and to
4124 SpamAssassin. Also added commented examples of av_scanner and
4125 spamd_address settings.
4127 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4128 and controls are allowed in which ACLs. There were a couple of minor
4129 errors. Some of the entries in the conditions table (which is a table of
4130 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4131 negation of where the condition IS allowed.
4133 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4135 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4136 header file does not have a version number, so I've had to invent a new
4137 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4138 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4139 radiusclient), but it was contributed by a Radius user.
4141 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4142 files or directories, for OpenSSL.
4144 PH/19 When an Exim process that is running as root has to create an Exim log
4145 file, it does so in a subprocess that runs as exim:exim so as to get the
4146 ownership right at creation (otherwise, other Exim processes might see
4147 the file with the wrong ownership). There was no test for failure of this
4148 fork() call, which would lead to the process getting stuck as it waited
4149 for a non-existent subprocess. Forks do occasionally fail when resources
4150 run out. I reviewed all the other calls to fork(); they all seem to check
4153 PH/20 When checking for unexpected SMTP input at connect time (before writing
4154 the banner), Exim was not dealing correctly with a non-positive return
4155 from the read() function. If the client had disconnected by this time,
4156 the result was a log entry for a synchronization error with an empty
4157 string after "input=" when read() returned zero. If read() returned -1
4158 (an event I could not check), uninitialized data bytes were printed.
4159 There were reports of junk text (parts of files, etc) appearing after
4162 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4164 PH/22 Added support for macro redefinition, and (re)definition in between
4165 driver and ACL definitions.
4167 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4168 forgetting to use the resulting value; it was using the unexpanded value.
4170 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4171 hadn't been configured. The fix is from Juergen Kreileder, who
4172 understands it better than I do:
4174 "Here's what I see happening with three configured cyrus_sasl
4175 authenticators configured (plain, login, cram-md5):
4177 On startup auth_cyrus_sasl_init() gets called for each of these.
4178 This means three calls to sasl_listmech() without a specified mech_list.
4179 => SASL tests which mechs of all available mechs actually work
4180 => three warnings about OTP not working
4181 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4183 With the patch, sasl_listmech() also gets called three times. But now
4184 SASL's mech_list option is set to the server_mech specified in the the
4185 authenticator. Or in other words, the answer from sasl_listmech()
4186 gets limited to just the mech you're testing for (which is different
4188 => the return list contains just 'plain' or 'login', 'cram-md5' or
4189 nothing depending on the value of ob->server_mech.
4191 I've just tested the patch: Authentication still works fine,
4192 unavailable mechs specified in the exim configuration are still
4193 caught, and the auth.log warnings about OTP are gone."
4195 PH/25 When debugging is enabled, the contents of the command line are added
4196 to the debugging output, even when log_selector=+arguments is not
4199 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4200 answer is "GNU", and only if the return is "GNU/something" is the answer
4203 PH/27 $acl_verify_message is now set immediately after the failure of a
4204 verification in an ACL, and so is available in subsequent modifiers. In
4205 particular, the message can be preserved by coding like this:
4207 warn !verify = sender
4208 set acl_m0 = $acl_verify_message
4210 Previously, $acl_verify_message was set only while expanding "message"
4211 and "log_message" when a very denied access.
4213 PH/28 Modified OS/os.c-Linux with
4215 -#ifndef OS_LOAD_AVERAGE
4216 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4218 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4219 nomenclature these days.)
4221 PH/29 Installed patch from the Sieve maintainer that adds the options
4222 sieve_useraddress and sieve_subaddress to the redirect router.
4224 PH/30 In these circumstances:
4225 . Two addresses routed to the same list of hosts;
4226 . First host does not offer TLS;
4227 . First host accepts first address;
4228 . First host gives temporary error to second address;
4229 . Second host offers TLS and a TLS session is established;
4230 . Second host accepts second address.
4231 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4232 and peerdn, if requested) that were in fact used only for the second
4235 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4236 attention to any local part prefix or suffix that was matched by the
4237 router that accepted the address. It now behaves in the same way as it
4238 does for delivery: the affixes are removed from the local part unless
4239 rcpt_include_affixes is set on the transport.
4241 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4242 timeout during the DATA phase of an incoming message.
4244 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4245 applied a patch sanctioned by the Sieve maintainer.
4247 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4248 the uid or gid is negative. A case of a negative gid caused this to be
4249 noticed. The fix allows for either to be negative.
4251 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4252 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4255 PH/36 The change PH/12 above was broken. Fixed it.
4257 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4258 the grounds that routing the same address twice would always produce the
4259 same answer. This might have been true once, but it is certainly no
4260 longer true now. Routing a child address may depend on the previous
4261 routing that produced that child. Some complicated redirection strategies
4262 went wrong when messages had multiple recipients, and made Exim's
4263 behaviour dependent on the order in which the addresses were given.
4265 I have moved the duplicate checking until after the routing is complete.
4266 Exim scans the addresses that are assigned to local and remote
4267 transports, and removes any duplicates. This means that more work will be
4268 done, as duplicates will always all be routed, but duplicates are
4269 presumably rare, so I don't expect this is of any significance.
4271 For deliveries to pipes, files, and autoreplies, the duplicate checking
4272 still happens during the routing process, since they are not going to be
4275 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4276 It corrects a timeout issue with spamd. This is Ian's comment: "The
4277 background is that sometimes spamd either never reads data from a
4278 connection it has accepted, or it never writes response data. The exiscan
4279 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4280 blindly assumes that writes won't block so it may never time out."
4282 PH/39 Allow G after quota size as well as K and M.
4284 PH/40 The value set for $authenticated_id in an authenticator may not contain
4285 binary zeroes or newlines because the value is written to log lines and
4286 to spool files. There was no check on this. Now the value is run through
4287 the string_printing() function so that such characters are converted to
4288 printable escape sequences.
4290 PH/41 $message_linecount is a new variable that contains the total number of
4291 lines in the message. Compare $body_linecount, which is the count for the
4294 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4295 bounce and delay warning messages, except in certain special cases, which
4298 (a) An SMTP error message from a remote host;
4299 (b) A message specified in a :fail: redirection;
4300 (c) A message specified in a "fail" command in a system filter;
4301 (d) A message specified in a FAIL return from the queryprogram router;
4302 (e) A message specified by the cannot_route_message router option.
4304 In these cases only, Exim does include the error details in bounce and
4305 warning messages. There are also a few cases where bland messages such
4306 as "unrouteable address" or "local delivery error" are given.
4308 PH/43 $value is now also set for the "else" part of a ${run expansion.
4310 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4311 being worked on, but at least Exim now implements the latest version to
4314 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4315 process to complete was treated as a delivery failure, a timeout while
4316 writing the message to the pipe was logged, but erroneously treated as a
4317 successful delivery. Such timeouts include transport filter timeouts. For
4318 consistency with the overall process timeout, these timeouts are now
4319 treated as errors, giving rise to delivery failures by default. However,
4320 there is now a new Boolean option for the pipe transport called
4321 timeout_defer, which, if set TRUE, converts the failures into defers for
4322 both kinds of timeout. A transport filter timeout is now identified in
4325 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4326 systems where "make" and "gmake" are different, calling "gmake" at top
4327 level broke things. I've arranged for the value of $(MAKE) to be passed
4328 from the Makefile to this script so that it can call the same version of
4332 A note about Exim versions 4.44 and 4.50
4333 ----------------------------------------
4335 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4336 changes of various kinds. As a consequence, a big documentation update was
4337 needed. This delayed the release for rather longer than seemed good, especially
4338 in the light of a couple of (minor) security issues. Therefore, the changes
4339 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4340 release. So 4.44 and 4.50 are in effect two different branches that both start
4343 I have left the 4.50 change log unchanged; it contains all the changes since
4344 4.43. The change log for 4.44 is below; many of its items are identical to
4345 those for 4.50. This seems to be the most sensible way to preserve the
4346 historical information.
4352 1. Minor wording change to the doc/README.SIEVE file.
4354 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4355 computation of the current number of files was incorrect.
4357 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4358 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4361 4. Give more explanation in the error message when the command for a transport
4362 filter fails to execute.
4364 5. There are several places where Exim runs a non-Exim command in a
4365 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4366 was being done only for the command run by the queryprogram router. It is
4367 now done for all such subprocesses. The other cases are: ${run, transport
4368 filters, and the commands run by the lmtp and pipe transports.
4370 6. Added CONFIGURE_GROUP build-time option.
4372 7. Some older OS have a limit of 256 on the maximum number of file
4373 descriptors. Exim was using setrlimit() to set 1000 as a large value
4374 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4375 systems. I've change it so that if it can't get 1000, it tries for 256.
4377 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4378 was an oversight, and furthermore, ever since the addition of extra
4379 controls (e.g. 4.43/32), the checks on when to allow different forms of
4380 "control" were broken. There should now be diagnostics for all cases when a
4381 control that does not make sense is encountered.
4383 9. Added the /retain_sender option to "control=submission".
4385 10. $recipients is now available in the predata ACL (oversight).
4387 11. Tidy the search cache before the fork to do a delivery from a message
4388 received from the command line. Otherwise the child will trigger a lookup
4389 failure and thereby defer the delivery if it tries to use (for example) a
4390 cached ldap connection that the parent has called unbind on.
4392 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4393 of $address_data from the recipient verification was clobbered by the
4394 sender verification.
4396 13. The value of address_data from a sender verification is now available in
4397 $sender_address_data in subsequent conditions in the ACL statement.
4399 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4401 15. Added a new option "connect=<time>" to callout options, to set a different
4404 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4405 was its contents. (It was OK if the option was not defined at all.)
4407 17. A "Completed" log line is now written for messages that are removed from
4408 the spool by the -Mrm option.
4410 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4411 information about exactly what failed.
4413 19. Added -dd to debug only the daemon process.
4415 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4416 handles timeouts, both on the server side and network timeouts. Renamed the
4417 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4419 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4420 It is now set to "smtps".
4422 22. $host_address is now set to the target address during the checking of
4423 ignore_target_hosts.
4425 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4426 being passed; this would have caused $sender_host_name to have been used if
4427 matching the list had actually called for a host name (not very likely,
4428 since this list is usually IP addresses). A host name is now passed as
4431 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4432 code that specifies a non-restarting handler (typically sigaction() in
4433 modern systems) in an attempt to fix a rare and obscure crash bug.
4435 25. Narrowed the window for a race in the daemon that could cause it to ignore
4436 SIGCHLD signals. This is not a major problem, because they are used only to
4437 wake it up if nothing else does.
4439 26. A malformed maildirsize file could cause Exim to calculate negative values
4440 for the mailbox size or file count. Odd effects could occur as a result.
4441 The maildirsize information is now recalculated if the size or filecount
4444 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4445 support for a long time. Removed HAVE_SYS_VFS_H.
4447 28. Installed the latest version of exipick from John Jetmore.
4449 29. In an address list, if the pattern was not a regular expression, an empty
4450 subject address (from a bounce message) matched only if the pattern was an
4451 empty string. Non-empty patterns were not even tested. This was the wrong
4452 because it is perfectly reasonable to use an empty address as part of a
4453 database query. An empty address is now tested by patterns that are
4454 lookups. However, all the other forms of pattern expect the subject to
4455 contain a local part and a domain, and therefore, for them, an empty
4456 address still always fails if the pattern is not itself empty.
4458 30. Exim went into a mad DNS loop when attempting to do a callout where the
4459 host was specified on an smtp transport, and looking it up yielded more
4460 than one IP address.
4462 31. Re-factored the code for checking spool and log partition space into a
4463 function that finds that data and another that does the check. The former
4464 is then used to implement four new variables: $spool_space, $log_space,
4465 $spool_inodes, and $log_inodes.
4467 32. The RFC2047 encoding function was originally intended for short strings
4468 such as real names; it was not keeping to the 75-character limit for
4469 encoded words that the RFC imposes. It now respects the limit, and
4470 generates multiple encoded words if necessary. To be on the safe side, I
4471 have increased the buffer size for the ${rfc2047: expansion operator from
4474 33. It is now permitted to omit both strings after an "if" condition; if the
4475 condition is true, the result is "true". As before, when the second string
4476 is omitted, a false condition yields an empty string. This makes it less
4477 cumbersome to write custom ACL and router conditions.
4479 34. Failure to deliver a bounce message always caused it to be frozen, even if
4480 there was an errors_to setting on the router. The errors_to setting is now
4483 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4484 canonical form (fully expanded) before being placed in
4485 $sender_host_address.
4487 36. The table in the code that translates DNS record types into text (T_A to
4488 "A" for instance) was missing entries for NS and CNAME. It is just possible
4489 that this could have caused confusion if both these types were looked up
4490 for the same domain, because the text type is used as part of Exim's
4491 per-process caching. But the chance of anyone hitting this buglet seems
4494 37. The dnsdb lookup has been extended in a number of ways.
4496 (1) There is a new type, "zns", which walks up the domain tree until it
4497 finds some nameserver records. It should be used with care.
4499 (2) There is a new type, "mxh", which is like "mx" except that it returns
4500 just the host names, not the priorities.
4502 (3) It is now possible to give a list of domains (or IP addresses) to be
4503 looked up. The behaviour when one of the lookups defers can be
4504 controlled by a keyword.
4506 (4) It is now possible to specify the separator character for use when
4507 multiple records are returned.
4509 38. The dnslists ACL condition has been extended: it is now possible to supply
4510 a list of IP addresses and/or domains to be looked up in a particular DNS
4513 39. Added log_selector=+queue_time_overall.
4515 40. When running the queue in the test harness, wait just a tad after forking a
4516 delivery process, to get repeatability of debugging output.
4518 41. Include certificate and key file names in error message when GnuTLS fails
4519 to set them up, because the GnuTLS error message doesn't include the name
4520 of the failing file when there is a problem reading it.
4522 42. Allow both -bf and -bF in the same test run.
4524 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4526 44. The "Exiscan patch" is now merged into the mainline Exim source.
4528 45. Sometimes the final signoff response after QUIT could fail to get
4529 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4530 < 0 before doing a fflush(). This bug looks as though it goes back to the
4531 introduction of TLS in release 3.20, but "sometimes" must have been rare
4532 because the tests only now provoked it.
4534 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4535 (this can affect the format of dates).
4537 47. exim_tidydb, when checking for the continued existence of a message for
4538 which it has found a message-specific retry record, was not finding
4539 messages that were in split spool directories. Consequently, it was
4540 deleting retry records that should have stayed in existence.
4542 48. Steve fixed some bugs in eximstats.
4544 49. The SPA authentication driver was not abandoning authentication and moving
4545 on to the next authenticator when an expansion was forced to fail,
4546 contradicting the general specification for all authenticators. Instead it
4547 was generating a temporary error. It now behaves as specified.
4549 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4550 (the order specifies the preference for clients). The order is now AES256,
4551 AES128, 3DES, ARCFOUR128.
4553 51. Small patch to Sieve code - explicitly set From: when generating an
4556 52. Exim crashed if a remote delivery caused a very long error message to be
4557 recorded - for instance if somebody sent an entire SpamAssassin report back
4558 as a large number of 550 error lines. This bug was coincidentally fixed by
4559 increasing the size of one of Exim's internal buffers (big_buffer) that
4560 happened as part of the Exiscan merge. However, to be on the safe side, I
4561 have made the code more robust (and fixed the comments that describe what
4564 53. Now that there can be additional text after "Completed" in log lines (if
4565 the queue_time_overall log selector is set), a one-byte patch to exigrep
4566 was needed to allow it to recognize "Completed" as not the last thing in
4569 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4570 patch that reportedly fixes this has been added. I am not expert enough to
4571 create a test for it. This is what the patch creator wrote:
4573 "I found a little strange behaviour of ldap code when working with
4574 Windows 2003 AD Domain, where users was placed in more than one
4575 Organization Units. When I tried to give exim partial DN, the exit code
4576 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4577 But simultaneously result of request was absolutely normal ldap result,
4578 so I produce this patch..."
4580 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4581 REFERENCE, so I have modified the code to exclude the patch when that macro
4584 55. Some experimental protocols are using DNS PTR records for new purposes. The
4585 keys for these records are domain names, not reversed IP addresses. The
4586 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4587 leaves it alone. Component reversal etc. now happens only for IP addresses.
4590 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4592 57. Double the size of the debug message buffer (to 2048) so that more of very
4593 long debug lines gets shown.
4595 58. The exicyclog utility now does better if the number of log files to keep
4596 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4598 59. Two changes related to the smtp_active_hostname option:
4600 (1) $smtp_active_hostname is now available as a variable.
4601 (2) The default for smtp_banner uses $smtp_active_hostname instead
4602 of $primary_hostname.
4604 60. The host_aton() function is supposed to be passed a string that is known
4605 to be a valid IP address. However, in the case of IPv6 addresses, it was
4606 not checking this. This is a hostage to fortune. Exim now panics and dies
4607 if the condition is not met. A case was found where this could be provoked
4608 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4609 components; fortuitously, this particular loophole had already been fixed
4610 by change 4.50/55 above.
4612 If there are any other similar loopholes, the new check in host_aton()
4613 itself should stop them being exploited. The report I received stated that
4614 data on the command line could provoke the exploit when Exim was running as
4615 exim, but did not say which command line option was involved. All I could
4616 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4617 running as the user.
4620 61. There was a buffer overflow vulnerability in the SPA authentication code
4621 (which came originally from the Samba project). I have added a test to the
4622 spa_base64_to_bits() function which I hope fixes it.
4625 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4626 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4628 63. The daemon start-up calls getloadavg() while still root for those OS that
4629 need the first call to be done as root, but it missed one case: when
4630 deliver_queue_load_max is set with deliver_drop_privilege. This is
4631 necessary for the benefit of the queue runner, because there is no re-exec
4632 when deliver_drop_privilege is set.
4634 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4635 This has been fixed.
4637 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4638 was in use, was not putting the data itself into the right store pool;
4639 consequently, it could be overwritten for a subsequent message in the same
4640 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4643 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4645 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4646 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4647 the function were treating the return as a boolean value, which happened to
4648 work because 0=false and not-0=true, but is not correct code.
4650 68. The host_aton() function was not handling scoped IPv6 addresses (those
4651 with, for example, "%eth0" on the end) correctly.
4653 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4654 negated items (that is, ~something) in unsigned ints. Some compilers
4655 apparently mutter when there is no cast.
4657 70. If an address verification called from an ACL failed, and did not produce a
4658 user-specific message (i.e. there was only a "system" message), nothing was
4659 put in $acl_verify_message. In this situation, it now puts the system
4662 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4663 session; change 4.31/43 added the unwanted input to the log line - except
4664 that it did not do this in the start of session case. It now does.
4666 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4667 This is wrong and can cause the other end to generate a synchronization
4668 error if it is another Exim or anything else that does the synchronization
4669 check. A QUIT command is no longer sent after a timeout.
4671 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4672 during host lookups.
4674 74. The defer_ok option of callout verification was not working if it was used
4675 when verifying addresses in header lines, that is, for this case:
4677 verify = header_sender/callout=defer_ok
4679 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4680 those file descriptors could be used for SMTP connections. If anything
4681 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4682 could be sent to the SMTP client, causing chaos. The daemon now opens
4683 stdin, stdout, and stderr to /dev/null when it puts itself into the
4686 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4687 log by default. The user can override this with suitable Perl magic.
4689 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4690 the log message when discard triggers, was not working for the DATA ACL or
4691 for the non-SMTP ACL.
4693 78. Error message wording change in sieve.c.
4695 79. If smtp_accept_max_per_host was set, the number of connections could be
4696 restricted to fewer than expected, because the daemon was trying to set up
4697 a new connection before checking whether the processes handling previous
4698 connections had finished. The check for completed processes is now done
4699 earlier. On busy systems, this bug wouldn't be noticed because something
4700 else would have woken the daemon, and it would have reaped the completed
4703 80. If a message was submitted locally by a user whose login name contained one
4704 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4705 It caused a spool format error. I have fixed the spool reading code. A
4706 related problem was that the "from" clause in the Received: line became
4707 illegal because of the space(s). It is now covered by ${quote_local_part.
4709 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4712 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4714 83. Patch from Sieve maintainer to fix unterminated string problem in
4715 "vacation" handling.
4717 84. Some minor changes to the Linux configuration files to help with other
4718 OS variants using glibc.
4720 85. One more patch for Sieve to update vacation handling to latest spec.
4723 ----------------------------------------------------
4724 See the note above about the 4.44 and 4.50 releases.
4725 ----------------------------------------------------
4731 1. Change 4.43/35 introduced a bug that caused file counts to be
4732 incorrectly computed when quota_filecount was set in an appendfile
4735 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4736 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4739 3. Give more explanation in the error message when the command for a transport
4740 filter fails to execute.
4742 4. There are several places where Exim runs a non-Exim command in a
4743 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4744 was being done only for the command run by the queryprogram router. It is
4745 now done for all such subprocesses. The other cases are: ${run, transport
4746 filters, and the commands run by the lmtp and pipe transports.
4748 5. Some older OS have a limit of 256 on the maximum number of file
4749 descriptors. Exim was using setrlimit() to set 1000 as a large value
4750 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4751 systems. I've change it so that if it can't get 1000, it tries for 256.
4753 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4754 was an oversight, and furthermore, ever since the addition of extra
4755 controls (e.g. 4.43/32), the checks on when to allow different forms of
4756 "control" were broken. There should now be diagnostics for all cases when a
4757 control that does not make sense is encountered.
4759 7. $recipients is now available in the predata ACL (oversight).
4761 8. Tidy the search cache before the fork to do a delivery from a message
4762 received from the command line. Otherwise the child will trigger a lookup
4763 failure and thereby defer the delivery if it tries to use (for example) a
4764 cached ldap connection that the parent has called unbind on.
4766 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4767 of $address_data from the recipient verification was clobbered by the
4768 sender verification.
4770 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4771 was its contents. (It was OK if the option was not defined at all.)
4773 11. A "Completed" log line is now written for messages that are removed from
4774 the spool by the -Mrm option.
4776 12. $host_address is now set to the target address during the checking of
4777 ignore_target_hosts.
4779 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4780 being passed; this would have caused $sender_host_name to have been used if
4781 matching the list had actually called for a host name (not very likely,
4782 since this list is usually IP addresses). A host name is now passed as
4785 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4786 code that specifies a non-restarting handler (typically sigaction() in
4787 modern systems) in an attempt to fix a rare and obscure crash bug.
4789 15. Narrowed the window for a race in the daemon that could cause it to ignore
4790 SIGCHLD signals. This is not a major problem, because they are used only to
4791 wake it up if nothing else does.
4793 16. A malformed maildirsize file could cause Exim to calculate negative values
4794 for the mailbox size or file count. Odd effects could occur as a result.
4795 The maildirsize information is now recalculated if the size or filecount
4798 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4799 support for a long time. Removed HAVE_SYS_VFS_H.
4801 18. Updated exipick to current release from John Jetmore.
4803 19. Allow an empty sender to be matched against a lookup in an address list.
4804 Previously the only cases considered were a regular expression, or an
4807 20. Exim went into a mad DNS lookup loop when doing a callout where the
4808 host was specified on the transport, if the DNS lookup yielded more than
4811 21. The RFC2047 encoding function was originally intended for short strings
4812 such as real names; it was not keeping to the 75-character limit for
4813 encoded words that the RFC imposes. It now respects the limit, and
4814 generates multiple encoded words if necessary. To be on the safe side, I
4815 have increased the buffer size for the ${rfc2047: expansion operator from
4818 22. Failure to deliver a bounce message always caused it to be frozen, even if
4819 there was an errors_to setting on the router. The errors_to setting is now
4822 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4823 canonical form (fully expanded) before being placed in
4824 $sender_host_address.
4826 24. Updated eximstats to version 1.33
4828 25. Include certificate and key file names in error message when GnuTLS fails
4829 to set them up, because the GnuTLS error message doesn't include the name
4830 of the failing file when there is a problem reading it.
4832 26. Expand error message when OpenSSL has problems setting up cert/key files.
4835 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4836 (this can affect the format of dates).
4838 28. exim_tidydb, when checking for the continued existence of a message for
4839 which it has found a message-specific retry record, was not finding
4840 messages that were in split spool directories. Consequently, it was
4841 deleting retry records that should have stayed in existence.
4843 29. eximstats updated to version 1.35
4844 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4845 1.35 - bugfix such that pie charts by volume are generated correctly
4847 30. The SPA authentication driver was not abandoning authentication and moving
4848 on to the next authenticator when an expansion was forced to fail,
4849 contradicting the general specification for all authenticators. Instead it
4850 was generating a temporary error. It now behaves as specified.
4852 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4853 (the order specifies the preference for clients). The order is now AES256,
4854 AES128, 3DES, ARCFOUR128.
4856 31. Small patch to Sieve code - explicitly set From: when generating an
4859 32. Exim crashed if a remote delivery caused a very long error message to be
4860 recorded - for instance if somebody sent an entire SpamAssassin report back
4861 as a large number of 550 error lines. This bug was coincidentally fixed by
4862 increasing the size of one of Exim's internal buffers (big_buffer) that
4863 happened as part of the Exiscan merge. However, to be on the safe side, I
4864 have made the code more robust (and fixed the comments that describe what
4867 33. Some experimental protocols are using DNS PTR records for new purposes. The
4868 keys for these records are domain names, not reversed IP addresses. The
4869 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4870 leaves it alone. Component reversal etc. now happens only for IP addresses.
4873 34. The host_aton() function is supposed to be passed a string that is known
4874 to be a valid IP address. However, in the case of IPv6 addresses, it was
4875 not checking this. This is a hostage to fortune. Exim now panics and dies
4876 if the condition is not met. A case was found where this could be provoked
4877 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4878 components; fortuitously, this particular loophole had already been fixed
4879 by change 4.50/55 or 4.44/33 above.
4881 If there are any other similar loopholes, the new check in host_aton()
4882 itself should stop them being exploited. The report I received stated that
4883 data on the command line could provoke the exploit when Exim was running as
4884 exim, but did not say which command line option was involved. All I could
4885 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4886 running as the user.
4889 35. There was a buffer overflow vulnerability in the SPA authentication code
4890 (which came originally from the Samba project). I have added a test to the
4891 spa_base64_to_bits() function which I hope fixes it.
4894 36. The daemon start-up calls getloadavg() while still root for those OS that
4895 need the first call to be done as root, but it missed one case: when
4896 deliver_queue_load_max is set with deliver_drop_privilege. This is
4897 necessary for the benefit of the queue runner, because there is no re-exec
4898 when deliver_drop_privilege is set.
4900 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4901 was in use, was not putting the data itself into the right store pool;
4902 consequently, it could be overwritten for a subsequent message in the same
4903 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4906 38. Sometimes the final signoff response after QUIT could fail to get
4907 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4908 < 0 before doing a fflush(). This bug looks as though it goes back to the
4909 introduction of TLS in release 3.20, but "sometimes" must have been rare
4910 because the tests only now provoked it.
4916 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4917 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4918 Now it returns an integer. A number of calls were still expecting a T/F
4919 return. Fortuitously, in all cases, the tests worked in OK situations,
4920 which is the norm. However, things would have gone wrong on any write
4921 failures on the smtp file descriptor. This function is used when sending
4922 messages over SMTP and also when doing verify callouts.
4924 2. When Exim is called to do synchronous delivery of a locally submitted
4925 message (the -odf or -odi options), it no longer closes stderr before doing
4928 3. Implemented the mua_wrapper option.
4930 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4932 5. Implemented the functions header_remove(), header_testname(),
4933 header_add_at_position(), and receive_remove_recipient(), and exported them
4936 6. If an ACL "warn" statement specified the addition of headers, Exim already
4937 inserted X-ACL-Warn: at the start if there was no header name. However, it
4938 was not making this test for the second and subsequent header lines if
4939 there were newlines in the string. This meant that an invalid header could
4940 be inserted if Exim was badly configured.
4942 7. Allow an ACL "warn" statement to add header lines at the start or after all
4943 the Received: headers, as well as at the end.
4945 8. Added the rcpt_4xx retry error code.
4947 9. Added postmaster_mailfrom=xxx to callout verification option.
4949 10. Added mailfrom=xxxx to the callout verification option, for verify=
4952 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4953 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4955 12. Inserted some casts to stop certain compilers warning when using pointer
4956 differences as field lengths or precisions in printf-type calls (mostly
4957 affecting debugging statements).
4959 13. Added optional readline() support for -be (dynamically loaded).
4961 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4962 same clock tick as a message's arrival, so that its received time was the
4963 same as the "first fail" time on the retry record, and that message
4964 remained on the queue past the ultimate address timeout, every queue runner
4965 would try a delivery (because it was past the ultimate address timeout) but
4966 after another failure, the ultimate address timeout, which should have then
4967 bounced the address, did not kick in. This was a "< instead of <=" error;
4968 in most cases the first failure would have been in the next clock tick
4969 after the received time, and all would be well.
4971 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4972 being recognized when the domain list was tested by the match_domain
4973 condition in an expansion string.
4975 16. Added the ${str2b64: operator.
4977 17. Exim was always calling setrlimit() to set a large limit for the number of
4978 processes, without checking whether the existing limit was already
4979 adequate. (It did check for the limit on file descriptors.) Furthermore,
4980 errors from getrlimit() and setrlimit() were being ignored. Now they are
4981 logged to the main and panic logs, but Exim does carry on, to try to do its
4982 job under whatever limits there are.
4984 18. Imported PCRE 5.0.
4986 19. Trivial typo in log message " temporarily refused connection" (the leading
4989 20. If the log selector return_path_on_delivery was set and an address was
4990 redirected to /dev/null, the delivery process crashed because it assumed
4991 that a return path would always be set for a "successful" delivery. In this
4992 case, the whole delivery is bypassed as an optimization, and therefore no
4995 21. Internal re-arrangement: the function for sending a challenge and reading
4996 a response while authentication was assuming a zero-terminated challenge
4997 string. It's now changed to take a pointer and a length, to allow for
4998 binary data in such strings.
5000 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5002 23. Exim was not respecting finduser_retries when seeking the login of the
5003 uid under which it was called; it was always trying 10 times. (The default
5004 setting of finduser_retries is zero.) Also, it was sleeping after the final
5005 failure, which is pointless.
5007 24. Implemented tls_on_connect_ports.
5009 25. Implemented acl_smtp_predata.
5011 26. If the domain in control=submission is set empty, Exim assumes that the
5012 authenticated id is a complete email address when it generates From: or
5013 Sender: header lines.
5015 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5016 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5017 chown and chgrp in /bin and hostname in /usr/bin.
5019 28. Exim was keeping the "process log" file open after each use, just as it
5020 does for the main log. This opens the possibility of it remaining open for
5021 long periods when the USR1 signal hits a daemon. Occasional processlog
5022 errors were reported, that could have been caused by this. Anyway, it seems
5023 much more sensible not to leave this file open at all, so that is what now
5026 29. The long-running daemon process does not normally write to the log once it
5027 has entered its main loop, and it closes the log before doing so. This is
5028 so that log files can straightforwardly be renamed and moved. However,
5029 there are a couple of unusual error situations where the daemon does write
5030 log entries, and I had neglected to close the log afterwards.
5032 30. The text of an SMTP error response that was received during a remote
5033 delivery was being truncated at 512 bytes. This is too short for some of
5034 the long messages that one sometimes sees. I've increased the limit to
5037 31. It is now possible to make retry rules that apply only when a message has a
5038 specific sender, in particular, an empty sender.
5040 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5041 it possible to be selective about when SMTP synchronization is enforced.
5043 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5045 32. Implemented hosts_connection_nolog.
5047 33. Added an ACL for QUIT.
5049 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5052 35. Added mailbox_size and mailbox_filecount to appendfile.
5054 36. Added control = no_multiline_responses to ACLs.
5056 37. There was a bug in the logic of the code that waits for the clock to tick
5057 in the case where the clock went backwards by a substantial amount such
5058 that the microsecond fraction of "now" was more than the microsecond
5059 fraction of "then" (but the whole seconds number was less).
5061 38. Added support for the libradius Radius client library this is found on
5062 FreeBSD (previously only the radiusclient library was supported).
5068 1. When certain lookups returned multiple values in the form name=value, the
5069 quoting of the values was not always being done properly. Specifically:
5070 (a) If the value started with a double quote, but contained no whitespace,
5072 (b) If the value contained whitespace other than a space character (i.e.
5073 tabs or newlines or carriage returns) it was not quoted.
5074 This fix has been applied to the mysql and pgsql lookups by writing a
5075 separate quoting function and calling it from the lookup code. The fix
5076 should probably also be applied to nisplus, ibase and oracle lookups, but
5077 since I cannot test any of those, I have not disturbed their existing code.
5079 2. A hit in the callout cache for a specific address caused a log line with no
5080 reason for rejecting RCPT. Now it says "Previous (cached) callout
5081 verification failure".
5083 3. There was an off-by-one bug in the queryprogram router. An over-long
5084 return line was truncated at 256 instead of 255 characters, thereby
5085 overflowing its buffer with the terminating zero. As well as fixing this, I
5086 have increased the buffer size to 1024 (and made a note to document this).
5088 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5089 when Exim is waiting for an SMTP response from a remote server, Exim
5090 restarts its select() call on the socket, thereby resetting its timeout.
5091 This is not a problem when such interrupts are rare. Somebody set up a cron
5092 job to run exiwhat every 2 minutes, which is less than the normal select()
5093 timeout (5 or 10 minutes). This meant that the select() timeout never
5094 kicked in because it was always reset. I have fixed this by comparing the
5095 time when an interrupt arrives with the time at the start of the first call
5096 to select(). If more time than the timeout has elapsed, the interrupt is
5097 treated as a timeout.
5099 5. Some internal re-factoring in preparation for the addition of Sieve
5100 extensions (by MH). In particular, the "personal" test is moved to a
5101 separate function, and given an option for scanning Cc: and Bcc: (which is
5102 not set for Exim filters).
5104 6. When Exim created an email address using the login of the caller as the
5105 local part (e.g. when creating a From: or Sender: header line), it was not
5106 quoting the local part when it contained special characters such as @.
5108 7. Installed new OpenBSD configuration files.
5110 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5111 try to make them clearer.
5113 9. Callout options, other than the timeout value, were being ignored when
5114 verifying sender addresses in header lines. For example, when using
5116 verify = header_sender/callout=no_cache
5118 the cache was (incorrectly) being used.
5120 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5121 only the Cygwin environment.
5123 11. When return_path_on_delivery was set as a log selector, if different remote
5124 addresses in the same message used different return paths and parallel
5125 remote delivery occurred, the wrong values would sometimes be logged.
5126 (Whenever a remote delivery process finished, the return path value from
5127 the most recently started remote delivery process was logged.)
5129 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5130 lines when AUTH and/or TLS are in use. This is the "received protocol"
5131 field. Exim used to use "asmtp" for authenticated SMTP, without any
5132 indication (in the protocol name) for TLS use. Now it follows the RFC and
5133 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5134 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5135 names appear in log lines as well as in Received: header lines.
5137 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5138 extensions, and comparison tests, and to fix some bugs.
5140 14. Changes to the "personal" filter test:
5142 (1) The test was buggy in that it was just doing the equivalent of
5143 "contains" tests on header lines. For example, if a user's address was
5144 anne@some.where, the "personal" test would incorrectly be true for
5146 To: susanne@some.where
5148 This test is now done by extracting each address from the header in turn,
5149 and checking the entire address. Other tests that are part of "personal"
5150 are now done using regular expressions (for example, to check local parts
5151 of addresses in From: header lines).
5153 (2) The list of non-personal local parts in From: addresses has been
5154 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5155 taken from the Sieve specification recommendations.
5157 (3) If the message contains any header line starting with "List-" it is
5158 treated as non-personal.
5160 (4) The test for "circular" in the Subject: header line has been removed
5161 because it now seems ill-conceived.
5163 15. Minor typos in src/EDITME comments corrected.
5165 16. Installed latest exipick from John Jetmore.
5167 17. If headers_add on a router specified a text string that was too long for
5168 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5169 of string_sprintf() is now avoided.
5171 18. $message_body_size was not set (it was always zero) when running the DATA
5172 ACL and the local_scan() function.
5174 19. For the "mail" command in an Exim filter, no default was being set for
5175 the once_repeat time, causing a random time value to be used if "once" was
5176 specified. (If the value happened to be <= 0, no repeat happened.) The
5177 default is now 0s, meaning "never repeat". The "vacation" command was OK
5178 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5179 (I found it when inspecting the code).
5181 20. There is now an overall timeout for performing a callout verification. It
5182 defaults to 4 times the callout timeout, which applies to individual SMTP
5183 commands during the callout. The overall timeout applies when there is more
5184 than one host that can be tried. The timeout is checked before trying the
5185 next host. This prevents very long delays if there are a large number of
5186 hosts and all are timing out (e.g. when the network connections are timing
5187 out). The value of the overall timeout can be changed by specifying an
5188 additional sub-option for "callout", called "maxwait". For example:
5190 verify = sender/callout=5s,maxwait=20s
5192 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5193 to the end before writing, but this should make it even safer).
5195 22. Exim was forgetting that it had advertised PIPELINING for the second and
5196 subsequent messages on an SMTP connection. It was also not resetting its
5197 memory on STARTTLS and an internal HELO.
5199 23. When Exim logs an SMTP synchronization error within a session, it now
5200 records whether PIPELINING has been advertised or not.
5202 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5203 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5204 rather than long int.
5206 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5208 26. Added the never_mail option to autoreply.
5214 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5215 crash if the getsockname() call failed; this can happen if a connection is
5216 closed very soon after it is established. The problem was simply in the
5217 order in which certain operations were done, causing Exim to try to write
5218 to the SMTP stream before it had set up the file descriptor. The bug has
5219 been fixed by making things happen in the correct order.
5225 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5226 before the connection was closed, thus losing the rejection response.
5228 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5229 some early Solaris releases, but causes trouble in current releases where
5230 socklen_t is defined.
5232 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5235 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5238 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5239 that releasing the top of it at the end releases what was used for sub-
5240 expansions (unless the block got too big). However, discard this block if
5241 the first thing is a variable or header, so that we can use its block when
5242 it is dynamic (useful for very large $message_headers, for example).
5244 6. Lookups now cache *every* query, not just the most recent. A new, separate
5245 store pool is used for this. It can be recovered when all lookup caches are
5246 flushed. Lookups now release memory at the end of their result strings.
5247 This has involved some general refactoring of the lookup sources.
5249 7. Some code has been added to the store_xxx() functions to reduce the amount
5250 of flapping under certain conditions.
5252 8. log_incoming_interface used to affect only the <= reception log lines. Now
5253 it causes the local interface and port to be added to several more SMTP log
5254 lines, for example "SMTP connection from", and rejection lines.
5256 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5258 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5260 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5261 could be overwritten at the end of the current message (or the start of a
5262 new message if it was set in a HELO ACL). The value is now preserved for
5263 the duration of the SMTP connection.
5265 12. If a transport had a headers_rewrite setting, and a matching header line
5266 contained an unqualified address, that address was qualified, even if it
5267 did not match any rewriting rules. The underlying bug was that the values
5268 of the flags that permit the existence of unqualified sender and recipient
5269 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5270 non-local messages, and by -bnq for local messages) were not being
5271 preserved with the message after it was received.
5273 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5274 "next input=" as part of the text comprising the host identity instead of
5275 the correct text. The code was using the same buffer for two different
5276 strings. However, depending on which order the printing function evaluated
5277 its arguments, the bug did not always show up. Under Linux, for example, my
5278 test suite worked just fine.
5280 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5281 doesn't work with some older versions of Perl. It has been changed to "my",
5282 which in any case is probably the better facility to use.
5284 15. A really picky compiler found some instances of statements for creating
5285 error messages that either had too many or two few arguments for the format
5288 16. The size of the buffer for calls to the DNS resolver has been increased
5289 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5290 for addresses that have a lot of PTR records. This alleviates a problem; it
5291 does not fully solve it.
5293 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5294 buffer now truncates the list and logs the incident, which is the same
5295 action as happens when Exim is looking up a host name and its aliases.
5296 Previously in this situation something unpredictable would happen;
5297 sometimes it was "internal error: store_reset failed".
5299 18. If a server dropped the connection unexpectedly when an Exim client was
5300 using GnuTLS and trying to read a response, the client delivery process
5301 crashed while trying to generate an error log message.
5303 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5304 string, for example:
5306 warn message = H1: something\nH2: something
5308 the text was added as a single header line from Exim's point of view
5309 though it ended up OK in the delivered message. However, searching for the
5310 second and subsequent header lines using $h_h2: did not work. This has been
5311 fixed. Similarly, if a system filter added multiple headers in this way,
5312 the routers could not see them.
5314 20. Expanded the error message when iplsearch is called with an invalid key to
5315 suggest using net-iplsearch in a host list.
5317 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5318 ACLs are no longer actually imposed (and a message to that effect is
5321 22. If a "gecos" field in a passwd entry contained escaped characters, in
5322 particular, if it contained a \" sequence, Exim got it wrong when building
5323 a From: or a Sender: header from that name. A second bug also caused
5324 incorrect handling when an unquoted " was present following a character
5325 that needed quoting.
5327 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5328 was not being matched caselessly.
5330 24. Arranged for all hyphens in the exim.8 source to be escaped with
5333 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5334 properly. Recipient callout cache records were still being keyed to include
5335 the sender, even when use_sender was set false. This led to far more
5336 callouts that were necessary. The sender is no longer included in the key
5337 when use_sender is false.
5339 26. Added "control = submission" modifier to ACLs.
5341 27. Added the ${base62d: operator to decode base 62 numbers.
5343 28. dnsdb lookups can now access SRV records.
5345 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5346 the configuration file.
5348 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5349 (-v) mode. This makes the output for a verbose queue run more intelligible.
5351 31. Added a use_postmaster feature to recipient callouts.
5353 32. Added the $body_zerocount variable, containing the number of binary zero
5354 bytes in the message body.
5356 33. The time of last modification of the "new" subdirectory is now used as the
5357 "mailbox time last read" when there is a quota error for a maildir
5360 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5362 35. Added +ignore_unknown as a special item in host lists.
5364 36. Code for decoding IPv6 addresses in host lists is now included, even if
5365 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5366 address was recognized as an IP address, but was then not correctly decoded
5367 into binary, causing unexpected and incorrect effects when compared with
5374 1. Very minor rewording of debugging text in manualroute to say "list of
5375 hosts" instead of "hostlist".
5377 2. If verify=header_syntax was set, and a header line with an unqualified
5378 address (no domain) and a large number of spaces between the end of the
5379 name and the colon was received, the reception process suffered a buffer
5380 overflow, and (when I tested it) crashed. This was caused by some obsolete
5381 code that should have been removed. The fix is to remove it!
5383 3. When running in the test harness, delay a bit after writing a bounce
5384 message to get a bit more predictability in the log output.
5386 4. Added a call to search_tidyup() just before forking a reception process. In
5387 theory, someone could use a lookup in the expansion of smtp_accept_max_
5388 per_host which, without the tidyup, could leave open a database connection.
5390 5. Added the variables $recipient_data and $sender_data which get set from a
5391 lookup success in an ACL "recipients" or "senders" condition, or a router
5392 "senders" option, similar to $domain_data and $local_part_data.
5394 6. Moved the writing of debug_print from before to after the "senders" test
5397 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5398 problems for message scanning, either using a data ACL, or using
5399 local_scan() because the Received: header was not generated till after they
5400 were called (in order to set the time as the time of reception completion).
5401 I have revised the way this works. The header is now generated after the
5402 body is received, but before the ACL or local_scan() are called. After they
5403 are run, the timestamp in the header is updated.
5409 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5410 before starting a queue runner without re-exec. This happened only when
5411 deliver_drop_privilege was set or when the Exim user was set to root. The
5412 effect of the bug was that timeouts during subsequent deliveries caused
5413 crashes instead of being properly handled. The handler is now left at its
5414 default (and expected) setting.
5416 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5417 message, again when deliver_drop_privilege is set or Exim is run as root.
5418 The bug described in (1) was not present in this case, but the tidying up
5419 of the other signals was missing. I have made the two cases consistent.
5421 3. The ignore_target_hosts setting on a manualroute router was being ignored
5422 for hosts that were looked up using the /MX notation.
5424 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5427 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5428 operated on the sender address. After changing the $sender_address to <>
5429 for the sender address verify, Exim was re-instated it as the original
5430 (before rewriting) address, but remembering that it had rewritten it, so it
5431 wasn't rewriting it again. This bug also had the effect of breaking the
5432 sender address verification caching when the sender address was rewritten.
5434 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5435 This has been changed so that if the ip literal address matches
5436 ignore_target_hosts, the router declines.
5438 7. Added expansion conditions match_domain, match_address, and match_local_
5439 part (NOT match_host).
5441 8. The placeholder for the Received: header didn't have a length field set.
5443 9. Added code to Exim itself and to exim_lock to test for a specific race
5444 condition that could lead to file corruption when using MBX delivery. The
5445 issue is with the lockfile that is created in /tmp. If this file is removed
5446 after a process has opened it but before that process has acquired a lock,
5447 there is the potential for a second process to recreate the file and also
5448 acquire a lock. This could lead to two Exim processes writing to the file
5449 at the same time. The added code performs the same test as UW imapd; it
5450 checks after acquiring the lock that its file descriptor still refers to
5451 the same named file.
5453 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5454 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5457 11. Added the smtp_active_hostname option. If used, this will typically be made
5458 to depend on the incoming interface address. Because $interface_address is
5459 not set up until the daemon has forked a reception process, error responses
5460 that can happen earlier (such as "too many connections") no longer contain
5463 12. If an expansion in a condition on a "warn" statement fails because a lookup
5464 defers, the "warn" statement is abandoned, and the next ACL statement is
5465 processed. Previously this caused the whole ACL to be aborted.
5467 13. Added the iplsearch lookup type.
5469 14. Added ident_timeout as a log selector.
5471 15. Added tls_certificate_verified as a log selector.
5473 16. Added a global option tls_require_ciphers (compare the smtp transport
5474 option of the same name). This controls incoming TLS connections.
5476 17. I finally figured out how to make tls_require_ciphers do a similar thing
5477 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5478 before starting the TLS session.
5480 18. Tabs are now shown as \t in -bP output.
5482 19. If the log selector return_path_on_delivery was set, Exim crashed when
5483 bouncing a message because it had too many Received: header lines.
5485 20. If two routers both had headers_remove settings, and the first one included
5486 a superfluous trailing colon, the final name in the first list and the
5487 first name in the second list were incorrectly joined into one item (with a
5488 colon in the middle).
5494 1. Added -C and -D options to the exinext utility, mainly to make it easier
5495 to include in the automated testing, but these could be helpful when
5496 multiple configurations are in use.
5498 2. The exinext utility was not formatting the output nicely when there was
5499 an alternate port involved in the retry record key, nor when there was a
5500 message id as well (for retries that were specific to a specific message
5501 and a specific host). It was also confused by IPv6 addresses, because of
5502 the additional colons they contain. I have fixed the IPv4 problem, and
5503 patched it up to do a reasonable job for IPv6.
5505 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5506 delivery, the log line now contains "pipelined" if PIPELINING was used.
5508 4. An SMTP transport process used to panic and die if the bind() call to set
5509 an explicit outgoing interface failed. This has been changed; it is now
5510 treated in the same way as a connect() failure.
5512 5. A reference to $sender_host_name in the part of a conditional expansion
5513 that was being skipped was still causing a DNS lookup. This no longer
5516 6. The def: expansion condition was not recognizing references to header lines
5517 that used bh_ and bheader_.
5519 7. Added the _cache feature to named lists.
5521 8. The code for checking quota_filecount in the appendfile transport was
5522 allowing one more file than it should have been.
5524 9. For compatibility with Sendmail, the command line option
5532 and sets the incoming protocol and host name (for trusted callers). The
5533 host name and its colon can be omitted when only the protocol is to be set.
5534 Note the Exim already has two private options, -pd and -ps, that refer to
5535 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5536 "s", but I don't think that's a major issue.
5538 10. A number of refactoring changes to the code, none of which should affect
5541 (a) The number of logging options was getting close to filling up the
5542 32-bit word that was used as a bit map. I have split them into two classes:
5543 those that are passed in the argument to log_write(), and those that are
5544 only ever tested independently outside of that function. These are now in
5545 separate 32-bit words, so there is plenty of room for expansion again.
5546 There is no change in the user interface or the logging behaviour.
5548 (b) When building, for example, log lines, the code previously used a
5549 macro that called string_cat() twice, in order to add two strings. This is
5550 not really sufficiently general. Furthermore, there was one instance where
5551 it was actually wrong because one of the argument was used twice, and in
5552 one call a function was used. (As it happened, calling the function twice
5553 did not affect the overall behaviour.) The macro has been replaced by a
5554 function that can join an arbitrary number of extra strings onto a growing
5557 (c) The code for expansion conditions now uses a table and a binary chop
5558 instead of a serial search (which was left over from when there were very
5559 few conditions). Also, it now recognizes conditions like "pam" even when
5560 the relevant support is not compiled in: a suitably worded error message is
5561 given if an attempt is made to use such a condition.
5563 11. Added ${time_interval:xxxxx}.
5565 12. A bug was causing one of the ddress fields not to be passed back correctly
5566 from remote delivery subprocesses. The field in question was not being
5567 subsequently used, so this caused to problems in practice.
5569 13. Added new log selectors queue_time and deliver_time.
5571 14. Might have fixed a bug in maildirsizefile handling that threw up
5572 "unexpected character" debug warnings, and recalculated the data
5573 unnecessarily. In any case, I expanded the warning message to give more
5576 15. Added the message "Restricted characters in address" to the statements in
5577 the default ACL that block characters like @ and % in local parts.
5579 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5580 Three changes have been made:
5582 (a) There was a serious bug; a negative response to MAIL caused the whole
5583 recipient domain to be cached as invalid, thereby blocking all messages
5584 to all local parts at the same domain, from all senders. This bug has
5585 been fixed. The domain is no longer cached after a negative response to
5586 MAIL if the sender used is not empty.
5588 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5591 (c) A new callout option, "use_sender" has been added for people who want
5592 the modified behaviour.
5598 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5601 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5602 indeed breaks things for older releases.
5604 3. Added additional logging to the case where there is a problem reading data
5605 from a filter that is running in a subprocess using a pipe, in order to
5606 try to track down a specific problem.
5608 4. Testing facility fudge: when running in the test harness and attempting
5609 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5610 getting "No route to host". Convert this to a timeout.
5612 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5615 6. Some OS don't have socklen_t but use size_t instead. This affects the
5616 fifth argument of getsockopt() amongst other things. This is now
5617 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5618 can be set for individual OS. I have set it for SunOS5, OSF1, and
5619 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5620 some earlier ones do not.
5622 7. Change 4.30/15 was not doing the test caselessly.
5624 8. The standard form for an IPv6 address literal was being rejected by address
5625 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5626 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5627 this, as well as the form without the "IPv6" on the front (but only when
5628 address literals are enabled, of course).
5630 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5632 10. Exim crashed if a message with an empty sender address specified by -f
5633 encountered a router with an errors_to setting. This could be provoked only
5634 by a command such as
5638 where an empty string was supplied; "<>" did not hit this bug.
5640 11. Installed PCRE release 4.5.
5642 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5643 remained set. It is now erased.
5645 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5646 times from message ids (which are base 36 rather than the normal 62).
5648 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5649 were being counted as actual protocol errors, and logged if the log
5650 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5651 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5652 and DATA following a set of rejected RCPTs do not count as protocol errors.
5653 In other words, Exim assumes they were pipelined, though this may not
5654 actually be the case. Of course, in all cases the client gets an
5655 appropriate error code.
5657 15. If a lookup fails in an ACL condition, a message about the failure may
5658 be available; it is used if testing the ACL cannot continue, because most
5659 such messages specify what the cause of the deferral is. However, some
5660 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5661 that caused an old message to be retained and used if a later statement
5662 caused a defer, replacing the real cause of the deferral.
5664 16. If an IP address had so many PTR records that the DNS lookup buffer
5665 was not large enough to hold them, Exim could crash while trying to process
5666 the truncated data. It now detects and logs this case.
5668 17. Further to 4.21/58, another change has been made: if (and only if) the
5669 first line of a message (the first header line) ends with CRLF, a bare LF
5670 in a subsequent header line has a space inserted after it, so as not to
5671 terminate the header.
5673 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5674 unnecessarily, used atoi() instead of strtol(), and didn't check the
5675 termination when getting file sizes from file names by regex.
5677 19. Completely re-implemented the support for maildirsize files, in the light
5678 of a number of problems with the previous contributed implementation
5679 (4.30/29). In particular:
5681 . If the quota is zero, the maildirsize file is maintained, but no quota is
5684 . If the maildir directory does not exist, it is created before any attempt
5685 to write a maildirsize file.
5687 . The quota value in the file is just a cache; if the quota is changed in
5688 the transport, the new value overrides.
5690 . A regular expression is available for excluding directories from the
5693 20. The autoreply transport checks the characters in options that define the
5694 message's headers; it allows continued headers, but it was checking with
5695 isspace() after an embedded newline instead of explicitly looking for a
5698 21. If all the "regular" hosts to which an address was routed had passed their
5699 expiry times, and had not reached their retry times, the address was
5700 bounced, even if fallback hosts were defined. Now Exim should go on to try
5703 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5704 equivalent code in the SMTP transport. Some hosts send humungous responses
5705 to HELO/EHLO, more than 1024 it seems.
5707 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5708 gives compiler warnings in some environments. I've now done it "properly",
5711 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5712 (because of problems with the built-in one) was declared to return uschar *
5713 instead of char *, causing compiler failure.
5715 25. Fixed a file descriptor leak when processing alias/forward files.
5717 26. Fixed a minor format string issue in dbfn.c.
5719 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5721 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5722 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5725 29. When a sender address is verified, it is cached, to save repeating the test
5726 when there is more than one recipient in a message. However, when the
5727 verification involves a callout, it is possible for different callout
5728 options to be set for different recipients. It is too complicated to keep
5729 track of this in the cache, so now Exim always runs a verification when a
5730 callout is required, relying on the callout cache for the optimization.
5731 The overhead is duplication of the address routing, but this should not be
5734 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5735 to be verified with callout=postmaster, and the main callout worked but the
5736 postmaster check failed, the verification correctly failed. However, if a
5737 subsequent RCPT command asked for sender verification *without* the
5738 postmaster check, incorrect caching caused this verification also to fail,
5741 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5742 it was not caching the DNS options (qualify_single, search_parents) that
5743 were used when the lookup failed. A subsequent lookup with different
5744 options therefore always gave the same answer, though there were cases
5745 where it should not have. (Example: a "domains = !$mx_any" option on a
5746 dnslookup router: the "domains" option is always processed without any
5747 widening, but the router might have qualify_single set.) Now Exim uses the
5748 cached value only when the same options are set.
5750 32. Added John Jetmore's "exipick" utility to the distribution.
5752 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5753 than a timeout (e.g. a certificate is required, and is not provided), an
5754 Exim server now closes the connection immediately. Previously it waited for
5755 the client to close - but if the client is SSL, it seems that they each
5756 wait for each other, leading to a delay before one of them times out.
5758 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5759 maintained 0.8.x compatibility because I don't think many are using it, and
5760 it is clearly obsolete.
5762 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5765 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5766 client certificate was expired. A simple patch fixes this, though I don't
5767 understand the full logic of why the verify callback is called multiple
5770 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5771 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5772 which causes problems with some clients (such as the Certicom SSL Plus
5773 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5774 disables the coutermeasure allowing Eudora to connect."
5776 38. Exim was not checking that a write() to a log file succeeded. This could
5777 lead to Bad Things if a log got too big, in particular if it hit a file
5778 size limit. Exim now panics and dies if it cannot write to a log file, just
5779 as it does if it cannot open a log file.
5781 39. Modified OS/Makefile-Linux so that it now contains
5783 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5785 The two -D definitions ensure that Exim is compiled with large file
5786 support, which makes it possible to handle log files that are bigger than
5789 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5790 instance) a domain was checked against a named list that involved a lookup,
5791 causing $domain_data to be set, then another domain was checked against the
5792 same list, then the first domain was re-checked, the value of $domain_data
5793 after the final check could be wrong. In particular, if the second check
5794 failed, it could be set empty. This bug probably also applied to
5797 41. The strip_trailing_dot option was not being applied to the address given
5798 with the -f command-line option.
5800 42. The code for reading a message's header from the spool was incrementing
5801 $received_count, but never initializing it. This meant that the value was
5802 incorrect (doubled) while delivering a message in the same process in which
5803 it was received. In the most common configuration of Exim, this never
5804 happens - a fresh exec is done - but it can happen when
5805 deliver_drop_privilege is set.
5807 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5808 it now includes up to 150 characters of the unexpected data in the log
5811 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5812 and building data strings. The size of both of these buffers was 10 000
5813 bytes - far larger than anybody would *ever* want, thought I. Needless to
5814 say, somebody hit the limit. I have increased the maximum line length to
5815 20 000 and the maximum data length of concatenated lines to 100 000. I have
5816 also fixed two bugs, because there was no checking on these buffers. Tsk,
5817 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5818 buffer is too small.
5820 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5821 lsearch lookups. Now it does.
5823 46. When parsing a route_list item in a manualroute router, a fixed-length
5824 buffer was used for the list of hosts. I made this 1024 bytes long,
5825 thinking that nobody would ever have a list of hosts that long. Wrong.
5826 Somebody had a whole pile of complicated expansion conditions, and the
5827 string was silently truncated, leading to an expansion error. It turns out
5828 that it is easier to change to an unlimited length (owing to other changes
5829 that have happened since this code was originally written) than to build
5830 structure for giving a limitation error. The length of the item that
5831 expands into the list of hosts is now unlimited.
5833 47. The lsearch lookup could not handle data where the length of text line was
5834 more than 4095 characters. Such lines were truncated, leading to shortened
5835 data being returned. It should now handle lines of any length.
5837 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5838 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5841 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5842 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5843 no longer happens in comment lines. A list of replacements is now placed
5844 at the head of all of the source files, except those whose only change is
5845 to replace PERL_COMMAND in the very first #! line.
5847 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5848 messages on the queue, with a bottom-up merge sort, using code contributed
5849 by Michael Haardt. This should make operations like -bp somewhat faster on
5850 large queues. It won't affect queue runners, except when queue_run_in_order
5853 51. Installed eximstats 1.31 in the distribution.
5855 52. Added support for SRV lookups to the dnslookup router.
5857 53. If an ACL referred to $message_body or $message_body_end, the value was not
5858 reset for any messages that followed in the same SMTP session.
5860 54. The store-handling optimization for building very long strings was not
5861 differentiating between the different store pools. I don't think this
5862 actually made any difference in practice, but I've tidied it.
5864 55. While running the routers to verify a sender address, $sender_address
5865 was still set to the sender address. This is wrong, because when routing to
5866 send a bounce to the sender, it would be empty. Therefore, I have changed
5867 it so that, while verifying a sender address, $sender_address is set to <>.
5868 (There is no change to what happens when verifying a recipient address.)
5870 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5871 target A or AAAA records (if not already returned) without resetting the
5872 qualify_single or search_parents options of the DNS resolver. These are
5873 inappropriate in this case because the targets of MX and SRV records must
5874 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5875 target that, when qualified, matched something in the local domain. These
5876 two options are now turned off when doing these lookups.
5878 57. It seems that at least some releases of Reiserfs (which does not have the
5879 concept of a fixed number of inodes) returns zero and not -1 for the
5880 number of available inodes. This interacted badly with check_spool_inodes,
5881 which assumed that -1 was the "no such thing" setting. What I have done is
5882 to check that the total number of inodes is greater than zero before doing
5883 the test of how many are available.
5885 58. When a "warn" ACL statement has a log_message modifier, the message is
5886 remembered, and not repeated. This is to avoid a lot of repetition when a
5887 message has many recipients that cause the same warning to be written.
5888 However, Exim was preserving the list of already written lines for an
5889 entire SMTP session, which doesn't seem right. The memory is now reset if a
5890 new message is started.
5892 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5893 debugging output unless log_rewrite was also set.
5895 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5896 of (int)(handle) when we know that handle contains (void *)(-1).
5898 61. The Exim daemon panic-logs an error return when it closes the incoming
5899 connection. However "connection reset by peer" seems to be common, and
5900 isn't really an error worthy of noting specially, so that particular error
5903 62. When Exim is trying to find all the local interfaces, it used to panic and
5904 die if the ioctl to get the interface flags failed. However, it seems that
5905 on at least one OS (Solaris 9) it is possible to have an interface that is
5906 included in the list of interfaces, but for which you get a failure error
5907 for this call. This happens when the interface is not "plumbed" into a
5908 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5909 failure of the "get flags" call assumes that the interface is down.
5911 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5912 makes life easier for people who are doing arithmetic on fields extracted
5913 from dates, where you often get leading zeros that should not be
5914 interpreted as octal.
5916 64. Added qualify_domain to the redirect router, to override the global
5919 65. If a pathologically long header line contained very many addresses (the
5920 report of this problem mentioned 10 000) and each of them was rewritten,
5921 Exim could use up a very large amount of memory. (It kept on making new
5922 copies of the header line as it rewrote, and never released the old ones.)
5923 At the expense of a bit more processing, the header rewriting function has
5924 been changed so that it no longer eats memory in this way.
5926 66. The generation of the Received: header has been moved from the time that a
5927 message starts to be received, to the time that it finishes. The timestamp
5928 in the Received: header should now be very close to that of the <= log
5929 line. There are two side-effects of this change:
5931 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5932 logged header lines no longer include the local Received: line, because
5933 it has not yet been created. The same applies to a copy of the message
5934 that is returned to a non-SMTP sender when a message is rejected.
5936 (b) When a filter file is tested using -bf, no additional Received: header
5937 is added to the test message. After some thought, I decided that this
5940 This change does not affect the value of $received_for. It is still set
5941 after address rewriting, but before local_scan() is called.
5943 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5945 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5946 gave an unhelpful panic error message, and a defer error. I have managed to
5947 change this behaviour so that it now rejects any supplied certificate,
5948 which seems right, as the list of acceptable certificates is empty.
5950 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5951 gave an unhelpful defer error. I have not managed to make this reject any
5952 supplied certificates, but the error message it gives is "no certificate
5953 supplied", which is not helpful.
5955 70. exigrep's output now also includes lines that are not associated with any
5956 message, but which match the given pattern. Implemented by a patch from
5957 Martin Sluka, which also tidied up the Perl a bit.
5959 71. Recipient callout verification, like sender verification, was using <> in
5960 the MAIL FROM command. This isn't really the right thing, since the actual
5961 sender may affect whether the remote host accepts the recipient or not. I
5962 have changed it to use the actual sender in the callout; this means that
5963 the cache record is now keyed on a recipient/sender pair, not just the
5964 recipient address. There doesn't seem to be a real danger of callout loops,
5965 since a callout by the remote host to check the sender would use <>.
5966 [SEE ABOVE: changed after hitting problems.]
5968 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5969 temporary errors. However, in the case of such a code being given after
5970 the end of a data transmission (i.e. after ".") Exim was failing to write
5971 a retry record for the message. (Yes, there was some broken host that was
5972 actually sending 8xx at this point.)
5974 73. An unknown lookup type in a host list could cause Exim to panic-die when
5975 the list was checked. (An example that provoked this was putting <; in the
5976 middle of a list instead of at the start.) If this happened during a DATA
5977 ACL check, a -D file could be left lying around. This kind of configuration
5978 error no longer causes Exim to die; instead it causes a defer error. The
5979 incident is still logged to the main and panic logs.
5981 74. Buglet left over from Exim 3 conversion. The message "too many messages
5982 in one connection" was written to the rejectlog but not the mainlog, except
5983 when address rewriting (yes!) was being logged.
5985 75. Added write_rejectlog option.
5987 76. When a system filter was run not as root (that is, when system_filter_user
5988 was set), the values of the $n variables were not being returned to the
5989 main process; thus, they were not subsequently available in the $sn
5992 77. Added +return_path_on_delivery log selector.
5994 78. A connection timeout was being treated differently from recipients deferred
5995 when testing hosts_max_try with a message that was older than the host's
5996 retry timeout. (The host should not be counted, thus allowing all hosts to
5997 be tried at least once before bouncing.) This may have been the cause of an
5998 occasionally reported bug whereby a message would remain on the queue
5999 longer than the retry timeout, but would be bounced if a delivery was
6000 forced. I say "may" because I never totally pinned down the problem;
6001 setting up timeout/retry tests is difficult. See also the next item.
6003 79. The ultimate address timeout was not being applied to errors that involved
6004 a combination of host plus message (for example, a timeout on a MAIL
6005 command). When an address resolved to a number of possible hosts, and they
6006 were not all tried for each delivery (e.g. because of hosts_max_try), a
6007 message could remain on the queue longer than the retry timeout.
6009 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6012 81. Fixed an obscure SMTP outgoing bug which required at least the following
6013 conditions: (a) there was another message waiting for the same server;
6014 (b) the server returned 5xx to all RCPT commands in the first message so
6015 that the message was not completed; (c) the server dropped the connection
6016 or gave a negative response to the RSET that Exim sends to abort the
6017 transaction. The observed case was a dropped connection after DATA that had
6018 been sent in pipelining mode. That is, the server had advertised PIPELINING
6019 but was not implementing it correctly. The effect of the bug was incorrect
6020 behaviour, such as trying another host, and this could lead to a crash.
6026 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6027 and daemon.c were passed as pointers to ints; they should have been
6028 pointers to socklen_t variables (which are typically unsigned ints).
6030 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6033 3. Fixed a really odd bug that affected only the testing scheme; patching a
6034 certain fixed string in the binary changed the value of another string that
6035 happened to be identical to the end of the original first string.
6037 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6038 name", it returns that address as the IP address. On some operating
6039 systems (e.g. Solaris), it also passes back the IP address string as the
6040 "host name". However, on others (e.g. Linux), it passes back an empty
6041 string. Exim wasn't checking for this, and was changing the host name to an
6042 empty string, assuming it had been canonicalized.
6044 5. Although rare, it is permitted to have more than one PTR record for a given
6045 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6046 all the names associated with an address, because they do in Solaris.
6047 However, it seems that they do not in Linux for data that comes from the
6048 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6049 I found this out when I moved to a new Linux workstation and tried to run
6050 the Exim test suite.
6052 To get round this problem I have changed the code so that it now does its
6053 own call to the DNS to look up PTR records when searching for a host name.
6054 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6055 addresses that are only in /etc/hosts are still found.
6057 This behaviour is, however, controlled by an option called host_lookup_
6058 order, which defaults to "bydns:byaddr". If people want to use the other
6059 order, or indeed, just use one or the other means of lookup, they can
6060 specify it in this variable.
6062 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6063 some operating systems, this comes back from gethostbyaddr() as an empty
6064 string, and this is what Exim used to test for. However, it seems that in
6065 other systems, "." is yielded. Exim now tests for this case too.
6067 7. The values of check_spool_space and check_log_space are now held internally
6068 as a number of kilobytes instead of an absolute number of bytes. If a
6069 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6070 kilobyte. This means that much larger values can be stored.
6072 8. Exim monitor: an attempt to get the action menu when not actually pointing
6073 at a message produces an empty menu entitled "No message selected". This
6074 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6075 no entries in it ("Shell widget menu has zero width and/or height"). So I
6076 have added a single, blank menu entry in this case.
6078 9. Added ${quote_local_part.
6080 10. MIME decoding is now applied to the contents of Subject: header lines when
6083 11. Now that a reference to $sender_host_address automatically causes a reverse
6084 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6085 host lookup before query-style lookups in lists that might use this
6086 variable. This has therefore been abolished, and the "net-" prefix is no
6087 longer necessary for query-style lookups.
6089 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6090 have been a typo for LFLAGS, so it has been changed.
6092 13. The install script calls Exim with "-C /dev/null" in order to find the
6093 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6094 to be output. However, since Exim outputs its version number before the
6095 error, it didn't break the script. It just looked ugly. I fixed this by
6096 always allowing "-C /dev/null" if the caller is root.
6098 14. Ignore overlarge ACL variable number when reading spool file - insurance
6099 against a later release with more variables having written the file.
6101 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6102 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6103 this, as well as the form without the "IPv6" on the front.
6105 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6106 OS/Makefile-Darwin file.
6108 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6109 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6111 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6112 variable. However, some OS use ints for inodes. I've added cast to long int
6113 to get rid of the compiler warning.
6115 19. I had forgotten to lock out "/../" in configuration file names when
6116 ALT_CONFIG_PREFIX was set.
6118 20. Routers used for verification do not need to specify transports. However,
6119 if such a router generated a host list, and callout was configured, Exim
6120 crashed, because it could not find a port number from the (non-existent)
6121 transport. It now assumes port 25 in this circumstance.
6123 21. Added the -t option to exigrep.
6125 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6126 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6127 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6128 them is set, LOOKUP_LSEARCH is forced.
6130 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6132 24. Added sender and host information to the "rejected by local_scan()" log
6133 line; previously there was no indication of these.
6135 25. Added .include_if_exists.
6137 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6138 Linux. It turns out that not all file systems support this. Apparently some
6139 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6140 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6141 not supported on the file descriptor, is now ignored when Exim is trying to
6142 sync a directory. This applies only to Linux.
6144 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6146 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6147 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6150 29. Added support for maildirsize files from supplied patch (modified a bit).
6152 30. The use of :fail: followed by an empty string could lead Exim to respond to
6153 sender verification failures with (e.g.):
6155 550 Verification failed for <xxx>
6156 550 Sender verify failed
6158 where the first response line was missing the '-' that indicates it is not
6159 the final line of the response.
6161 31. The loop for finding the name of the user that called Exim had a hardwired
6162 limit of 10; it now uses the value of finduser_retries, which is used for
6163 all other user lookups.
6165 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6168 33. Exim was neglecting to zero errno before one call of strtol() when
6169 expanding a string and expecting an integer value. On some systems this
6170 resulted in spurious "integer overflow" errors. Also, it was casting the
6171 result into an int without checking.
6173 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6174 did not work. The code looks as if it has *never* worked, though it appears
6175 to have been documented since at least release 1.62. I have made it work.
6177 35. The "timeout_DNS" error in retry rules, also documented since at least
6178 1.62, also never worked. As it isn't clear exactly what this means, and
6179 clearly it isn't a major issue, I have abolished the feature by treating it
6180 as "timeout", and writing a warning to the main and panic logs.
6182 36. The display of retry rules for -brt wasn't always showing the error code
6185 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6186 timeout_connect_A, timeout_connect_MX.
6188 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6189 to the empty sender.
6191 39. The daemon was not analysing the content of -oX till after it had closed
6192 stderr and disconnected from the controlling terminal. This meant that any
6193 syntax errors were only noted on the panic log, and the return code from
6194 the command was 0. By re-arranging the code a little, I've made the
6195 decoding happen first, so such errors now appear on stderr, and the return
6196 code is 1. However, the actual setting up of the sockets still happens in
6197 the disconnected process, so errors there are still only recorded on the
6200 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6201 connections (as happens on some IP stacks) was logged at start up time as
6202 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6203 it from "IPv6 and IPv4", which means that two separate sockets are being
6206 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6207 says whether AF_INET or AF_INET6 was passed as an argument.
6209 42. Exiwhat output was messed up when time zones were included in log
6212 43. Exiwhat now gives more information about the daemon's listening ports,
6213 and whether -tls-on-connect was used.
6215 44. The "port" option of the smtp transport is now expanded.
6217 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6218 silently ignored. Now an error message is written to the main and panic
6221 46. There's a new ACL modifier called "logwrite" which writes to a log file
6222 as soon as it is encountered.
6224 47. Added $local_user_uid and $local_user_gid at routing time.
6226 48. Exim crashed when trying to verify a sender address that was being
6229 49. Exim was recognizing only a space character after ".include". It now also
6230 recognizes a tab character.
6232 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6233 extracting the relevant information from the specification. The man page no
6234 longer contains scrambled data for the -d option, and I've added a section
6235 at the front about calling Exim under different names.
6237 51. Added "extra_headers" argument to the "mail" command in filter files.
6239 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6242 53. Installed eximstats 1.29.
6244 54. Added transport_filter_timeout as a generic transport option.
6246 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6247 Cc: header lines. This was required by RFC 822, but it not required by RFC
6250 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6251 incoming messages that did not have them. Now it does so only if the
6252 message originates locally, that is, if there is no associated remote host
6253 address. When Resent- header lines are present, this applies to the Resent-
6254 lines rather than the non-Resent- lines.
6256 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6257 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6259 58. Messages for configuration errors now include the name of the main
6260 configuration file - useful now that there may be more than one file in a
6261 list (.included file names were always shown).
6263 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6264 for those rare installations that do not start the daemon as root or run it
6265 setuid root. I've cut out the call to initgroups() if the daemon is not
6268 60. The Exim user and group can now be bound into the binary as text strings
6269 that are looked up at the start of Exim's processing.
6271 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6273 62. Added $mailstore_basename variable.
6275 63. Installed patch to sieve.c from Michael Haardt.
6277 64. When Exim failed to open the panic log after failing to open the main log,
6278 the original message it was trying to log was written to stderr and debug
6279 output, but if they were not available (the usual case in production), it
6280 was lost. Now it is written to syslog before the two lines that record the
6281 failures to open the logs.
6283 65. Users' Exim filters run in subprocesses under the user's uid. It is
6284 possible for a "deliver" command or an alias in a "personal" command to
6285 provoke an address rewrite. If logging of address rewriting is configured,
6286 this fails because the process is not running as root or exim. There may be
6287 a better way of dealing with this, but for the moment (because 4.30 needs
6288 to be released), I have disabled address rewrite logging when running a
6289 filter in a non-root, non-exim process.
6295 1. The buildconfig auxiliary program wasn't quoting the value set for
6296 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6297 not defined. This bug was masked in 4.22 by the effect that was fixed in
6300 2. Some messages that were rejected after a message id was allocated were
6301 shown as "incomplete" by exigrep. It no longer does this for messages that
6302 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6304 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6305 have allow_domain_literals set, the ID did not get logged in the <= line.
6306 Domain literals are now always recognized in Message-ID: header lines.
6308 4. The first argument for a ${extract expansion item is the key name or field
6309 number. Leading and trailing spaces in this item were not being ignored,
6310 causing some misleading effects.
6312 5. When deliver_drop_privilege was set, single queue runner processes started
6313 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6314 same command in the process it spins off) were not dropping privilege.
6316 6. When the daemon running as "exim" started a queue runner, it always
6317 re-executed Exim in the spun-off process. This is a waste of effort when
6318 deliver_drop_privilege is set. The new process now just calls the
6319 queue-runner function directly.
6325 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6328 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6329 ignored. Though the use of -oP was forcing the writing of a pid file, it
6330 was always written to the default place.
6332 3. If the message "no IP address found for host xxxx" is generated during
6333 incoming verification, it is now followed by identification of the incoming
6334 connection (so you can more easily find what provoked it).
6336 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6338 5. Added some features to "harden" Exim a bit more against certain attacks:
6340 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6341 be put in Local/Makefile. This is like the never_users runtime option,
6342 but it cannot be overridden. The default setting is "root".
6344 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6345 prefix string with which any file named in a -C command line option
6348 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6349 is retained for -C and -D only if the caller of Exim is root. Without
6350 it, the exim user may also use -C and -D and retain privilege.
6352 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6353 command line option is disabled.
6355 6. Macro names set by the -D option must start with an upper case letter, just
6356 like macro names defined in the configuration file.
6358 7. Added "dereference=" facility to LDAP.
6360 8. Two instances of the typo "uknown" in the source files are fixed.
6362 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6363 the Configure-Makefile script screwed up while processing it.
6365 10. Incorporated PCRE 4.4.
6367 11. The SMTP synchronization check was not operating right at the start of an
6368 SMTP session. For example, it could not catch a HELO sent before the client
6369 waited for the greeting. There is now a check for outstanding input at the
6370 point when the greeting is written. Because of the duplex, asynchronous
6371 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6372 way, but not yet received, when the check is performed.
6374 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6375 on TCP/IP sockets, because this apparently causes some broken clients to
6378 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6379 unchanged) from the Cygwin maintainer.
6381 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6382 was supported instead of testing for mbx. Effectively a typo.
6384 15. The spa authenticator server code was not checking that the input it
6385 received was valid base64.
6387 16. The debug output line for the "set" modifier in ACLs was not showing the
6388 name of the variable that was being set.
6390 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6392 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6393 lookup on its own. Something else was needed to trigger the lookup. For
6394 example, a match in host_lookup or the need for a host name in a host list.
6395 Now, if $sender_host_name is referenced and the host name has not yet been
6396 looked up, a lookup is performed. If the lookup fails, the variable remains
6397 empty, and $host_lookup_failed is set to "1".
6399 19. Added "eqi" as a case-independent comparison operator.
6401 20. The saslauthd authentication condition could segfault if neither service
6402 nor realm was specified.
6404 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6405 error message that was logged was misleading, and incoming SMTP
6406 connections were dropped. The message is now more accurate, and temporary
6407 errors are given to SMTP connections.
6409 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6410 (except RSET) with a 421 error, until QUIT is received. However, it was
6411 failing to send a response to QUIT.
6413 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6414 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6415 ACL is now run after the helo_try_verify_hosts code.
6417 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6418 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6419 case-independent, so other case variants are also recognized). Apparently
6420 some systems use these upper case variants.
6422 25. If more than two messages were waiting for the same host, and a transport
6423 filter was specified for the transport, Exim sent two messages over the
6424 same TCP/IP connection, and then failed with "socket operation on non-
6425 socket" when it tried to send the third.
6427 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6429 27. The extern definition of crypt16() in expand.c was not being excluded when
6430 the OS had its own crypt16() function.
6432 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6433 as a preferred synonym for return_size_limit, both as an option and as an
6436 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6438 30. Changed the default configuration ACL to relax the local part checking rule
6439 for addresses that are not in any local domains. For these addresses,
6440 slashes and pipe symbols are allowed within local parts, but the sequence
6441 /../ is explicitly forbidden.
6443 31. SPA server authentication was not clearing the challenge buffer before
6446 32. log_message in a "warn" ACL statement was writing to the reject log as
6447 well as to the main log, which contradicts the documentation and doesn't
6448 seem right (because no rejection is happening). So I have stopped it.
6450 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6451 However, I am unable to do any testing of this.
6453 34. Fixed an infelicity in the appendfile transport. When checking directories
6454 for a mailbox, to see if any needed to be created, it was accidentally
6455 using path names with one or more superfluous leading slashes; tracing
6456 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6458 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6459 contents are added to the log line that is written for every discarded
6460 recipient. (Previously a log_message setting was ignored.)
6462 36. The ${quote: operator now quotes the string if it is empty.
6464 37. The install script runs exim in order to find its version number. If for
6465 some reason other than non-existence or emptiness, which it checks, it
6466 could not run './exim', it was installing it with an empty version number,
6467 i.e. as "exim-". This error state is now caught, and the installation is
6470 38. An argument was missing from the function that creates an error message
6471 when Exim fails to connect to the socket for saslauthd authentication.
6472 This could cause Exim to crash, or give a corrupted message.
6474 39. Added isip, isip4, and isip6 to ${if conditions.
6476 40. The ACL variables $acl_xx are now saved with the message, and can be
6477 accessed later in routers, transports, and filters.
6479 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6480 strings in the file are not string-expanded.
6482 42. If a MAIL command specified a SIZE value that was too large to fit into an
6483 int variable, the check against message_size_limit failed. Such values are
6484 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6485 day this will have to be increased, but I don't think I want to be around
6486 when emails are that large.
6493 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6494 iconv() is not standard in FreeBSD.
6496 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6497 IPv6 enabled. The observed symptom was a segmentation fault on return from
6498 the function os_common_find_running_interfaces() in src/os.c.
6500 3. In the check_special_case() function in daemon.c I had used "errno" as an
6501 argument name, which causes warnings on some systems. This was basically a
6502 typo, since it was named "eno" in the comments!
6504 4. The code that waits for the clock to tick (at a resolution of some fraction
6505 of a second) so as to ensure message-id uniqueness was always waiting for
6506 at least one whole tick, when it could have waited for less. [This is
6507 almost certainly not relevant at current processor speeds, where it is
6508 unlikely to ever wait at all. But we try to future-proof.]
6510 5. The function that sleeps for a time interval that includes fractions of a
6511 second contained a race. It did not block SIGALRM between setting the
6512 timer, and suspending (a couple of lines later). If the interval was short
6513 and the sigsuspend() was delayed until after it had expired, the suspension
6514 never ended. On busy systems this could lead to processes getting stuck for
6517 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6518 process, before it forks any delivery processes. The open lookup caching
6519 mechanism meant that the open file or database connection was passed into
6520 the delivery process. The problem was that delivery processes always tidy
6521 up cached lookup data. This could cause a problem for the next delivery
6522 process started by the queue runner, because the external queue runner
6523 process does not know about the closure. So the next delivery process
6524 still has data in the lookup cache. In the case of a file lookup, there was
6525 no problem because closing a file descriptor in a subprocess doesn't affect
6526 the parent. However, if the lookup was caching a connection to a database,
6527 the connection was closed, and the second delivery process was likely to
6528 see errors such as "PGSQL: query failed: server closed the connection
6529 unexpectedly". The problem has been fixed by closing all cached lookups
6530 in a queue runner before running a delivery process.
6532 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6533 seem to have the "const" qualifier which it has on other OS. I've
6536 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6537 *of the same type* (client or server) with the same public name that an
6538 error should be diagnosed.
6540 9. When Exim looked up a host name for an IP address, but failed to find the
6541 original IP address when looking up the host name (a safety check), it
6542 output the message "<ip address> does not match any IP for NULL", which was
6543 confusing, to say the least. The bug was that the host name should have
6544 appeared instead of "NULL".
6546 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6547 user that is built into the binary, and the -C or -D options is used, root
6548 privilege is dropped before the configuration file is read. In addition,
6549 logging is switched to stderr instead of the normal log files. If the
6550 configuration then re-defines the Exim user, the unprivileged environment
6551 is probably not what is expected, so Exim logs a panic warning message (but
6554 However, if deliver_drop_privilege is set, the unprivileged state may well
6555 be exactly what is intended, so the warning has been cut out in that case,
6556 and Exim is allowed to try to write to its normal log files.
6562 1. smtp_return_error_details was not giving details for temporary sender
6563 or receiver verification errors.
6565 2. Diagnose a configuration error if two authenticators have the same public
6568 3. Exim used not to create the message log file for a message until the first
6569 delivery attempt. This could be confusing when incoming messages were held
6570 for policy or load reasons. The message log file is now created at the time
6571 the message is received, and an initial "Received" line is written to it.
6573 4. The automatically generated man page for command line options had a minor
6574 bug that caused no ill effects; however, a more serious problem was that
6575 the procedure for building the man page automatically didn't always
6576 operate. Consequently, release 4.20 contains an out-of-date version. This
6577 shouldn't happen again.
6579 5. When building Exim with embedded Perl support, the script that builds the
6580 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6581 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6583 6. The freeze_tell option was not being used for messages that were frozen on
6584 arrival, either by an ACL or by local_scan().
6586 7. Added the smtp_incomplete_transaction log selector.
6588 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6589 was accepting AUTH without a new EHLO.
6591 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6592 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6595 10. Logging of TCP/IP connections (when configured) now happens in the main
6596 daemon process instead of the child process, so that the TCP/IP connection
6597 count is more accurate (but it can never be perfect).
6599 11. The use of "drop" in a nested ACL was not being handled correctly in the
6600 outer ACL. Now, if condition failure induced by the nested "drop" causes
6601 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6602 or "require"), the connection is dropped.
6604 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6605 that yield "discard" can now be used with an "accept" or a "discard" verb,
6606 but an error is generated for any others (because I can't see a useful way
6607 to define what should happen).
6609 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6610 are now processed in the same way as lines in the Exim configuration file.
6611 In particular, continuation lines are supported.
6613 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6615 15. Added -ti meaning -t -i.
6617 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6618 domains, and warn by logging if others are found.
6620 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6621 returned from GIFCONF when Exim is trying to find the list of interfaces on
6622 a host. The code in os.c has been modified to copy each ifreq to an aligned
6623 structure in all cases.
6625 Also, in some cases, the returned ifreq's were being copied to a 'struct
6626 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6627 means the last couple of bytes of an IPv6 address could be chopped if the
6628 ifreq contained only a normal sockaddr (14 bytes storage).
6630 18. Named domain lists were not supported in the hosts_treat_as_local option.
6631 An entry such as +xxxx was not recognized, and was treated as a literal
6634 19. Ensure that header lines added by a DATA ACL are included in the reject log
6635 if the ACL subsequently rejects the message.
6637 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6638 MD5 (which is deprecated).
6640 21. When testing a filter file using -bf, Exim was writing a message when it
6641 took the sender from a "From " line in the message, but it was not doing so
6642 when it took $return_path from a Return-Path: header line. It now does.
6644 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6645 with a valid header line field name (a series of printing characters
6646 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6648 23. Changed "disc" in the source to "disk" to conform to the documentation and
6649 the book and for uniformity.
6651 24. Ignore Sendmail's -Ooption=value command line item.
6653 25. When execve() failed while trying to run a command in a pipe transport,
6654 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6655 could be confused with a return value of 69 from the command itself. This
6656 has been changed to 127, the value the shell returns if it is asked to run
6657 a non-existent command. The wording for the related log line suggests a
6658 non-existent command as the problem.
6660 26. If received_header_text expands to an empty string, do not add a Received:
6661 header line to the message. (Well, it adds a token one on the spool, but
6662 marks it "old" so that it doesn't get used or transmitted.)
6664 27. Installed eximstats 1.28 (addition of -nt option).
6666 28. There was no check for failure on the call to getsockname() in the daemon
6667 code. This can fail if there is a shortage of resources on the system, with
6668 ENOMEM, for example. A temporary error is now given on failure.
6670 29. Contrary to the C standard, it seems that in some environments, the
6671 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6672 program. Exim now does this explicitly; it affects the formatting of
6673 timestamps using strftime().
6675 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6676 complaints. I've now initialized all the variables, to avoid this.
6678 32. Header lines added by a system filter were not being "seen" during
6679 transport-time rewrites.
6681 33. The info_callback() function passed to OpenSSL is set up with type void
6682 (*)(SSL *, int, int), as described somewhere. However, when calling the
6683 function (actually a macro) that sets it up, the type void(*)() is
6684 expected. I've put in a cast to prevent warnings from picky compilers.
6686 34. If a DNS black list lookup found a CNAME record, but there were no A
6687 records associated with the domain it pointed at, Exim crashed.
6689 35. If a DNS black list lookup returned more than one A record, Exim ignored
6690 all but the first. It now scans all returned addresses if a particular IP
6691 value is being sought. In this situation, the contents of the
6692 $dnslist_value variable are a list of all the addresses, separated by a
6695 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6696 to accept a host name and all its aliases if the forward lookup for any of
6697 them yielded the IP address of the incoming connection. Now it accepts only
6698 those names whose forward lookup yields the correct IP address. Any other
6699 names are discarded. This closes a loophole whereby a rogue DNS
6700 administrator could create reverse DNS records to break through a
6701 wildcarded host restriction in an ACL.
6703 37. If a user filter or a system filter that ran in a subprocess used any of
6704 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6705 the wrong values were passed to the pipe command ($thisaddress had the
6706 value of $0, $0 had the value of $1, etc). This bug was introduced by
6707 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6709 38. Improved the line breaking for long SMTP error messages from ACLs.
6710 Previously, if there was no break point between 40 and 75 characters, Exim
6711 left the rest of the message alone. Two changes have been made: (a) I've
6712 reduced the minimum length to 35 characters; (b) if it can't find a break
6713 point between 35 and 75 characters, it looks ahead and uses the first one
6714 that it finds. This may give the occasional overlong line, but at least the
6715 remaining text gets split now.
6717 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6718 file descriptors might be low, and that setting 1000 would always raise it.
6719 It turns out that in some environments, the limit is already over 1000 and
6720 that lowering it causes trouble. So now Exim takes care not to decrease it.
6722 40. When delivering a message, the value of $return_path is set to $sender_
6723 address at the start of routing (routers may change the value). By an
6724 oversight, this default was not being set up when an address was tested by
6725 -bt or -bv, which affected the outcome if any router or filter referred to
6728 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6729 log only when "message" or "log_message" are set. However, if one of the
6730 conditions was an address verification, or a call to a nested ACL, the
6731 messages generated by the underlying test were being passed through. This
6732 no longer happens. The underlying message is available in $acl_verify_
6733 message for both "message" and "log_message" expansions, so it can be
6734 passed through if needed.
6736 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6737 new expansion $bh_ to give the encoded byte string without charset
6738 translation. Translation happens only if iconv() is available; HAVE_ICONV
6739 indicates this at build time. HEADERS_CHARSET gives the charset to
6740 translate to; headers_charset can change it in the configuration, and
6741 "headers charset" can change it in an individual filter file.
6743 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6744 that creates RFC 2047 encoded "words" labels them as that charset instead
6745 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6746 expansion operator; (ii) when Exim creates a From: line for a local
6747 message; (iii) when a header line is rewritten to include a "phrase" part.
6749 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6750 buggy, causing it to skip the first lines of messages whose message ID
6751 ended in 'D'. This would not have bitten before Exim release 4.14, because
6752 message IDs were unlikely to end in 'D' before then. The effect was to have
6753 incorrect size information for certain domains.
6755 45. #include "config.h" was missing at the start of the crypt16.c module. This
6756 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6759 46. If there was a timeout during a "random" callout check, Exim treated it as
6760 a failure of the random address, and carried on sending RSET and the real
6761 address. If the delay was just some slowness somewhere, the response to the
6762 original RCPT would be taken as a response to RSET and so on, causing
6763 mayhem of various kinds.
6765 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6766 when I implemented it. It didn't allow for the fact that some option values
6767 may legitimately be negative (e.g. size_addition), and it didn't even do
6768 the right test for positive values.
6770 48. Domain names in DNS records are case-independent. Exim always looks them up
6771 in lower case. Some resolvers return domain names in exactly the case they
6772 appear in the zone file, that is, they may contain uppercase letters. Not
6773 all resolvers do this - some return always lower case. Exim was treating a
6774 change of case by a resolver as a change of domain, similar to a widening
6775 of a domain abbreviation. This triggered its re-routing code and so it was
6776 trying to route what was effectively the same domain again. This normally
6777 caused routing to fail (because the router wouldn't handle the domain
6778 twice). Now Exim checks for this case specially, and just changes the
6779 casing of the domain that it ultimately uses when it transmits the message
6782 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6785 50. If a filter generated a file delivery with a non-absolute name (possible if
6786 no home directory exists for the router), the forbid_file option was not
6789 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6790 the existing equality matching.
6792 52. Exim was using ints instead of ino_t variables in some places where it was
6793 dealing with inode numbers.
6795 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6796 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6797 called TMPDIR, and if it finds it is different, it changes its value.
6799 54. The smtp_printf() function is now made available to local_scan() so
6800 additional output lines can be written before returning. There is also an
6801 smtp_fflush() function to enable the detection of a dropped connection.
6802 The variables smtp_input and smtp_batched_input are exported to
6805 55. Changed the default runtime configuration: the message "Unknown user"
6806 has been removed from the ACL, and instead placed on the localuser router,
6807 using the cannot_route_message feature. This means that any verification
6808 failures that generate their own messages won't get overridden. Similarly,
6809 the "Unrouteable address" message that was in the ACL for unverifiable
6810 relay addresses has also been removed.
6812 56. Added hosts_avoid_esmtp to the smtp transport.
6814 57. The exicyclog script was not checking for the esoteric option
6815 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6816 will work only if exicyclog is run under the appropriate euid.
6818 58. Following a discussion on the list, the rules by which Exim recognises line
6819 endings on incoming messages have been changed. The -dropcr and drop_cr
6820 options are now no-ops, retained only for backwards compatibility. The
6821 following line terminators are recognized: LF CRLF CR. However, special
6822 processing applies to CR:
6824 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6825 nor a local message in the state where . is a terminator.
6827 (ii) If a bare CR is encountered in a header line, an extra space is added
6828 after the line terminator so as not to end the header. The reasoning
6829 behind this is that bare CRs in header lines are most likely either
6830 to be mistakes, or people trying to play silly games.
6832 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6833 was being incorrectly given as 18 bytes larger than it should have been.
6834 This is a VOB (very old bug).
6836 60. This may never have affected anything current, but just in case it has:
6837 When the local host is found other than at the start of a list of hosts,
6838 the local host, those with the same MX, and any that follow, are discarded.
6839 When the list in question was part of a longer list of hosts, the following
6840 hosts (not currently being processed) were also being discarded. This no
6841 longer happens. I'm not sure if this situation could ever has previously
6844 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6847 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6848 header. This is something that is recommended in a new Internet Draft, and
6849 is something that is documented as being done by Sendmail. There are two
6850 possible values. For messages generated by the autoreply transport, Exim
6853 Auto-Submitted: auto-replied
6855 whereas for all other generated messages (e.g. bounces) it adds
6857 Auto-Submitted: auto-generated
6859 63. The "personal" condition in filters now includes a test for the
6860 Auto-Submitted: header. If it contains the string "auto-" the message it
6861 not considered personal.
6863 64. Added rcpt_include_affixes as a generic transport option.
6865 65. Added queue_only_override (default true).
6867 66. Added the syslog_duplication option.
6869 67. If what should have been the first header line of a message consisted of
6870 a space followed by a colon, Exim was mis-interpreting it as a header line.
6871 It isn't of course - it is syntactically invalid and should therefore be
6872 treated as the start of the message body. The misbehaviour could have
6873 caused a number of strange effects, including loss of data in subsequent
6874 header lines, and spool format errors.
6876 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6877 client host had authenticated. This control can now be exercised by an ACL
6878 for more flexibility.
6880 69. By default, callouts do not happen when testing with -bh. There is now a
6881 variant, -bhc, which does actually run the callout code, including
6882 consulting and updating the callout cache.
6884 70. Added support for saslauthd authentication, courtesy of Alexander
6887 71. If statvfs() failed on the spool or log directories while checking their
6888 size for availability, Exim confusingly gave the error "space shortage".
6889 Furthermore, in debugging mode it crashed with a floating point exception.
6890 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6891 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6892 serious problem, Exim now writes to the main and panic logs when this
6893 happens, with details of the failure. It then refuses to accept the
6894 incoming message, giving the message "spool directory problem" or "log
6895 directory problem" with a 421 code for SMTP messages.
6897 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6898 0, 1, and 2 exist, because some OS complain for execs without them (see
6899 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6900 descriptors. However, the code omitted to check that the open succeeded,
6901 causing mysterious errors if for some reason the permissions on /dev/null
6902 got screwed. Now Exim writes a message to the main and panic logs, and
6903 bombs out if it can't open /dev/null.
6905 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6906 interact so that it is all more flexible. It is supposed to remain
6907 backwards compatible. Also added extra_local_interfaces.
6909 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6910 to bomb out with an assertion failure - to the client this appears as a
6911 connection drop. This problem occurs in the part of the code that was taken
6912 from the Samba project. Fortunately, the assertion is in a very simple
6913 function, so I have fixed this by reproducing the function inline in the
6914 one place where it is called, and arranging for authentication to fail
6915 instead of killing the process with assert().
6917 75. The SPA client code was not working when the server requested OEM rather
6918 than Unicode encoding.
6920 76. Added code to make require_files with a specific uid setting more usable in
6921 the case where statting the file as root fails - usually a non-root-mounted
6922 NFS file system. When this happens and the failure is EACCES, Exim now
6923 forks a subprocess and does the per-uid checking as the relevant uid.
6925 77. Added process_log_path.
6927 78. If log_file_path was not explicitly set, a setting of check_log_space or
6928 check_log_inodes was ignored.
6930 79. If a space check for the spool or log partitions fails, the incident is now
6931 logged. Of course, in the latter case the data may get lost...
6933 80. Added the %p formatting code to string_format() so that it can be used to
6934 print addresses in debug_print(). Adjusted all the address printing in the
6935 debugging in store.c to use %p rather than %d.
6937 81. There was a concern that a line of code in smtp_in.c could overflow a
6938 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6939 initially expressed, the concern was not well-founded, because trailing
6940 spaces are removed early. However, if the trailing spaces were followed by
6941 a NULL, they did not get removed, so the overflow was possible. Two fixes
6944 (a) I re-wrote the offending code in a cleaner fashion.
6945 (b) If an incoming SMTP command contains a NULL character, it is rejected
6948 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6949 runs initgroups(), so that if the Exim user is in any additional groups,
6950 they will be used during message reception.
6956 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob