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
198 freed. CVE-2017-16943.
200 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
201 from SMTP input. Previously it was always done; now only done for DATA
202 and not BDAT commands. CVE-2017-16944.
208 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
209 than -2003 did; needs libidn2 in addition to libidn.
211 JH/02 The path option on a pipe transport is now expanded before use.
213 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
214 Patch provided by "Björn", documentation fix added too.
216 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
217 missing a wire-to-host endian conversion.
219 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
220 close after a BDAT command line could be taken as a following command,
221 giving a synch failure. Fix by only checking for synch immediately
222 before acknowledging the chunk.
224 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
225 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
226 macro. Patches provided by Josh Soref.
228 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
229 Previously we did not; the RFC seems ambiguous and VRFY is not listed
230 by IANA as a service extension. However, John Klensin suggests that we
233 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
234 the dkim code may be unix-mode line endings rather than smtp wire-format
235 CRLF, so prepend a CR to any bare LF.
237 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
238 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
240 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
241 the first were themselves being wrongly included in the feed into dkim
242 processing; with most chunk sizes in use this resulted in an incorrect
243 body hash calculated value.
245 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
246 DKIM signature block, for verification. Although advised against by
247 standards it is specifically not ruled illegal.
249 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
251 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
252 missing a body hash (the bh= tag).
254 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
255 It seems that HAProxy sends the Proxy Protocol information in clear and
256 only then does a TLS startup, so do the same.
258 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
259 TCP connections (such as for Spamd) unless the daemon successfully set
260 Fast Open mode on its listening sockets. This fixes breakage seen on
261 too-old kernels or those not configured for Fast Open, at the cost of
262 requiring both directions being enabled for TFO, and TFO never being used
263 by non-daemon-related Exim processes.
265 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
266 endings, at least on the first header line. Try to canonify any that get
267 past that check, despite the cost.
269 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
270 now limited to an arbitrary five deep, while parsing addresses with the
271 strip_excess_angle_brackets option enabled.
273 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
274 instead leave the unprompted TLS handshake in socket buffer for the
275 TLS library to consume.
277 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
279 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
281 JH/16 Drop variables when they go out of scope. Memory management drops a whole
282 region in one operation, for speed, and this leaves assigned pointers
283 dangling. Add checks run only under the testsuite which checks all
284 variables at a store-reset and panics on a dangling pointer; add code
285 explicitly nulling out all the variables discovered. Fixes one known
286 bug: a transport crash, where a dangling pointer for $sending_ip_address
287 originally assigned in a verify callout, is re-used.
289 PP/06 Drop '.' from @INC in various Perl scripts.
291 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
293 PP/08 Reduce a number of compilation warnings under clang; building with
294 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
295 should be warning-free.
297 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
299 HS/01 Fix portability problems introduced by PP/08 for platforms where
300 realloc(NULL) is not equivalent to malloc() [SunOS et al].
302 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
303 chunk. This allows us to accept broken chunked messages. We need a more
304 general solution here.
306 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
307 already-broken messages in the queue.
309 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
311 JH/19 Fix reference counting bug in routing-generated-address tracking.
317 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
318 supports it and a size is available (ie. the sending peer gave us one).
320 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
321 years of being deprecated). The replacements are the ACLs
322 acl_smtp_mime and acl_not_smtp_mime.
324 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
325 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
326 or even an in-clear connection were permitted. Now, if the host lookup
327 was dnssec and dane was requested then the host is only used if the
328 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
329 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
330 if one fails this test.
331 This means that a poorly-configured remote DNS will make it incommunicado;
332 but it protects against a DNS-interception attack on it.
334 JH/04 Bug 1810: make continued-use of an open smtp transport connection
335 non-noisy when a race steals the message being considered.
337 JH/05 If main configuration option tls_certificate is unset, generate a
338 self-signed certificate for inbound TLS connections.
340 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
341 in rewrites and routers.
343 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
344 and logged a warning sing 4.83; now they are a configuration file error.
346 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
347 (lacking @domain). Apply the same qualification processing as RCPT.
349 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
351 JH/10 Support ${sha256:} applied to a string (as well as the previous
354 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
355 a cutthrough deliver is pending, as we always want to make a connection.
356 This also avoids re-routing the message when later placing the cutthrough
357 connection after a verify cache hit.
358 Do not update it with the verify result either.
360 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
361 when routing results in more than one destination address.
363 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
364 signing (which inhibits the cutthrough capability). Previously only
365 the presence of an option was tested; now an expansion evaluating as
366 empty is permissible (obviously it should depend only on data available
367 when the cutthrough connection is made).
369 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
370 the relevant preceding SMTP command did not note the pipelining mode.
372 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
373 Previously they were not counted.
375 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
376 as one having no matching records. Previously we deferred the message
377 that needed the lookup.
379 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
380 distinguished as "(=".
382 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
383 for missing MX records. Previously it only worked for missing A records.
385 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
387 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
388 after the data-go-ahead and data-ack. Patch from Jason Betts.
390 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
391 even for a "none" policy. Patch from Tony Meyer.
393 JH/22 Fix continued use of a connection for further deliveries. If a port was
394 specified by a router, it must also match for the delivery to be
397 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
398 When one of the recipients of a message was unsuitable for the connection
399 (has no matching addresses), we lost track of needing to mark it
400 deferred. As a result mail would be lost.
402 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
404 JH/25 Decoding ACL controls is now done using a binary search; the source code
405 takes up less space and should be simpler to maintain. Merge the ACL
406 condition decode tables also, with similar effect.
408 JH/26 Fix problem with one_time used on a redirect router which returned the
409 parent address unchanged. A retry would see the parent address marked as
410 delivered, so not attempt the (identical) child. As a result mail would
413 JH/27 Fix a possible security hole, wherein a process operating with the Exim
414 UID can gain a root shell. Credit to http://www.halfdog.net/ for
415 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
418 JH/28 Enable {spool,log} filesystem space and inode checks as default.
419 Main config options check_{log,spool}_{inodes,space} are now
420 100 inodes, 10MB unless set otherwise in the configuration.
422 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
423 Previously it only applied to the main-section connection policy
426 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
428 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
429 by me. Added RFC7919 DH primes as an alternative.
431 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
432 are not in the system include path.
434 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
435 GnuTLS, when a session startup failed (eg because the client disconnected)
436 Exim did stdio operations after fclose. This was exposed by a recent
437 change which nulled out the file handle after the fclose.
439 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
440 signed directly by the cert-signing cert, rather than an intermediate
441 OCSP-signing cert. This is the model used by LetsEncrypt.
443 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
445 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
446 an incoming connection.
448 HS/02 Bug 1802: Do not half-close the connection after sending a request
451 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
452 fallback to "prime256v1".
454 JH/34 SECURITY: Use proper copy of DATA command in error message.
455 Could leak key material. Remotely exploitable. CVE-2016-9963.
461 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
462 and 3.4.4 - once the server is enabled to respond to an OCSP request
463 it does even when not requested, resulting in a stapling non-aware
464 client dropping the TLS connection.
466 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
467 support variable-length bit vectors. No functional change.
469 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
470 The I= interface field on outgoing lines is now after the H= remote
471 host field, same as incoming lines. There is a separate
472 outgoing_interface log selector which allows you to disable the
475 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
476 If not running log_selector +smtp_connection the mainlog would be held
477 open indefinitely after a "too many connections" event, including to a
478 deleted file after a log rotate. Leave the per net connection logging
479 leaving it open for efficiency as that will be quickly detected by the
480 check on the next write.
482 HS/01 Bug 1671: Fix post transport crash.
483 Processing the wait-<transport> messages could crash the delivery
484 process if the message IDs didn't exist for some reason. When
485 using 'split_spool_directory=yes' the construction of the spool
486 file name failed already, exposing the same netto behaviour.
488 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
489 mime_regex ACL conditions.
491 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
492 to DSN fail messages (bounces): remote IP, remote greeting, remote response
493 to HELO, local diagnostic string.
495 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
496 log line to debug. Even when configured with a tls authenticator many
497 client connections are expected to not authenticate in this way, so
498 an authenticate fail is not an error.
500 HS/02 Add the Exim version string to the process info. This way exiwhat
501 gives some more detail about the running daemon.
503 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
504 matter for fast-change records such as DNSBLs.
506 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
507 retry record, even if constant. There may be multiple transports with
508 different interface settings and the retry behaviour needs to be kept
511 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
513 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
515 JH/10 Bug 840: fix log_defer_output option of pipe transport
517 JH/11 Bug 830: use same host for all RCPTS of a message, even under
518 hosts_randomize. This matters a lot when combined with mua_wrapper.
520 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
521 ${quote_pgsql:<string>} operator.
523 JH/13 Bug 1708: avoid misaligned access in cached lookup.
525 JH/14 Change header file name for freeradius-client. Relevant if compiling
526 with Radius support; from the Gentoo tree and checked under Fedora.
528 JH/15 Bug 1712: Introduce $prdr_requested flag variable
530 JH/16 Bug 1714: Permit an empty string as expansion result for transport
531 option transport_filter, meaning no filtering.
533 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
535 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
536 defaults to "*" (all hosts). The variable is now available when not built
537 with TLS, default unset, mainly to enable keeping the testsuite sane.
538 If a server certificate is not supplied (via tls_certificate) an error is
539 logged, and clients will find TLS connections fail on startup. Presumably
540 they will retry in-clear.
541 Packagers of Exim are strongly encouraged to create a server certificate
542 at installation time.
544 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
545 with the $config_file variable.
547 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
548 in transport context, after the attempt, and per-recipient. The latter type
549 is per host attempted. The event data is the error message, and the errno
550 information encodes the lookup type (A vs. MX) used for the (first) host,
551 and the trailing two digits of the smtp 4xx response.
553 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
554 to write to mainlog (or rejectlog, paniclog) in the window between file
555 creation and permissions/ownership being changed. Particularly affects
556 installations where exicyclog is run as root, rather than exim user;
557 result is that the running daemon panics and dies.
559 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
561 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
562 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
563 "pri" and "weight". Note that the previous implicit priority given by the
564 list order is no longer honoured.
566 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
569 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
570 by defining SUPPORT_SOCKS.
572 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
573 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
574 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
575 variables are renamed to proxy_{local,external}_{address,port}.
577 JH/27 Move Internationalisation support from Experimental to mainline, enabled
578 for a build by defining SUPPORT_I18N
580 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
581 of the query string, and make ${quote_redis:} do that quoting.
583 JH/29 Move Events support from Experimental to mainline, enabled by default
584 and removable for a build by defining DISABLE_EVENT.
586 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
588 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
589 cached by the daemon.
591 JH/32 Move Redis support from Experimental to mainline, enabled for a build
592 by defining LOOKUP_REDIS. The libhiredis library is required.
594 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
595 keys are given for lookup.
597 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
598 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
599 only supported when built with TLS support. The PolarSSL SHA routines
600 are still used when the TLS library is too old for convenient support.
602 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
603 openssl_options), for security. OpenSSL forces this from version 1.1.0
604 server-side so match that on older versions.
606 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
607 allocation for $value could be released as the expansion processing
608 concluded, but leaving the global pointer active for it.
610 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
611 and to use the domains and local_parts ACL conditions.
613 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
614 incorrectly not doubled on cutthrough transmission, hence seen as a
615 body-termination at the receiving system - resulting in truncated mails.
616 Commonly the sender saw a TCP-level error, and retransmitted the message
617 via the normal store-and-forward channel. This could result in duplicates
618 received - but deduplicating mailstores were liable to retain only the
619 initial truncated version.
621 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
623 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
625 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
626 we're in there, support oversigning also; bug 1309.
628 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
630 HS/04 Add support for keep_environment and add_environment options.
632 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
633 either intentional arithmetic overflow during PRNG, or testing config-
636 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
637 delivery resulted in actual delivery. Cancel cutthrough before DATA
640 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
641 rejects a recipient: pass the reject to the originator.
643 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
644 Many were false-positives and ignorable, but it's worth fixing the
647 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
648 for the new environment-manipulation done at startup. Move the routines
649 from being local to tls.c to being global via the os.c file.
651 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
652 an extract embedded as result-arg for a map, the first arg for extract
653 is unavailable so we cannot tell if this is a numbered or keyed
654 extraction. Accept either.
660 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
663 JH/02 The smtp transport option "multi_domain" is now expanded.
665 JH/03 The smtp transport now requests PRDR by default, if the server offers
668 JH/04 Certificate name checking on server certificates, when exim is a client,
669 is now done by default. The transport option tls_verify_cert_hostnames
670 can be used to disable this per-host. The build option
671 EXPERIMENTAL_CERTNAMES is withdrawn.
673 JH/05 The value of the tls_verify_certificates smtp transport and main options
674 default to the word "system" to access the system default CA bundle.
675 For GnuTLS, only version 3.0.20 or later.
677 JH/06 Verification of the server certificate for a TLS connection is now tried
678 (but not required) by default. The verification status is now logged by
679 default, for both outbound TLS and client-certificate supplying inbound
682 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
685 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
686 Status Notification (bounce) messages are now MIME format per RFC 3464.
687 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
688 under the control of the dsn_advertise_hosts option, and routers may
689 have a dsn_lasthop option.
691 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
692 default, modifiable by a malware= option. The list separator for
693 the options can now be changed in the usual way. Bug 68.
695 JH/10 The smtp_receive_timeout main option is now expanded before use.
697 JH/11 The incoming_interface log option now also enables logging of the
698 local interface on delivery outgoing connections.
700 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
701 if the interface and destination host and port all match.
703 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
706 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
707 Patch from Andrew Lewis.
709 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
710 now supports optional time-restrictions, weighting, and priority
711 modifiers per server. Patch originally by <rommer@active.by>.
713 JH/16 The spamd_address main option now supports a mixed list of local
714 and remote servers. Remote servers can be IPv6 addresses, and
715 specify a port-range.
717 JH/17 Bug 68: The spamd_address main option now supports an optional
718 timeout value per server.
720 JH/18 Bug 1581: Router and transport options headers_add/remove can
721 now have the list separator specified.
723 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
726 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
729 JH/21 Support for the A6 type of dns record is withdrawn.
731 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
732 rather than the verbs used.
734 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
735 from 255 to 1024 chars.
737 JH/24 Verification callouts now attempt to use TLS by default.
739 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
740 are generic router options now. The defaults didn't change.
742 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
743 Original patch from Alexander Shikoff, worked over by JH.
745 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
746 files if ZCAT_COMMAND is not executable.
748 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
750 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
752 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
753 Normally benign, it bites when the pair was led to by a CNAME;
754 modern usage is to not canonicalize the domain to a CNAME target
755 (and we were inconsistent anyway for A-only vs AAAA+A).
757 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
759 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
760 when evaluating $sender_host_dnssec.
762 JH/31 Check the HELO verification lookup for DNSSEC, adding new
763 $sender_helo_dnssec variable.
765 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
767 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
769 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
771 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
772 documented as working, but never had. Support all but $spam_report.
774 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
775 added for tls authenticator.
777 HS/03 Add perl_taintmode main config option
783 TL/01 When running the test suite, the README says that variables such as
784 no_msglog_check are global and can be placed anywhere in a specific
785 test's script, however it was observed that placement needed to be near
786 the beginning for it to behave that way. Changed the runtest perl
787 script to read through the entire script once to detect and set these
788 variables, reset to the beginning of the script, and then run through
789 the script parsing/test process like normal.
791 TL/02 The BSD's have an arc4random API. One of the functions to induce
792 adding randomness was arc4random_stir(), but it has been removed in
793 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
794 function when detected.
796 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
797 cause callback expansion.
799 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
800 syntax errors in an expansion can be treated as a string instead of
801 logging or causing an error, due to the internal use of bool_lax
802 instead of bool when processing it.
804 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
805 server certificates when making smtp deliveries.
807 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
809 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
811 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
813 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
814 Merged patch from Sebastian Wiedenroth.
816 JH/05 Fix results-pipe from transport process. Several recipients, combined
817 with certificate use, exposed issues where response data items split
818 over buffer boundaries were not parsed properly. This eventually
819 resulted in duplicates being sent. This issue only became common enough
820 to notice due to the introduction of connection certificate information,
821 the item size being so much larger. Found and fixed by Wolfgang Breyha.
823 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
824 size buffer was used, resulting in syntax errors when an expansion
827 JH/07 Add support for directories of certificates when compiled with a GnuTLS
828 version 3.3.6 or later.
830 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
831 is EXPERIMENTAL_EVENT, the main-configuration and transport options
832 both become "event_action", the variables become $event_name, $event_data
833 and $event_defer_errno. There is a new variable $verify_mode, usable in
834 routers, transports and related events. The tls:cert event is now also
835 raised for inbound connections, if the main configuration event_action
838 TL/06 In test suite, disable OCSP for old versions of openssl which contained
839 early OCSP support, but no stapling (appears to be less than 1.0.0).
841 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
842 server certificate names available under the smtp transport option
843 "tls_verify_cert_hostname" now do not permit multi-component wildcard
846 JH/10 Time-related extraction expansions from certificates now use the main
847 option "timezone" setting for output formatting, and are consistent
848 between OpenSSL and GnuTLS compilations. Bug 1541.
850 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
851 encoded parameter in the incoming message. Bug 1558.
853 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
854 include certificate info, eximon was claiming there were spoolfile
857 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
859 JH/14 Log delivery-related information more consistently, using the sequence
860 "H=<name> [<ip>]" wherever possible.
862 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
863 are problematic for Debian distribution, omit them from the release
866 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
868 JH/16 Fix string representation of time values on 64bit time_t architectures.
871 JH/17 Fix a null-indirection in certextract expansions when a nondefault
872 output list separator was used.
877 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
878 checkers that were complaining about end of non-void function with no
881 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
882 This was a regression introduced in 4.83 by another bugfix.
884 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
886 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
887 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
893 TF/01 Correctly close the server side of TLS when forking for delivery.
895 When a message was received over SMTP with TLS, Exim failed to clear up
896 the incoming connection properly after forking off the child process to
897 deliver the message. In some situations the subsequent outgoing
898 delivery connection happened to have the same fd number as the incoming
899 connection previously had. Exim would try to use TLS and fail, logging
900 a "Bad file descriptor" error.
902 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
903 utilities have not been installed.
905 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
906 temporary space as the ACL may create new global variables.
908 TL/01 LDAP support uses per connection or global context settings, depending
909 upon the detected version of the libraries at build time.
911 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
912 to extract and use the src ip:port in logging and expansions as if it
913 were a direct connection from the outside internet. PPv2 support was
914 updated based on HAProxy spec change in May 2014.
916 JH/02 Add ${listextract {number}{list}{success}{fail}}.
918 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
919 Properly escape header and check for NULL return.
921 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
924 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
926 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
927 characters in header names, implemented as a verify condition.
928 Contributed by Michael Fischer v. Mollard.
930 TL/05 Rename SPF condition results err_perm and err_temp to standardized
931 results permerror and temperror. Previous values are deprecated but
932 still accepted. In a future release, err_perm and err_temp will be
933 completely removed, which will be a backward incompatibility if the
934 ACL tests for either of these two old results. Patch contributed by
935 user bes-internal on the mailing list.
937 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
939 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
940 selectors, in both main and reject logs.
942 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
945 JH/07 Add malware type "sock" for talking to simple daemon.
947 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
949 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
950 routers/transports under cutthrough routing.
952 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
953 numbers. Touch up "bool" conditional to keep the same definition.
955 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
957 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
959 JH/12 Expand items in router/transport headers_add or headers_remove lists
960 individually rather than the list as a whole. Bug 1452.
962 Required for reasonable handling of multiple headers_ options when
963 they may be empty; requires that headers_remove items with embedded
964 colons must have them doubled (or the list-separator changed).
966 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
967 view the policy declared in the DMARC record. Currently, $dmarc_status
968 is a combined value of both the record presence and the result of the
971 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
973 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
974 dnslookup router and the smtp transport (applying to the forward
977 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
978 of ldap servers used for a specific lookup. Patch provided by Heiko
981 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
982 New variable $lookup_dnssec_authenticated for observability.
984 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
985 Patch submitted by Lars Timman.
987 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
989 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
990 Requires trusted mode and valid format message id, aborts otherwise.
991 Patch contributed by Heiko Schlichting.
993 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
994 certextract with support for various fields. Bug 1358.
996 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
997 is requested by default, modifiable by smtp transport option
1000 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1001 operate on certificate variables to give certificate fingerprints
1002 Also new ${sha256:cert_variable}.
1004 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1006 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1007 Christian Aistleitner.
1009 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1011 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1012 file. Patch from Wolfgang Breyha.
1014 JH/25 Expand the coverage of the delivery $host and $host_address to
1015 client authenticators run in verify callout. Bug 1476.
1017 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1018 align with daemon_smtp_ports. Bug 72.
1020 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1021 support and error reporting did not work properly.
1023 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1024 and is readable. Patch from Andrew Colin Kissa.
1026 TL/14 Enhance documentation of ${run expansion and how it parses the
1027 commandline after expansion, particularly in the case when an
1028 unquoted variable expansion results in an empty value.
1030 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1032 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1033 by renaming away from it. Addresses GH issue 15, reported by
1036 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1038 TL/15 SECURITY: prevent double expansion in math comparison functions
1039 (can expand unsanitized data). Not remotely exploitable.
1046 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1048 PP/02 Make -n do something, by making it not do something.
1049 When combined with -bP, the name of an option is not output.
1051 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1054 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1055 $sender_host_name and config options to manage this, and basic check
1058 PP/05 DSCP support for outbound connections and control modifier for inbound.
1060 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1061 (Only plugin which currently uses this is kerberos4, which nobody should
1062 be using, but we should make it available and other future plugins might
1063 conceivably use it, even though it would break NAT; stuff *should* be
1064 using channel bindings instead).
1066 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1067 name; added for Sendmail compatibility; requires admin caller.
1068 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1069 just ignore it); requires trusted caller.
1070 Also parse but ignore: -Ac -Am -X<logfile>
1073 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1075 TL/02 Add +smtp_confirmation as a default logging option.
1077 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1078 Patch by Magnus Holmgren from 2007-02-20.
1080 TL/04 Bugzilla 1281 - Spec typo.
1081 Bugzilla 1283 - Spec typo.
1082 Bugzilla 1290 - Spec grammar fixes.
1084 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1086 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1088 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1089 mailing list by by Dmitry Isaikin.
1091 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1093 JH/02 Support "G" suffix to numbers in ${if comparisons.
1095 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1097 NM/01 Bugzilla 1197 - Spec typo
1098 Bugzilla 1196 - Spec examples corrections
1100 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1102 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1103 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1106 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1107 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1109 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1110 "acl {{name}{arg}...}", and optional args on acl condition
1113 JH/05 Permit multiple router/transport headers_add/remove lines.
1115 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1117 JH/07 Avoid using a waiting database for a single-message-only transport.
1118 Performance patch from Paul Fisher. Bugzilla 1262.
1120 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1123 JH/09 Add $headers_added variable, with content from use of ACL modifier
1124 add_header (but not yet added to the message). Bugzilla 199.
1126 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1127 Pulled from Bugzilla 817 by Wolfgang Breyha.
1129 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1131 (nb: this is the same fix as in Exim 4.80.1)
1133 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1136 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1139 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1141 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1142 advertise SMTP AUTH mechanism to us, instead of a generic
1143 protocol violation error. Also, make Exim more robust to bad
1144 data from the Dovecot auth socket.
1146 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1148 When a queue runner is handling a message, Exim first routes the
1149 recipient addresses, during which it prunes them based on the retry
1150 hints database. After that it attempts to deliver the message to
1151 any remaining recipients. It then updates the hints database using
1154 So if a recipient address works intermittently, it can get repeatedly
1155 deferred at routing time. The retry hints record remains fresh so the
1156 address never reaches the final cutoff time.
1158 This is a fairly common occurrence when a user is bumping up against
1159 their storage quota. Exim had some logic in its local delivery code
1160 to deal with this. However it did not apply to per-recipient defers
1161 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1163 This change adds a proper retry rule check during routing so that the
1164 final cutoff time is checked against the message's age. We only do
1165 this check if there is an address retry record and there is not a
1166 domain retry record; this implies that previous attempts to handle
1167 the address had the retry_use_local_parts option turned on. We use
1168 this as an approximation for the destination being like a local
1169 delivery, as in LMTP.
1171 I suspect this new check makes the old local delivery cutoff check
1172 redundant, but I have not verified this so I left the code in place.
1174 TF/02 Correct gecos expansion when From: is a prefix of the username.
1176 Test 0254 submits a message to Exim with the header
1180 When I ran the test suite under the user fanf2, Exim expanded
1181 the header to contain my full name, whereas it should have added
1182 a Resent-Sender: header. It erroneously treats any prefix of the
1183 username as equal to the username.
1185 This change corrects that bug.
1187 GF/01 DCC debug and logging tidyup
1188 Error conditions log to paniclog rather than rejectlog.
1189 Debug lines prefixed by "DCC: " to remove any ambiguity.
1191 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1193 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1194 Bug spotted by Jeremy Harris; was flawed since initial commit.
1195 Would have resulted in OCSP responses post-SNI triggering an Exim
1196 NULL dereference and crash.
1198 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1200 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1201 Bug detection, analysis and fix by Samuel Thibault.
1202 Bugzilla 1331, Debian bug #698092.
1204 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1206 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1207 Server implementation by Todd Lyons, client by JH.
1208 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1209 config variable "prdr_enable" controls whether the server
1210 advertises the facility. If the client requests PRDR a new
1211 acl_data_smtp_prdr ACL is called once for each recipient, after
1212 the body content is received and before the acl_smtp_data ACL.
1213 The client is controlled by both of: a hosts_try_prdr option
1214 on the smtp transport, and the server advertisement.
1215 Default client logging of deliveries and rejections involving
1216 PRDR are flagged with the string "PRDR".
1218 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1219 fclose(). Diagnosis by Todd Lyons.
1221 PP/17 Update configure.default to handle IPv6 localhost better.
1222 Patch by Alain Williams (plus minor tweaks).
1225 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1226 This is now consistent with GnuTLS, and is now documented: the
1227 previous undocumented portable approach to treating the option as
1228 unset was to force an expansion failure. That still works, and
1229 an empty string is now equivalent.
1231 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1232 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1233 not performing validation itself.
1235 PP/20 Added force_command boolean option to pipe transport.
1236 Patch from Nick Koston, of cPanel Inc.
1238 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1241 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1243 PP/21 Fix eximon continuous updating with timestamped log-files.
1244 Broken in a format-string cleanup in 4.80, missed when I repaired the
1245 other false fix of the same issue.
1246 Report and fix from Heiko Schlichting.
1249 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1250 Report from Prashanth Katuri.
1252 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1253 It's SecureTransport, so affects any MacOS clients which use the
1254 system-integrated TLS libraries, including email clients.
1256 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1257 using a MIME ACL for non-SMTP local injection.
1258 Report and assistance in diagnosis by Warren Baker.
1260 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1262 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1264 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1265 last id that failed so it may be referenced in subsequent ACL's.
1267 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1270 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1271 ldap library initialization, allowing self-signed CA's to be
1272 used. Also properly sets require_cert option later in code by
1273 using NULL (global ldap config) instead of ldap handle (per
1274 session). Bug diagnosis and testing by alxgomz.
1276 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1277 the src/util/ subdirectory.
1279 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1280 renamed to Transport Post Delivery Action by Jeremy Harris, as
1283 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1284 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1285 redis_servers = needs to be configured which will be used by the redis
1286 lookup. Patch from Warren Baker, of The Packet Hub.
1288 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1290 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1291 hostname or reverse DNS when processing a host list. Used suggestions
1292 from multiple comments on this bug.
1294 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1296 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1297 Missed a few lines, added it to make the runtest require no keyboard
1300 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1301 contains upper case chars. Make router use caseful_local_part.
1303 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1304 support when GnuTLS has been built with p11-kit.
1310 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1312 This, or similar/improved, will also be change PP/11 of 4.82.
1318 PP/01 Handle short writes when writing local log-files.
1319 In practice, only affects FreeBSD (8 onwards).
1320 Bugzilla 1053, with thanks to Dmitry Isaikin.
1322 NM/01 Bugzilla 949 - Documentation tweak
1324 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1327 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1329 PP/02 Implemented gsasl authenticator.
1331 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1333 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1334 `pkg-config foo` for cflags/libs.
1336 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1337 with rest of GSASL and with heimdal_gssapi.
1339 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1340 `pkg-config foo` for cflags/libs for the TLS implementation.
1342 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1343 properties get this fed in as external SSF. A number of robustness
1344 and debugging improvements to the cyrus_sasl authenticator.
1346 PP/08 cyrus_sasl server now expands the server_realm option.
1348 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1349 Patch by Jeremy Harris.
1351 PP/10 Added dbmjz lookup type.
1353 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1355 PP/12 MAIL args handles TAB as well as SP, for better interop with
1356 non-compliant senders.
1357 Analysis and variant patch by Todd Lyons.
1359 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1360 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1361 Patch from Dirk Mueller <dmueller@suse.com>
1363 PP/13 tls_peerdn now print-escaped for spool files.
1364 Observed some $tls_peerdn in wild which contained \n, which resulted
1365 in spool file corruption.
1367 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1368 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1369 or write after TLS renegotiation, which otherwise led to messages
1372 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1373 as a tracking header (ie: a signed header comes before the signature).
1374 Patch from Wolfgang Breyha.
1376 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1377 comma-sep list; embedded commas doubled.
1379 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1381 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1383 Report and patch from Dmitry Banschikov.
1385 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1386 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1387 failures. We appear to now support TLS1.1+ with Exim.
1389 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1390 lets Exim select keys and certificates based upon TLS SNI from client.
1391 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1392 before an outbound SMTP session. New log_selector, +tls_sni.
1394 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1395 NULL dereference. Report and patch from Alun Jones.
1397 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1398 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1399 Not seeing resolver debug output on NetBSD, but suspect this is a
1400 resolver implementation change.
1402 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1403 Left warnings. Added "eximon gdb" invocation mode.
1405 PP/21 Defaulting "accept_8bitmime" to true, not false.
1407 PP/22 Added -bw for inetd wait mode support.
1409 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1410 locate the relevant includes and libraries. Made this the default.
1412 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1413 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1415 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1416 This may cause build issues on older platforms.
1418 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1419 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1420 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1421 Added SNI support via GnuTLS too.
1422 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1424 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1426 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1427 Applied second patch from Janne, implementing suggestion to default
1428 multiple-strings-in-record handling to match SPF spec.
1430 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1432 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1433 read-only, out of scope).
1434 Patch from Wolfgang Breyha, report from Stuart Northfield.
1436 PP/29 Fix three issues highlighted by clang analyser static analysis.
1437 Only crash-plausible issue would require the Cambridge-specific
1438 iplookup router and a misconfiguration.
1439 Report from Marcin Mirosław.
1441 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1443 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1444 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1445 As part of this, removing so much warning spew let me fix some minor
1446 real issues in debug logging.
1448 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1449 assignment on my part. Fixed.
1451 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1452 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1453 Janne Snabb (who went above and beyond: thank you).
1455 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1456 string otherwise requires a connection and a bunch more work and it's
1457 relatively easy to get wrong. Should also expose TLS library linkage
1460 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1461 64-bit ${eval} (JH/03).
1463 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1464 GNU libc to support some of the 64-bit stuff, should not lead to
1465 conflicts. Defined before os.h is pulled in, so if a given platform
1466 needs to override this, it can.
1468 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1469 protection layer was required, which is not implemented.
1470 Bugzilla 1254, patch from Wolfgang Breyha.
1472 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1473 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1474 tls_dhparam take prime identifiers. Also unbreak combination of
1475 OpenSSL+DH_params+TLSSNI.
1477 PP/39 Disable SSLv2 by default in OpenSSL support.
1483 PP/01 Solaris build fix for Oracle's LDAP libraries.
1484 Bugzilla 1109, patch from Stephen Usher.
1486 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1488 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1491 TF/02 Fix a couple more cases where we did not log the error message
1492 when unlink() failed. See also change 4.74-TF/03.
1494 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1495 lock up or crash if it happened to be inside a call to libc when it
1496 got a SIGUSR1 from exiwhat.
1498 The SIGUSR1 handler appends the current process status to the process
1499 log which is later printed by exiwhat. It used to use the general
1500 purpose logging code to do this, but several functions it calls are
1501 not safe for signals.
1503 The new output code in the SIGUSR1 handler is specific to the process
1504 log, and simple enough that it's easy to inspect for signal safety.
1505 Removing some special cases also simplifies the general logging code.
1506 Removing the spurious timestamps from the process log simplifies
1509 TF/04 Improved ratelimit ACL condition.
1511 The /noupdate option has been deprecated in favour of /readonly which
1512 has clearer semantics. The /leaky, /strict, and /readonly update modes
1513 are mutually exclusive. The update mode is no longer included in the
1514 database key; it just determines when the database is updated. (This
1515 means that when you upgrade Exim will forget old rate measurements.)
1517 Exim now checks that the per_* options are used with an update mode that
1518 makes sense for the current ACL. For example, when Exim is processing a
1519 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1520 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1521 must specify per_mail/readonly. If you omit the update mode it defaults to
1522 /leaky where that makes sense (as before) or /readonly where required.
1524 The /noupdate option is now undocumented but still supported for
1525 backwards compatibility. It is equivalent to /readonly except that in
1526 ACLs where /readonly is required you may specify /leaky/noupdate or
1527 /strict/noupdate which are treated the same as /readonly.
1529 A useful new feature is the /count= option. This is a generalization
1530 of the per_byte option, so that you can measure the throughput of other
1531 aggregate values. For example, the per_byte option is now equivalent
1532 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1534 The per_rcpt option has been generalized using the /count= mechanism
1535 (though it's more complicated than the per_byte equivalence). When it is
1536 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1537 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1538 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1539 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1540 Note that using per_rcpt with a non-readonly update mode in more than
1541 one ACL will cause the recipients to be double-counted. (The per_mail
1542 and per_byte options don't have this problem.)
1544 The handling of very low rates has changed slightly. If the computed rate
1545 is less than the event's count (usually one) then this event is the first
1546 after a long gap. In this case the rate is set to the same as this event's
1547 count, so that the first message of a spam run is counted properly.
1549 The major new feature is a mechanism for counting the rate of unique
1550 events. The new per_addr option counts the number of different
1551 recipients that someone has sent messages to in the last time period. It
1552 behaves like per_rcpt if all the recipient addresses are different, but
1553 duplicate recipient addresses do not increase the measured rate. Like
1554 the /count= option this is a general mechanism, so the per_addr option
1555 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1556 example, measure the rate that a client uses different sender addresses
1557 with the options per_mail/unique=$sender_address. There are further
1558 details in the main documentation.
1560 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1562 TF/06 Removed a few PCRE remnants.
1564 TF/07 Automatically extract Exim's version number from tags in the git
1565 repository when doing development or release builds.
1567 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1568 Bugzilla 879. Patch from Paul Fisher.
1570 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1571 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1574 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1576 PP/05 Variable $av_failed, true if the AV scanner deferred.
1577 Bugzilla 1078. Patch from John Horne.
1579 PP/06 Stop make process more reliably on build failure.
1580 Bugzilla 1087. Patch from Heiko Schlittermann.
1582 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1583 Bugzilla 1089. Patch from Heiko Schlittermann.
1585 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1586 Bugzilla 1131. Patch from Holger Weiß.
1588 PP/09 Handle IPv6 addresses with SPF.
1589 Bugzilla 860. Patch from Wolfgang Breyha.
1591 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1593 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1596 PP/11 match_* no longer expand right-hand-side by default.
1597 New compile-time build option, EXPAND_LISTMATCH_RHS.
1598 New expansion conditions, "inlist", "inlisti".
1600 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1602 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1604 PP/14 fix log_write() format string regression from TF/03.
1605 Bugzilla 1152. Patch from Dmitry Isaikin.
1611 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1613 PP/02 Harmonised TLS library version reporting; only show if debugging.
1614 Layout now matches that introduced for other libraries in 4.74 PP/03.
1616 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1618 PP/04 New "dns_use_edns0" global option.
1620 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1623 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1624 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1626 TK/01 Updated PolarSSL code to 0.14.2.
1627 Bugzilla 1097. Patch from Andreas Metzler.
1629 PP/07 Catch divide-by-zero in ${eval:...}.
1630 Fixes bugzilla 1102.
1632 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1635 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1636 format-string attack -- SECURITY: remote arbitrary code execution.
1638 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1639 time unintentionally subject to list matching rules, letting the header
1640 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1641 arbitrary string expansion). This allowed for information disclosure.
1643 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1644 INT_MIN/-1 -- value coerced to INT_MAX.
1650 NM/01 Workaround for PCRE version dependency in version reporting
1653 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1654 This fixes portability to compilers other than gcc, notably
1655 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1657 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1658 makefiles for portability to HP-UX and POSIX correctness.
1660 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1661 Also via indented variable definition in the Makefile.
1662 (Debugging by Oliver Heesakkers).
1664 PP/02 Restore caching of spamd results with expanded spamd_address.
1665 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1667 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1668 Improves build reliability. Fix from: Frank Elsner
1670 NM/02 Fix wide character breakage in the rfc2047 coding
1671 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1673 NM/03 Allow underscore in dnslist lookups
1674 Fixes bug 1026. Patch from Graeme Fowler
1676 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1677 Code patches from Adam Ciarcinski of NetBSD.
1679 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1682 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1683 is logged, to avoid truncation. Patch from John Horne.
1685 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1686 Patch from Jakob Hirsch.
1688 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1689 SQL string expansion failure details.
1690 Patch from Andrey Oktyabrski.
1692 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1693 Patch from Simon Arlott.
1695 PP/09 New lookups functionality failed to compile on old gcc which rejects
1696 extern declarations in function scope.
1697 Patch from Oliver Fleischmann
1699 PP/10 Use sig_atomic_t for flags set from signal handlers.
1700 Check getgroups() return and improve debugging.
1701 Fixed developed for diagnosis in bug 927 (which turned out to be
1704 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1705 Patch from Mark Zealey.
1707 PP/12 Bugzilla 1056: Improved spamd server selection.
1708 Patch from Mark Zealey.
1710 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1711 Based on patch from Heiko Schlittermann.
1713 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1714 Patch from Uwe Doering, sign-off by Michael Haardt.
1716 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1717 variable declaration deep within a block. Bug and patch from
1720 PP/15 lookups-Makefile IRIX compatibility coercion.
1722 PP/16 Make DISABLE_DKIM build knob functional.
1724 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1725 Patch by Simon Arlott
1727 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1728 variable argument macros. Our copy now differs from upstream.
1734 TF/01 Failure to get a lock on a hints database can have serious
1735 consequences so log it to the panic log.
1737 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1738 controlled using the smtp_confirmation log selector.
1740 TF/03 Include the error message when we fail to unlink a spool file.
1742 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1743 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1744 for maintaining out-of-tree patches for some time.
1746 PP/01 Bugzilla 139: Documentation and portability issues.
1747 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1748 Handle per-OS dynamic-module compilation flags.
1750 PP/02 Let /dev/null have normal permissions.
1751 The 4.73 fixes were a little too stringent and complained about the
1752 permissions on /dev/null. Exempt it from some checks.
1753 Reported by Andreas M. Kirchwitz.
1755 PP/03 Report version information for many libraries, including
1756 Exim version information for dynamically loaded libraries. Created
1757 version.h, now support a version extension string for distributors
1758 who patch heavily. Dynamic module ABI change.
1760 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1761 privilege escalation vulnerability whereby the Exim run-time user
1762 can cause root to append content of the attacker's choosing to
1765 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1768 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1769 If dropping privileges for untrusted macros, we disabled normal logging
1770 on the basis that it would fail; for the Exim run-time user, this is not
1771 the case, and it resulted in successful deliveries going unlogged.
1772 Fixed. Reported by Andreas Metzler.
1778 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1779 only prepend for the Resent-* case. Fixes regression introduced in
1780 Exim 4.70 by NM/22 for Bugzilla 607.
1782 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1783 increasing numbers of administrators be bitten by this.
1785 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1787 PP/03 Bugzilla 994: added openssl_options main configuration option.
1789 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1791 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1793 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1795 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1796 without a peer certificate, leading to a segfault because of an
1797 assumption that peers always have certificates. Be a little more
1798 paranoid. Problem reported by Martin Tscholak.
1800 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1801 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1802 NB: ClamAV planning to remove STREAM in "middle of 2010".
1803 CL also introduces -bmalware, various -d+acl logging additions and
1804 more caution in buffer sizes.
1806 PP/09 Implemented reverse_ip expansion operator.
1808 PP/10 Bugzilla 937: provide a "debug" ACL control.
1810 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1812 PP/12 Bugzilla 973: Implement --version.
1814 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1816 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1818 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1820 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1821 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1822 ignore trailing whitespace.
1824 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1826 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1829 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1830 Notification from Dr Andrew Aitchison.
1832 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1833 ExtendedDetectionInfo response format.
1834 Notification from John Horne.
1836 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1839 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1840 XSL and documented dependency on system catalogs, with examples of how
1843 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1846 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1847 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1848 configuration file which is writeable by the Exim user or group.
1850 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1851 of configuration files to cover files specified with the -C option if
1852 they are going to be used with root privileges, not just the default
1855 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1856 option (effectively making it always true).
1858 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1859 files to be used while preserving root privileges.
1861 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1862 that rogue child processes cannot use them.
1864 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1865 run-time user, instead of root.
1867 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1868 Exim run-time user without dropping privileges.
1870 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1871 result string, instead of calling string_vformat() twice with the same
1874 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1875 for other users. Others should always drop root privileges if they use
1876 -C on the command line, even for a whitelisted configure file.
1878 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1880 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1886 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1887 $data_path, and $header_path variables; fixed documentation bugs and
1890 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1891 exipick to access non-standard spools, including the "frozen" queue
1894 NM/01 Bugzilla 965: Support mysql stored procedures.
1895 Patch from Alain Williams
1897 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1899 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1900 Patch from Andreas Metzler
1902 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1903 Patch from Kirill Miazine
1905 NM/05 Bugzilla 671: Added umask to procmail example.
1907 JJ/03 installed exipick 20100323.0, fixing doc bug
1909 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1910 directory. Notification and patch from Dan Rosenberg.
1912 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1914 TK/02 Improve log output when DKIM signing operation fails.
1916 MH/01 Treat the transport option dkim_domain as a colon separated
1917 list, not as a single string, and sign the message with each element,
1918 omitting multiple occurences of the same signer.
1920 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1921 Bugzilla 985, 986. Patch by Simon Arlott
1923 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1924 Patch by Simon Arlott
1926 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1927 MBX locking. Notification from Dan Rosenberg.
1933 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1935 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1937 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1939 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1941 NM/04 Bugzilla 915: Flag broken perl installation during build.
1947 TK/01 Added patch by Johannes Berg that expands the main option
1948 "spamd_address" if it starts with a dollar sign.
1950 TK/02 Write list of recipients to X-Envelope-Sender header when building
1951 the mbox-format spool file for content scanning (suggested by Jakob
1954 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1955 (http://www.dcc-servers.net/) support via dccifd. Activated by
1956 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1958 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1959 by Mark Daniel Reidel <mr@df.eu>.
1961 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1962 When building exim an external PCRE library is now needed -
1963 PCRE is a system library on the majority of modern systems.
1964 See entry on PCRE_LIBS in EDITME file.
1966 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1967 conversation. Added nologin parameter to request.
1968 Patch contributed by Kirill Miazine.
1970 TF/01 Do not log submission mode rewrites if they do not change the address.
1972 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1974 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1975 log files in place. Contributed by Roberto Lima.
1977 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1979 TF/03 Bugzilla 615: When checking the local_parts router precondition
1980 after a local_part_suffix or local_part_prefix option, Exim now
1981 does not use the address's named list lookup cache, since this
1982 contains cached lookups for the whole local part.
1984 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1985 Robert Millan. Documentation is in experimental-spec.txt.
1987 TF/04 Bugzilla 668: Fix parallel build (make -j).
1989 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1991 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1992 Patch provided by Jan Srzednicki.
1994 TF/05 Leading white space used to be stripped from $spam_report which
1995 wrecked the formatting. Now it is preserved.
1997 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1998 that they are available at delivery time.
2000 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2002 TF/08 TLS error reporting now respects the incoming_interface and
2003 incoming_port log selectors.
2005 TF/09 Produce a more useful error message if an SMTP transport's hosts
2006 setting expands to an empty string.
2008 NM/06 Bugzilla 744: EXPN did not work under TLS.
2009 Patch provided by Phil Pennock.
2011 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2012 Patch provided by Richard Godbee.
2014 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2015 acl_smtp_notquit, added index entry.
2017 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2018 Patch provided by Eugene Bujak.
2020 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2021 accept(). Patch provided by Maxim Dounin.
2023 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2024 Patch provided by Phil Pennock.
2026 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2028 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2029 Patch provided by Brad "anomie" Jorsch.
2031 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2032 Patch provided by Dean Brooks.
2034 TK/05 Add native DKIM support (does not depend on external libraries).
2036 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2037 Patch provided by Graeme Fowler.
2039 NM/16 Bugzilla 851: Documentation example syntax fix.
2041 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2043 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2046 NM/19 Bugzilla 745: TLS version reporting.
2047 Patch provided by Phil Pennock.
2049 NM/20 Bugzilla 167: bool: condition support.
2050 Patch provided by Phil Pennock.
2052 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2053 clients. Patch provided by Phil Pennock.
2055 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2056 Patch provided by Brad "anomie" Jorsch.
2058 NM/23 Bugzilla 687: Fix misparses in eximstats.
2059 Patch provided by Heiko Schlittermann.
2061 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2062 Patch provided by Heiko Schlittermann.
2064 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2065 plus update to original patch.
2067 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2069 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2070 Patch provided by David Brownlee.
2072 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2074 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2076 NM/30 Bugzilla 888: TLS documentation bugfixes.
2078 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2080 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2081 Unlike the original bugzilla I have changed all shell scripts in src tree.
2083 NM/33 Bugzilla 898: Transport filter timeout fix.
2084 Patch by Todd Rinaldo.
2086 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2087 Patch by Serge Demonchaux.
2089 NM/35 Bugzilla 39: Base64 decode bug fixes.
2090 Patch by Jakob Hirsch.
2092 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2094 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2096 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2098 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2104 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2105 ALT-N's libdkim that I have put here:
2106 http://duncanthrax.net/exim-experimental/
2108 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2109 were called 'true' and it seems that C99 defines that as a reserved
2110 keyword to be used with 'bool' variable types. That means you could
2111 not include C99-style headers which use bools without triggering
2112 build errors in sieve.c.
2114 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2115 as mailq or other aliases. Changed the --help handling significantly
2116 to do whats expected. exim_usage() emits usage/help information.
2118 SC/01 Added the -bylocaldomain option to eximstats.
2120 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2122 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2124 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2130 PH/01 Another patch from the Sieve maintainer.
2132 PH/02 When an IPv6 address is converted to a string for single-key lookup
2133 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2134 dots are used instead of colons so that keys in lsearch files need not
2135 contain colons. This was done some time before quoting was made available
2136 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2137 (notated using the quote facility) so as to distinguish them from IPv4
2138 keys. This meant that lookups for IP addresses in host lists did not work
2139 for iplsearch lookups.
2141 This has been fixed by arranging for IPv6 addresses to be expressed with
2142 colons if the lookup type is iplsearch. This is not incompatible, because
2143 previously such lookups could never work.
2145 The situation is now rather anomalous, since one *can* have colons in
2146 ordinary lsearch keys. However, making the change in all cases is
2147 incompatible and would probably break a number of configurations.
2149 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2152 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2153 conversion specification without a maximum field width, thereby enabling
2154 a rogue spamd server to cause a buffer overflow. While nobody in their
2155 right mind would setup Exim to query an untrusted spamd server, an
2156 attacker that gains access to a server running spamd could potentially
2157 exploit this vulnerability to run arbitrary code as the Exim user.
2159 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2160 $primary_hostname instead of what libspf2 thinks the hosts name is.
2162 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2163 a directory entry by the name of the lookup key. Previously, if a
2164 symlink pointed to a non-existing file or a file in a directory that
2165 Exim lacked permissions to read, a lookup for a key matching that
2166 symlink would fail. Now it is enough that a matching directory entry
2167 exists, symlink or not. (Bugzilla 503.)
2169 PH/03 The body_linecount and body_zerocount variables are now exported in the
2172 PH/04 Added the $dnslist_matched variable.
2174 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2175 This means they are set thereafter only if the connection becomes
2178 PH/06 Added the client_condition to authenticators so that some can be skipped
2179 by clients under certain conditions.
2181 PH/07 The error message for a badly-placed control=no_multiline_responses left
2182 "_responses" off the end of the name.
2184 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2186 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2187 (without spaces) instead of just copying the configuration text.
2189 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2191 PH/11 Added $max_received_linelength.
2193 PH/12 Added +ignore_defer and +include_defer to host lists.
2195 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2196 way in which PCRE > 7.0 is built.
2198 PH/14 Implemented queue_only_load_latch.
2200 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2201 MAIL command. The effect was to mangle the value on 64-bit systems.
2203 PH/16 Another patch from the Sieve maintainer.
2205 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2207 PH/18 If a system quota error occurred while trying to create the file for
2208 a maildir delivery, the message "Mailbox is full" was not appended to the
2209 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2210 only to a quota excession during the actual writing of the file.
2212 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2213 characters?) which causes problems in log lines. The DN values are now
2214 passed through string_printing() before being added to log lines.
2216 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2217 and InterBase are left for another time.)
2219 PH/21 Added message_body_newlines option.
2221 PH/22 Guard against possible overflow in moan_check_errorcopy().
2223 PH/23 POSIX allows open() to be a macro; guard against that.
2225 PH/24 If the recipient of an error message contained an @ in the local part
2226 (suitably quoted, of course), incorrect values were put in $domain and
2227 $local_part during the evaluation of errors_copy.
2233 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2234 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2237 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2238 issue a MAIL command.
2240 PH/02 In an ACL statement such as
2242 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2244 if a client was not listed at all, or was listed with a value other than
2245 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2246 the condition was not true (as it should be), so access was not denied.
2247 The bug was that the ! inversion was incorrectly passed on to the second
2248 item. This has been fixed.
2250 PH/03 Added additional dnslists conditions == and =& which are different from
2251 = and & when the dns lookup returns more than one IP address.
2253 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2254 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2256 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2257 FSYNC, which compiles an option called disable_fsync that allows for
2258 bypassing fsync(). The documentation is heavily laced with warnings.
2260 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2262 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2263 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2264 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2265 including adding "make clean"; (3) Added -fPIC when compiling the test
2266 dynamically loaded module, to get rid of a warning.
2268 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2269 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2270 = 0s. The bug is otherwise harmless.
2272 PH/07 There was a bug in the dovecot authenticator such that the value of
2273 $auth1 could be overwritten, and so not correctly preserved, after a
2274 successful authentication. This usually meant that the value preserved by
2275 the server_setid option was incorrect.
2277 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2279 PH/09 Installed PCRE release 7.0.
2281 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2282 run for batched SMTP input. It is now run at the start of every message
2283 in the batch. While fixing this I discovered that the process information
2284 (output by running exiwhat) was not always getting set for -bs and -bS
2285 input. This is fixed, and it now also says "batched" for BSMTP.
2287 PH/11 Added control=no_pipelining.
2289 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2290 patch, slightly modified), and move the expansion of helo_data till after
2291 the connection is made in the smtp transport (so it can use these
2294 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2296 PH/14 Added log_selector = +pid.
2298 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2300 PH/16 Add ${if forany and ${if forall.
2302 PH/17 Added dsn_from option to vary the From: line in DSNs.
2304 PH/18 Flush SMTP output before performing a callout, unless control =
2305 no_callout_flush is set.
2307 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2308 was true (the default) a successful delivery failed to delete the retry
2309 item, thus causing premature timeout of the address. The bug is now
2312 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2314 PH/21 Long custom messages for fakedefer and fakereject are now split up
2315 into multiline responses in the same way that messages for "deny" and
2316 other ACL rejections are.
2318 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2319 with slight modification.
2321 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2322 draft, changing the syntax and factoring some duplicate code".
2324 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2325 for deliveries of the second and subsequent messages over the same SMTP
2328 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2329 ${reduce, with only minor "tidies".
2331 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2333 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2334 expansion side effects.
2336 PH/27 When a message times out after an over-quota error from an Exim-imposed
2337 quota, the bounce message says "mailbox is full". This message was not
2338 being given when it was a system quota that was exceeded. It now should
2341 MH/03 Made $recipients available in local_scan(). local_scan() already has
2342 better access to the recipient list through recipients_list[], but
2343 $recipients can be useful in postmaster-provided expansion strings.
2345 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2346 in the case of a MAIL command with additional options following the
2347 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2348 were accidentally chopped off.
2350 PH/29 SMTP synchronization checks are implemented when a command is read -
2351 there is a check that no more input is waiting when there shouldn't be
2352 any. However, for some commands, a delay in an ACL can mean that it is
2353 some time before the response is written. In this time, more input might
2354 arrive, invalidly. So now there are extra checks after an ACL has run for
2355 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2356 pipelining has not been advertised.
2358 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2360 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2361 returned in the SMTP response when smtp_return_error_details was false.
2362 This has been fixed.
2364 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2365 instead of the C I/O that was originally supplied, because problems were
2366 reported on Solaris.
2368 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2369 Exim which did not show up earlier: it was assuming that a call to
2370 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2371 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2372 was a macro that became an assignment, so it seemed to work. This has
2373 changed to a proper function call with a void return, hence the compile
2374 error. Exim's code has been fixed.
2376 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2379 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2381 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2383 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2384 because it turns out to be too big or there is a timeout) there is no
2385 "Completed" line in the log. When some messages of this type were
2386 selected by exigrep, they were listed as "not completed". Others were
2387 picked up by some special patterns. I have improved the selection
2388 criteria to be more general.
2390 PH/38 The host_find_failed option in the manualroute router can now be set
2391 to "ignore", to completely ignore a host whose IP address cannot be
2392 found. If all hosts are ignored, the behaviour is controlled by the new
2393 host_all_ignored option.
2395 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2396 homing or because of multiple MX records with /mx) generated more than
2397 one IP address, and the following item turned out to be the local host,
2398 all the secondary addresses of the first item were incorrectly removed
2399 from the list, along with the local host and any following hosts (which
2400 is what is supposed to happen).
2402 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2403 whoever called Exim into the -H file. In the case of the daemon it was
2404 behaving confusingly. When first started, it used values for whoever
2405 started the daemon, but after a SIGHUP it used the Exim user (because it
2406 calls itself on a restart). I have changed the code so that it now always
2409 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2410 message are rejected with the same error (e.g. no authentication or bad
2411 sender address), and a DATA command is nevertheless sent (as can happen
2412 with PIPELINING or a stupid MUA), the error message that was given to the
2413 RCPT commands is included in the rejection of the DATA command. This is
2414 intended to be helpful for MUAs that show only the final error to their
2417 PH/42 Another patch from the Sieve maintainer.
2419 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2420 Eximstats - Fixed some broken HTML links and added missing column headers
2422 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2423 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2425 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2427 PH/43 Yet another patch from the Sieve maintainer.
2429 PH/44 I found a way to check for a TCP/IP connection going away before sending
2430 the response to the final '.' that terminates a message, but only in the
2431 case where the client has not sent further data following the '.'
2432 (unfortunately, this is allowed). However, in many cases there won't be
2433 any further data because there won't be any more messages to send. A call
2434 to select() can be used: if it shows that the input is "ready", there is
2435 either input waiting, or the socket has been closed. An attempt to read
2436 the next input character can distinguish the two cases. Previously, Exim
2437 would have sent an OK response which the client would never have see.
2438 This could lead to message repetition. This fix should cure that, at
2439 least in a lot of common cases.
2441 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2442 advertised in response to EHLO.
2448 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2449 fixed by 4.65/MH/01 (is this a record?) are fixed:
2451 (i) An empty string was always treated as zero by the numeric comparison
2452 operators. This behaviour has been restored.
2454 (ii) It is documented that the numeric comparison operators always treat
2455 their arguments as decimal numbers. This was broken in that numbers
2456 starting with 0 were being interpreted as octal.
2458 While fixing these problems I realized that there was another issue that
2459 hadn't been noticed. Values of message_size_limit (both the global option
2460 and the transport option) were treated as octal if they started with 0.
2461 The documentation was vague. These values are now always treated as
2462 decimal, and I will make that clear in the documentation.
2468 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2469 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2472 MH/01 Don't check that the operands of numeric comparison operators are
2473 integers when their expansion is in "skipping" mode (fixes bug
2474 introduced by 4.64-PH/07).
2476 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2477 child addresses, Exim now panics and dies. Previously, because the count
2478 is held in a short int, deliveries were likely to be lost. As such a
2479 large number of recipients for a single message is ridiculous
2480 (performance will be very, very poor), I have chosen to impose a limit
2481 rather than extend the field.
2487 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2488 leftover -K file (the existence of which was triggered by #402).
2489 While we were at it, introduced process PID as part of the -K
2490 filename. This should rule out race conditions when creating
2493 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2494 processing considerably. Previous code took too long for large mails,
2495 triggering a timeout which in turn triggers #401.
2497 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2498 in the DK code in transports.c. sendfile() is not really portable,
2499 hence the _LINUX specificness.
2501 TF/01 In the add_headers option to the mail command in an Exim filter,
2502 there was a bug that Exim would claim a syntax error in any
2503 header after the first one which had an odd number of characters
2506 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2507 callout verification, Exim cached a "reject" for the entire domain. This
2508 is correct for most verifications, but it is not correct for a recipient
2509 verification with use_sender or use_postmaster set, because in that case
2510 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2511 case of MAIL FROM:<> rejection from other early rejections (e.g.
2512 rejection of HELO). When verifying a recipient using a non-null MAIL
2513 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2514 Whatever the result of the callout, the value of the domain cache is
2515 left unchanged (for any other kind of callout, getting as far as trying
2516 RCPT means that the domain itself is ok).
2518 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2521 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2522 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2523 session, the daemon ignores ECONNECT errors and logs others; it now
2524 ignores EPIPE as well.
2526 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2527 (quoted-printable decoding).
2529 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2530 later the small subsequent patch to fix an introduced bug.
2532 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2534 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2536 PH/08 An error is now given if message_size_limit is specified negative.
2538 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2539 to be given (somewhat) arbitrary names.
2541 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2544 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2545 miscellaneous code fixes
2547 PH/10 Added the log_reject_target ACL modifier to specify where to log
2550 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2551 hostname. This is wrong, because it relates to the incoming message (and
2552 probably the interface on which it is arriving) and not to the outgoing
2553 callout (which could be using a different interface). This has been
2554 changed to use the value of the helo_data option from the smtp transport
2555 instead - this is what is used when a message is actually being sent. If
2556 there is no remote transport (possible with a router that sets up host
2557 addresses), $smtp_active_hostname is used.
2559 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2560 tweaks were necessary in order to get it to work (see also 21 below):
2561 (a) The code assumed that strncpy() returns a negative number on buffer
2562 overflow, which isn't the case. Replaced with Exim's string_format()
2564 (b) There were several signed/unsigned issues. I just did the minimum
2565 hacking in of casts. There is scope for a larger refactoring.
2566 (c) The code used strcasecmp() which is not a standard C function.
2567 Replaced with Exim's strcmpic() function.
2568 (d) The code set only $1; it now sets $auth1 as well.
2569 (e) A simple test gave the error "authentication client didn't specify
2570 service in request". It would seem that Dovecot has changed its
2571 interface. Fortunately there's a specification; I followed it and
2572 changed what the client sends and it appears to be working now.
2574 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2577 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2578 address A is aliased to B and C, where B exists and C does not. Without
2579 -v the output is "A verified" because verification stops after a
2580 successful redirection if more than one address is generated. However,
2581 with -v the child addresses are also verified. Exim was outputting "A
2582 failed to verify" and then showing the successful verification for C,
2583 with its parentage. It now outputs "B failed to verify", showing B's
2584 parentage before showing the successful verification of C.
2586 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2587 look up a TXT record in a specific list after matching in a combined
2590 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2591 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2592 they consult the DNS. I had assumed they would set it the way they
2593 wanted; and indeed my experiments on Linux seem to show that in some
2594 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2595 To be on the safe side, however, I have now made the interface to
2596 host_find_byname() similar to host_find_bydns(), with an argument
2597 containing the DNS resolver options. The host_find_byname() function now
2598 sets these options at its start, just as host_find_bydns() does. The smtp
2599 transport options dns_qualify_single and dns_search_parents are passed to
2600 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2601 of host_find_byname() use the default settings of RES_DEFNAMES
2602 (qualify_single) but not RES_DNSRCH (search_parents).
2604 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2605 spool_read_header() do less string testing, by means of a preliminary
2606 switch on the second character of optional "-foo" lines. (This is
2607 overdue, caused by the large number of possibilities that now exist.
2608 Originally there were few.) While I was there, I also converted the
2609 str(n)cmp tests so they don't re-test the leading "-" and the first
2610 character, in the hope this might squeeze out yet more improvement.
2612 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2613 flag allowing group syntax was set by the header_syntax check but not
2614 turned off, possible causing trouble later; (2) The flag was not being
2615 set at all for the header_verify test, causing "group"-style headers to
2616 be rejected. I have now set it in this case, and also caused header_
2617 verify to ignore an empty address taken from a group. While doing this, I
2618 came across some other cases where the code for allowing group syntax
2619 while scanning a header line wasn't quite right (mostly, not resetting
2620 the flag correctly in the right place). These bugs could have caused
2621 trouble for malformed header lines. I hope it is now all correct.
2623 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2624 with the "reply" argument non-NULL. The code, however (which originally
2625 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2626 but it didn't always do it. This confused somebody who was copying the
2627 code for some other use. I have removed all the tests.
2629 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2630 feature that was used to support insecure browsers during the U.S. crypto
2631 embargo. It requires special client support, and Exim is probably the
2632 only MTA that supported it -- and would never use it because real RSA is
2633 always available. This code has been removed, because it had the bad
2634 effect of slowing Exim down by computing (never used) parameters for the
2635 RSA_EXPORT functionality.
2637 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2638 authenticator to fail if there's a tab character in the incoming data
2639 (there should never be unless someone is messing about, as it's supposed
2640 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2641 if the connection is using TLS or if the remote IP is the same as the
2642 local IP, and the "valid-client-cert option" if a client certificate has
2645 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2646 authenticators. This can be used for authorization after authentication
2647 succeeds. (In the case of plaintext, it servers for both authentication
2650 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2651 if any retry times were supplied.
2653 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2654 connection, where there is no client IP address to check. In this
2655 situation, the verify now always succeeds.
2657 PH/25 Applied John Jetmore's -Mset patch.
2659 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2661 PH/27 In a string expansion for a processed (not raw) header when multiple
2662 headers of the same name were present, leading whitespace was being
2663 removed from all of them, but trailing whitespace was being removed only
2664 from the last one. Now trailing whitespace is removed from each header
2665 before concatenation. Completely empty headers in a concatenation (as
2666 before) are ignored.
2668 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2669 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2671 PH/29 [Removed. This was a change that I later backed out, and forgot to
2672 correct the ChangeLog entry (that I had efficiently created) before
2673 committing the later change.]
2675 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2676 address errors (4xx response to RCPT) over the same connection as other
2677 messages routed to the same hosts. Such deliveries are always "forced",
2678 so retry times are not inspected. This resulted in far too many retries
2679 for the affected addresses. The effect occurred only when there were more
2680 hosts than the hosts_max_try setting in the smtp transport when it had
2681 the 4xx errors. Those hosts that it had tried were not added to the list
2682 of hosts for which the message was waiting, so if all were tried, there
2683 was no problem. Two fixes have been applied:
2685 (i) If there are any address or message errors in an SMTP delivery, none
2686 of the hosts (tried or untried) are now added to the list of hosts
2687 for which the message is waiting, so the message should not be a
2688 candidate for sending over the same connection that was used for a
2689 successful delivery of some other message. This seems entirely
2690 reasonable: after all the message is NOT "waiting for some host".
2691 This is so "obvious" that I'm not sure why it wasn't done
2692 previously. Hope I haven't missed anything, but it can't do any
2693 harm, as the worst effect is to miss an optimization.
2695 (ii) If, despite (i), such a delivery is accidentally attempted, the
2696 routing retry time is respected, so at least it doesn't keep
2697 hammering the server.
2699 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2700 in ${readsocket because some servers need this prod.
2702 PH/32 Added some extra debug output when updating a wait-xxx database.
2704 PH/33 The hint "could be header name not terminated by colon", which has been
2705 given for certain expansion errors for a long time, was not being given
2706 for the ${if def:h_colon_omitted{... case.
2708 PH/34 The spec says: "With one important exception, whenever a domain list is
2709 being scanned, $domain contains the subject domain." There was at least
2710 one case where this was not true.
2712 PH/35 The error "getsockname() failed: connection reset by peer" was being
2713 written to the panic log as well as the main log, but it isn't really
2714 panic-worthy as it just means the connection died rather early on. I have
2715 removed the panic log writing for the ECONNRESET error when getsockname()
2718 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2719 runs only) independently of the message's sender address. This meant
2720 that, if the 4xx error was in fact related to the sender, a different
2721 message to the same recipient with a different sender could confuse
2722 things. In particular, this can happen when sending to a greylisting
2723 server, but other circumstances could also provoke similar problems.
2724 I have changed the default so that the retry time for these errors is now
2725 based a combination of the sender and recipient addresses. This change
2726 can be overridden by setting address_retry_include_sender=false in the
2729 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2730 remote server are returned as part of bounce messages. This was not
2731 happening for LMTP over a pipe (the lmtp transport), but now it is the
2732 same for both kinds of LMTP.
2734 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2735 in header lines that were in fact CNAMEs. This is no longer the case.
2737 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2738 and queue runs started by the daemon processed all messages. This has
2739 been fixed so that -R and -S can now usefully be given with -q<time>.
2741 PH/40 Import PCRE release 6.7 (fixes some bugs).
2743 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2745 PH/42 Give an error if -q is specified more than once.
2747 PH/43 Renamed the variables $interface_address and $interface_port as
2748 $received_ip_address and $received_port, to make it clear that these
2749 values apply to message reception, and not to the outgoing interface when
2750 a message is delivered. (The old names remain recognized, of course.)
2752 PH/44 There was no timeout on the connect() call when using a Unix domain
2753 socket in the ${readsocket expansion. There now is.
2755 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2756 be meaningful with "accept".
2758 SC/01 Eximstats V1.43
2759 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2761 SC/02 Eximstats V1.44
2762 Use a glob alias rather than an array ref in the generated
2763 parser. This improves both readability and performance.
2765 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2766 Collect SpamAssassin and rejection statistics.
2767 Don't display local sender or destination tables unless
2768 there is data to show.
2769 Added average volumes into the top table text output.
2771 SC/04 Eximstats V1.46
2772 Collect data on the number of addresses (recipients)
2773 as well as the number of messages.
2775 SC/05 Eximstats V1.47
2776 Added 'Message too big' to the list of mail rejection
2777 reasons (thanks to Marco Gaiarin).
2779 SC/06 Eximstats V1.48
2780 Mainlog lines which have GMT offsets and are too short to
2781 have a flag are now skipped.
2783 SC/07 Eximstats V1.49 (Alain Williams)
2784 Added the -emptyok flag.
2786 SC/08 Eximstats V1.50
2787 Fixes for obtaining the IP address from reject messages.
2789 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2790 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2791 whitespace changes from 4.64-PH/27
2793 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2796 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2797 are found, allow negative numbers in numeric criteria)
2799 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2801 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2802 to match changes made in 4.64-PH/43
2804 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2806 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2807 "domain" is actually an IP address, and force a failure. This locks out
2808 those revolvers/nameservers that support "A-for-A" lookups, in
2809 contravention of the specifications.
2811 PH/48 When a host name was looked up from an IP address, and the subsequent
2812 forward lookup of the name timed out, the host name was left in
2813 $sender_host_name, contrary to the specification.
2815 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2816 restricted to single-key lookups, Exim was not diagnosing an error if
2817 * or *@ was used with a query-style lookup.
2819 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2821 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2822 long ago, but noone interested enough thought of it. Let's just say that
2823 the "1.1" means that there are some new functions that weren't there at
2824 some point in the past.
2826 PH/51 Error processing for expansion failure of helo_data from an smtp
2827 transport during callout processing was broken.
2829 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2830 tested/used via the -bh/-bhc/-bs options.
2832 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2833 bug, fixed in subsequent PCRE releases).
2835 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2836 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2838 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2844 SC/01 Use a glob alias rather than an array ref in eximstats generated
2845 parser. This improves both readability and performance.
2847 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2848 Don't display local sender or destination tables in eximstats unless
2849 there is data to show.
2850 Added average volumes into the eximstats top table text output.
2852 SC/03 Collect data on the number of addresses (recipients) as well
2853 as the number of messages in eximstats.
2855 TF/01 Correct an error in the documentation for the redirect router. Exim
2856 does (usually) call initgroups() when daemonizing.
2858 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2859 with consistent privilege compared to when running as a daemon.
2861 TF/03 Note in the spec that $authenticated_id is not set for local
2862 submissions from trusted users.
2864 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2865 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2867 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2868 by adding some example configuration directives to the default
2869 configuration file. A little bit of work is required to uncomment the
2870 directives and define how usernames and passwords are checked, but
2871 there is now a framework to start from.
2873 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2874 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2875 without this. I don't know how relevant this is to other LDAP libraries.
2877 PH/02 Add the verb name to the "unknown ACL verb" error.
2879 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2881 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2883 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2884 directory not expanded when it should be if an expanded home directory
2885 was set for the address (which is overridden by the transport).
2887 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2890 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2891 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2892 because it is too late at that time, and has no effect.
2894 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2895 security issue with \' (bugzilla #107). I could not use the
2896 PQescapeStringConn() function, because it needs a PGconn value as one of
2899 PH/08 When testing addresses using -bt, indicate those final addresses that
2900 are duplicates that would not cause an additional delivery. At least one
2901 person was confused, thinking that -bt output corresponded to deliveries.
2902 (Suppressing duplicates isn't a good idea as you lose the information
2903 about possibly different redirections that led to the duplicates.)
2905 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2906 systems where poll() doesn't work, in particular OS X.
2908 PH/10 Added more information to debugging output for retry time not reached.
2910 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2911 operations in malware.c.
2913 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2916 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2917 syslog_duplication set false, log lines that would normally be written
2918 both the the main log and to the reject log were not written to syslog at
2921 PH/14 In the default configuration, change the use of "message" in ACL warn
2922 statements to "add_header".
2924 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2925 not followed by a command (e.g. "seen endif").
2927 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2928 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2931 PH/17 Added extra conditions to the default value of delay_warning_condition
2935 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2936 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2937 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2940 The Auto-Submitted: and various List- headers are standardised, whereas I
2941 don't think Precedence: ever was.
2943 PH/18 Refactored debugging code in route_finduser() to show more information,
2944 in particular, the error code if getpwnam() issues one.
2946 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2947 This is apparently needed in addition to the PH/07 change above to avoid
2948 any possible encoding problems.
2950 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2951 but not after initializing Perl.
2953 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2954 output them only if debugging. By default they are written stderr,
2955 apparently, which is not desirable.
2957 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2960 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2963 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2965 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2966 authenticated or an ident call has been made. Suppress the default
2967 values for $authenticated_id and $authenticated_sender (but permit -oMai
2968 and -oMas) when testing with -bh.
2970 PH/24 Re-jigged the order of the tests in the default configuration so that the
2971 tests for valid domains and recipients precede the DNS black list and CSA
2972 tests, on the grounds that those ones are more expensive.
2974 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2975 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2976 This bug exists in every version of Exim that I still have, right back to
2979 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2980 However, an attempt to turn on case-sensitivity in a regex key by
2981 including (?-i) didn't work because the subject string was already
2982 lowercased, and the effects were non-intuitive. It turns out that a
2983 one-line patch can be used to allow (?-i) to work as expected.
2989 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2990 other effects) broke the use of negated acl sub-conditions.
2992 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2995 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2996 "Deny" causes Exim to reject the incoming connection with a 554 error.
2997 Unfortunately, if there is a major crisis, such as a disk failure,
2998 tcp-wrappers gives "deny", whereas what one would like would be some
2999 kind of temporary error. A kludge has been added to help with this.
3000 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3001 554 error is used if errno is still zero or contains ENOENT (which occurs
3002 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3005 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3007 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3008 errors. Otherwise a message that provokes a temporary error (when other
3009 messages do not) can cause a whole host to time out.
3011 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3012 addresses were routed directly to files or pipes from a redirect router.
3013 File deliveries just didn't batch; pipe deliveries might have suffered
3016 PH/06 A failure to get a lock for a hints database would erroneously always say
3017 "Failed to get write lock", even when it was really a read lock.
3019 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3020 of 0600. This has been changed to use the value of the lockfile_mode
3021 option (which defaults to 0600).
3023 PH/08 Applied small patch from the Sieve maintainer.
3025 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3026 folder from quota calculations, a direct delivery into this folder messed
3027 up the contents of the maildirsize file. This was because the regex was
3028 used only to exclude .Trash (or whatever) when the size of the mailbox
3029 was calculated. There was no check that a delivery was happening into an
3030 excluded directory. This bug has been fixed by ignoring all quota
3031 processing for deliveries into excluded directories.
3033 PH/10 Added the maildirfolder_create_regex option to appendfile.
3039 PH/01 The code for finding all the local interface addresses on a FreeBSD
3040 system running IPv6 was broken. This may well have applied to all BSD
3041 systems, as well as to others that have similar system calls. The broken
3042 code found IPv4 interfaces correctly, but gave incorrect values for the
3043 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3044 that it would not match correctly against @[] and not recognize the IPv6
3047 PH/02 The ipliteral router was not recognizing addresses of the form user@
3048 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3050 PH/03 Added disable_ipv6.
3052 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3053 decoded form, because it is most often used to construct To: headers
3054 lines in autoreplies, and the decoded form may well be syntactically
3055 invalid. However, $reply_address has leading white space removed, and all
3056 newlines turned into spaces so that the autoreply transport does not
3059 PH/05 If group was specified without a user on a router, and no group or user
3060 was specified on a transport, the group from the router was ignored.
3062 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3063 for visible compile-time settings that can be used to change these
3064 numbers, for those that want even more. Backwards compatibility with old
3065 spool files has been maintained. However, going back to a previous Exim
3066 release will lost any variables that are in spool files.
3068 PH/07 Two small changes when running in the test harness: increase delay when
3069 passing a TCP/IP connection to a new process, in case the original
3070 process has to generate a bounce, and remove special handling of
3071 127.0.0.2 (sic), which is no longer necessary.
3073 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3074 be the same on different OS.
3076 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3079 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3080 whether --show-vars was specified or not
3082 JJ/02 exipick: Added support for new ACL variable spool format introduced
3085 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3086 syntactically invalid From: or Reply-to: line, and a filter used this to
3087 generate an autoreply, and therefore failed to obtain an address for the
3088 autoreply, Exim could try to deliver to a non-existent relative file
3089 name, causing unrelated and misleading errors. What now happens is that
3090 it logs this as a hard delivery error, but does not attempt to create a
3093 PH/11 The exinext utility has a -C option for testing purposes, but although
3094 the given file was scanned by exinext itself; it wasn't being passed on
3095 when Exim was called.
3097 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3098 an end-of-file indication when reading a command response.
3100 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3101 compiled. In many other places in Exim, IPv6 addresses are always
3102 recognized, so I have changed this. It also means that IPv4 domain
3103 literals of the form [IPV4:n.n.n.n] are now always recognized.
3105 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3106 used if the router is not running as root, for example, when verifying at
3107 ACL time, or when using -bh. The debugging output from this situation was
3108 non-existent - all you got was a failure to exec. I have made two
3111 (a) Failures to set uid/gid, the current directory, or a process leader
3112 in a subprocess such as that created by queryprogram now generate
3113 suitable debugging output when -d is set.
3115 (b) The queryprogram router detects when it is not running as root,
3116 outputs suitable debugging information if -d is set, and then runs
3117 the subprocess without attempting to change uid/gid.
3119 PH/15 Minor change to Makefile for building test_host (undocumented testing
3122 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3123 additional section of a DNS packet that returns MX or SRV records.
3124 Instead, it always explicitly searches for A/AAAA records. This avoids
3125 major problems that occur when a DNS server includes only records of one
3126 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3127 fixed another bug: if SRV records were looked up and the corresponding
3128 address records were *not* found in the additional section, the port
3129 values from the SRV records were lost.
3131 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3132 using the correct key (the original address) when searching the retry
3133 rules in order to find which one to use for generating the retry hint.
3135 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3136 adding the default one. Similarly, if it contains a Reply-To: header, the
3137 errors_reply_to option, if set, is not used.
3139 PH/19 When calculating a retry time, Exim used to measure the "time since
3140 failure" by looking at the "first failed" field in the retry record. Now
3141 it does not use this if it is later than than the arrival time of the
3142 message. Instead it uses the arrival time. This makes for better
3143 behaviour in cases where some deliveries succeed, thus re-setting the
3144 "first failed" field. An example is a quota failure for a huge message
3145 when small messages continue to be delivered. Without this change, the
3146 "time since failure" will always be short, possible causing more frequent
3147 delivery attempts for the huge message than are intended.
3148 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3150 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3151 $1, $2, $3) because the numerical variables can be reset during some
3152 expansion items (e.g. "match"), thereby losing the authentication data.
3154 PH/21 Make -bV show the size of off_t variables so that the test suite can
3155 decide whether to run tests for quotas > 2G.
3157 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3158 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3159 filecount value is greater than 2G or if a quota value is greater than 2G
3160 on a system where the size of off_t is not greater than 4, a panic error
3163 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3164 never match. The debug and -bh output now contains an explicit error
3165 message indicating a malformed IPv4 address or mask.
3167 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3168 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3169 PH/23 above applies.
3171 PH/25 Do not write to syslog when running in the test harness. The only
3172 occasion when this arises is a failure to open the main or panic logs
3173 (for which there is an explicit test).
3175 PH/26 Added the /no_tell option to "control=freeze".
3177 PH/27 If a host name lookup failed very early in a connection, for example, if
3178 the IP address matched host_lookup and the reverse lookup yielded a name
3179 that did not have a forward lookup, an error message of the form "no IP
3180 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3181 could be logged. Now it outputs the IP address instead of "NULL".
3183 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3184 allows the sender and the authenticated sender to be set when
3185 submitting a message from within Exim. Since child_open_exim() is
3186 documented for local_scan(), the new function should be too.
3188 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3189 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3190 results in an empty string is now treated as unset.
3192 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3194 PH/31 Added sender_verify_fail logging option.
3196 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3197 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3198 tidied the source and removed it altogether.
3200 PH/33 When a queue run was abandoned because the load average was too high, a
3201 log line was always written; now it is written only if the queue_run log
3202 selector is set. In addition, the log line for abandonment now contains
3203 information about the queue run such as the pid. This is always present
3204 in "start" and "stop" lines but was omitted from the "abandon" line.
3206 PH/34 Omit spaces between a header name and the colon in the error message that
3207 is given when verify = headers_syntax fails (if there are lots of them,
3208 the message gets confusing).
3210 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3211 names, as there are now some PTR records that contain slashes. This check
3212 is only to protect against broken name servers that fall over on strange
3213 characters, so the fact that it applies to all lookups doesn't matter.
3215 PH/36 Now that the new test suite is complete, we can remove some of the
3216 special code in Exim that was needed for the old test suite. For example,
3217 sorting DNS records because real resolvers return them in an arbitrary
3218 order. The new test suite's fake resolver always returns records in the
3221 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3222 bounces) except when queue_only is set, to avoid logging races between
3223 the different processes.
3225 PH/38 Panic-die if .include specifies a non-absolute path.
3227 PH/39 A tweak to the "H" retry rule from its user.
3229 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3230 a label. They prevented compilation on older perls.
3232 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3233 a warning to be raised on newish perls.
3235 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3236 on queue. Changes to match documented behaviour of showing count of
3237 messages matching specified criteria.
3239 PH/40 Changed the default ident timeout from 30s to 5s.
3241 PH/41 Added support for the use of login_cap features, on those BSD systems
3242 that have them, for controlling the resources used by pipe deliveries.
3244 PH/42 The content-scanning code uses fopen() to create files in which to put
3245 message data. Previously it was not paying any attention to the mode of
3246 the files. Exim runs with umask(0) because the rest of the code creates
3247 files with open(), and sets the required mode explicitly. Thus, these
3248 files were ending up world-writeable. This was not a big issue, because,
3249 being within the spool directory, they were not world-accessible. I have
3250 created a function called modefopen, which takes an additional mode
3251 argument. It sets umask(777), creates the file, chmods it to the required
3252 mode, then resets the umask. All the relevant calls to fopen() in the
3253 content scanning code have been changed to use this function.
3255 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3256 to 24 hours. This avoids potential overflow problems when processing G
3257 and H retry rules. I suspect nobody ever tinkers with this value.
3259 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3261 PH/45 When the plaintext authenticator is running as a client, the server's
3262 challenges are checked to ensure they are valid base64 strings. By
3263 default, the authentication attempt is cancelled if an invalid string is
3264 received. Setting client_ignore_invalid_base64 true ignores these errors.
3265 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3266 they are received. Thus, the responses can be made to depend on the
3267 challenges. If an invalid string is ignored, an empty string is placed in
3270 PH/46 Messages that are created by the autoreply transport now contains a
3271 References: header, in accordance with RFCs 2822 and 3834.
3273 PH/47 Added authenticated_sender_force to the smtp transport.
3275 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3277 PH/49 Installed latest patch from the Sieve maintainer.
3279 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3280 was also set, the appendfile transport was unnecessarily scanning a
3281 directory of message files (e.g. for maildir delivery) to find the count
3282 of files (along with the size), even though it did not need this
3283 information. It now does the scan only if it needs to find either the
3284 size of the count of files.
3286 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3288 PH/52 Two bugs concerned with error handling when the smtp transport is
3291 (i) Exim was not creating retry information for temporary errors given
3292 for individual recipients after the DATA command when the smtp transport
3293 was used in LMTP mode. This meant that they could be retried too
3294 frequently, and not timed out correctly.
3296 (ii) Exim was setting the flag that allows error details to be returned
3297 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3298 recipients that were returned after the DATA command.
3300 PH/53 This is related to PH/52, but is more general: for any failing address,
3301 when detailed error information was permitted to be returned to the
3302 sender, but the error was temporary, then after the final timeout, only
3303 "retry timeout exceeded" was returned. Now it returns the full error as
3304 well as "retry timeout exceeded".
3306 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3307 do this, and (what is worse) MTAs that accept it.
3309 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3310 will now be deprecated.
3312 PH/56 New os.c-cygwin from the Cygwin maintainer.
3314 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3315 formats (previously only available in exim formats via -bpr, -bpru,
3316 and -bpra. Now also available in native and exiqgrep formats)
3318 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3319 with very large, slow to parse queues
3321 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3323 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3325 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3326 responses to authentication challenges, though it was showing the
3327 challenges; (ii) I've removed the CR characters from the debug output for
3330 PH/58 Allow for the insertion of a newline as well as a space when a string
3331 is turned into more than one encoded-word during RFC 2047 encoding. The
3332 Sieve code now uses this.
3334 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3335 data_4xx, lost_connection, tls_required.
3337 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3338 message was being sent as an SMTP response.
3340 PH/61 Add -l and -k options to exicyclog.
3342 PH/62 When verifying, if an address was redirected to one new address, so that
3343 verification continued, and the new address failed or deferred after
3344 having set something in $address_data, the value of $address_data was not
3345 passed back to the ACL. This was different to the case when no
3346 redirection occurred. The value is now passed back in both cases.
3348 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3349 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3350 use login_cap.h, so on its own it isn't the distinguishing feature. The
3351 new name refers directly to the setclassresources() function.
3353 PH/65 Added configuration files for NetBSD3.
3355 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3357 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3358 is preferred over IPv4.
3360 PH/68 The bounce_return_message and bounce_return_body options were not being
3361 honoured for bounces generated during the reception of non-SMTP messages.
3362 In particular, this applied to messages rejected by the ACL. This bug has
3363 been fixed. However, if bounce_return_message is true and bounce_return_
3364 body is false, the headers that are returned for a non-SMTP message
3365 include only those that have been read before the error was detected.
3366 (In the case of an ACL rejection, they have all been read.)
3368 PH/69 The HTML version of the specification is now built in a directory called
3369 spec_html instead of spec.html, because the latter looks like a path with
3370 a MIME-type, and this confuses some software.
3372 PH/70 Catch two compiler warnings in sieve.c.
3374 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3375 function verify_get_ident() calls ip_connect() to connect a socket, but
3376 if the "connect()" function timed out, ip_connect() used to close the
3377 socket. However, verify_get_ident() also closes the socket later, and in
3378 between Exim writes to the log, which may get opened at this point. When
3379 the socket was closed in ip_connect(), the log could get the same file
3380 descriptor number as the socket. This naturally causes chaos. The fix is
3381 not to close the socket in ip_connect(); the socket should be closed by
3382 the function that creates it. There was only one place in the code where
3383 this was missing, in the iplookup router, which I don't think anybody now
3384 uses, but I've fixed it anyway.
3386 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3387 well as to direct DNS lookups. Otherwise the handling of names in host
3388 lists is inconsistent and therefore confusing.
3394 PH/01 Two changes to the default runtime configuration:
3396 (1) Move the checks for relay_from_hosts and authenticated clients from
3397 after to before the (commented out) DNS black list checks.
3399 (2) Add control=submission to the relay_from_hosts and authenticated
3400 clients checks, on the grounds that messages accepted by these
3401 statements are most likely to be submissions.
3403 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3405 (1) Generate an error if the third argument for the ${prvs expansion is
3408 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3411 (3) Reset the variables that are obtained from the first argument of
3412 ${prvscheck and used in the second argument before leaving the code,
3413 because their memory is reclaimed, so using them afterwards may do
3416 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3417 one (it's much easier than Tom thought :-).
3419 (5) Because of (4), we can now allow for the use of $prvscheck_result
3420 inside the third argument.
3422 PH/03 For some reason, the default setting of PATH when running a command from
3423 a pipe transport was just "/usr/bin". I have changed it to
3426 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3427 anything to be listed in the output from -bV.
3429 PH/05 When a filter generated an autoreply, the entire To: header line was
3430 quoted in the delivery log line, like this:
3432 => >A.N.Other <ano@some.domain> <original@ddress> ...
3434 This has been changed so that it extracts the operative address. There
3435 may be more than one such address. If so, they are comma-separated, like
3438 => >ano@some.domain,ona@other.domain <original@ddress> ...
3440 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3441 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3442 not being looked up in the rDNS to get a host name, Exim was showing the
3443 IP address twice in Received: lines, even though the IP addresses were
3444 identical. For example:
3446 Received: from [1.2.3.4] (helo=[1.2.3.4])
3448 However, if the real host name was known, it was omitting the HELO data
3449 if it matched the actual IP address. This has been tidied up so that it
3450 doesn't show the same IP address twice.
3452 PH/07 When both +timestamp and +memory debugging was on, the value given by
3453 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3454 called by the debug printing, thereby overwriting the timestamp buffer.
3455 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3457 PH/08 When the original message was included in an autoreply transport, it
3458 always said "this is a copy of the message, including all the headers",
3459 even if body_only or headers_only was set. It now gives an appropriate
3462 PH/09 Applied a patch from the Sieve maintainer which:
3464 o fixes some comments
3465 o adds the (disabled) notify extension core
3466 o adds some debug output for the result of if/elsif tests
3467 o points to the current vacation draft in the documentation
3468 and documents the missing references header update
3472 o fixes a bug in processing the envelope test (when testing
3473 multiple envelope elements, the last element determined the
3476 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3477 Electronic Mail") by including:
3479 Auto-submitted: auto-generated
3481 in the messages that it generates (bounce messages and others, such as
3482 warnings). In the case of bounce messages for non-SMTP messages, there was
3483 also a typo: it was using "Auto_submitted" (underscore instead of
3484 hyphen). Since every message generated by Exim is necessarily in response
3485 to another message, thes have all been changed to:
3487 Auto-Submitted: auto-replied
3489 in accordance with these statements in the RFC:
3491 The auto-replied keyword:
3493 - SHOULD be used on messages sent in direct response to another
3494 message by an automatic process,
3496 - MUST NOT be used on manually-generated messages,
3498 - MAY be used on Delivery Status Notifications (DSNs) and Message
3499 Disposition Notifications (MDNs),
3501 - MUST NOT be used on messages generated by automatic or periodic
3502 processes, except for messages which are automatic responses to
3505 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3506 to the default Received: header definition.
3508 PH/12 Added log selector acl_warn_skipped (default on).
3510 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3511 variables because (a) they are in the wrong storage pool and (b) even if
3512 they were copied, it wouldn't work properly because of the caching.
3514 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3515 checking when decoding. Apparently there are clients that generate
3516 overlong encoded strings. Why am I not surprised?
3518 PH/15 If the first argument of "${if match_address" was not empty, but did not
3519 contain an "@" character, Exim crashed. Now it writes a panic log message
3520 and treats the condition as false.
3522 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3524 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3525 extension "envelope-auth". The code is finished and in agreement with
3526 other implementations, but there is no documentation so far and in fact,
3527 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3528 not changing the active code.
3530 Print executed "if" and "elsif" statements when debugging is used. This
3531 helps a great deal to understand what a filter does.
3533 Document more things not specified clearly in RFC3028. I had all this
3534 sorted out, when out of a sudden new issues came to my mind. Oops."
3536 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3539 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3540 canonical form (as documented). However, after a host name lookup from
3541 the IP address, check_host() was doing a simple string comparison with
3542 addresses acquired from the DNS when checking that the found name did
3543 have the original IP as one of its addresses. Since any found IPv6
3544 addresses are likely to be in abbreviated form, the comparison could
3545 fail. Luckily, there already exists a function for doing the comparison
3546 by converting both addresses to binary, so now that is used instead of
3547 the text comparison.
3549 PH/20 There was another similar case to PH/19, when a complete host name was
3550 given in a host list; looking up its IP address could give an abbreviated
3551 form, whereas the current host's name might or might not be abbreviated.
3552 The same fix has been applied.
3558 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3559 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3562 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3563 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3565 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3567 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3568 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3569 Also, for "Auto-Submitted", treat anything other than "no" as
3570 non-personal, in accordance with RFC 3834. (Previously it treated
3571 anything starting "auto-" as non-personal.)
3573 TF/01 The control=submission/name=... option had a problem with syntax
3574 errors if the name included a slash character. The /name= option
3575 now slurps the rest of the string, so it can include any characters
3576 but it must come last in the list of options (after /sender_retain
3579 PH/05 Some modifications to the interface to the fake nameserver for the new
3587 TK/01 Added the "success_on_redirect" address verification option. See
3588 NewStuff for rationale and an example.
3590 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3592 PH/02 Patch to exigrep to allow it to work on syslog lines.
3594 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3595 fread() to skip over the body file's header line, because in Cygwin the
3596 header line is locked and is inaccessible.
3598 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3599 co-exist for some time) to make it clear that it is the Exim ID that is
3600 referenced, not the Message-ID: header line.
3602 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3603 string_format() function, because snprintf() does not exist on all
3606 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3607 ${stat: expansion item.
3609 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3610 protocol synchronization error", to keep the pedants happy.
3612 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3613 well as for IRIX systems, when gcc is being used. See the host.c source
3616 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3618 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3621 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3622 transport and to the smtp transport in LMTP mode.
3624 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3626 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3627 run a filter in a subprocess. This could lead to confusion in subsequent
3628 lookups in the parent process. There should also be a search_tidyup() at
3629 the end of the subprocess.
3631 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3632 only if the host matched helo_try_verify_hosts, which caused the
3633 verification to occur when the EHLO/HELO command was issued. The ACL just
3634 tested the remembered result. Now, if a previous verification attempt has
3635 not happened, "verify = helo" does it there and then.
3637 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3639 TK/03 Fix log output including CR from clamd.
3641 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3642 exist provoked a memory error which could cause a segfault.
3644 PH/15 Installed PCRE 6.2
3646 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3648 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3649 of the problem. Specifically, suggested +O2 rather than +O1 for the
3652 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3654 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3655 record, the port was not correctly logged when the outgoing_port log
3656 selector was set (it logged the transort's default port).
3658 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3659 fallback_hosts, and "hosts" in the smtp transport.
3661 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3662 host errors such as "Connection refused".
3664 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3665 authentication with radiusclient 0.4.9:
3667 - Error returned from rc_read_config was caught wrongly
3668 - Username/password not passed on to radius server due to wrong length.
3670 The presumption is that some radiusclient API changes for 4.51/PH/17
3671 were not taken care of correctly. The code is still untested by me (my
3672 Linux distribution still has 0.3.2 of radiusclient), but it was
3673 contributed by a Radius user.
3675 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3676 expanding the "port" option of the smtp transport.
3678 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3679 while reading a MIME header. Thanks to Tom Hughes for a patch.
3681 PH/24 Include config.h inside local_scan.h so that configuration settings are
3684 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3685 that in an ACL for RCPT (for example), you can examine exactly what was
3688 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3689 commands, but it was not correctly comparing the address with the actual
3690 client host address. Thus, it would show the EHLO address in Received:
3691 header lines when this was not necessary.
3693 PH/27 Added the % operator to ${eval:}.
3695 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3696 it should be ignoring failures (because with -C, for example, it has lost
3697 privilege). It wasn't ignoring creation failures other than "already
3700 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3703 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3704 because an input line was too long, either on its own, or by virtue of
3705 too many continuations, the temporary file was not being removed, and the
3706 return code was incorrect.
3708 PH/31 Missing "BOOL" in function definition in filtertest.c.
3710 PH/32 Applied Sieve patches from the maintainer.
3712 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3714 PH/33 Added "verify = not_blind".
3716 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3717 Local/Makefile (with some defaults set). These are used in built scripts
3718 such as exicyclog, but they have never been used in the exim_install
3719 script (though there are many overriding facilities there). I have
3720 arranged that the exim_install script now takes note of these two
3723 PH/35 Installed configuration files for Dragonfly.
3725 PH/36 When a locally submitted message by a trusted user did not contain a
3726 From: header, and the sender address was obtained from -f or from an SMTP
3727 MAIL command, and the trusted user did not use -F to supply a sender
3728 name, $originator_name was incorrectly used when constructing a From:
3729 header. Furthermore, $originator_name was used for submission mode
3730 messages from external hosts without From: headers in a similar way,
3731 which is clearly wrong.
3733 PH/37 Added control=suppress_local_fixups.
3735 PH/38 When log_selector = +received_sender was set, and the addition of the
3736 sender made the log line's construction buffer exactly full, or one byte
3737 less than full, an overflow happened when the terminating "\n" was
3740 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3741 when the result of a list match is failure because a DNS lookup failed.
3743 PH/40 RM_COMMAND is now used in the building process.
3745 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3746 the "build-* directories that it finds.
3748 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3749 address in a domain literal was a prefix of an interface address.
3751 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3752 when verifying a sender address, unless rewrite_headers is false.
3754 PH/44 Wrote a long comment about why errors_to addresses are verified as
3755 recipients, not senders.
3757 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3758 the ratelimit ACL was added.
3760 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3762 PH/46 Added extra information about PostgreSQL errors to the error string.
3764 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3765 avoiding the need to install special zones in a real server. This is
3766 backwards compatible; if it can't find the fake resolver, it drops back.
3767 Thus, both old and new test suites can be run.
3769 TF/02 Added util/ratelimit.pl
3771 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3772 clock is set back in time.
3774 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3775 Candler <B.Candler@pobox.com>.
3777 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3778 OK for addresses that are the result of redirections.
3780 PH/48 A number of further additions for the benefit of the new test suite,
3781 including a fake gethostbyname() that interfaces to the fake DNS resolver
3784 TF/06 The fix for widen_domains has also been applied to qualify_single and
3785 search_parents which are the other dnslookup options that can cause
3788 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3791 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3793 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3794 service" if it encountered an MX record with an empty target hostname.
3795 The message is now "an MX or SRV record indicated no SMTP service".
3797 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3798 if the DNS of the sending site is misconfigured. This is quite a
3799 common situation. This change restores the behaviour of treating a
3800 helo verification defer as a failure.
3802 PH/51 If self=fail was set on a router, the bounce message did not include the
3803 actual error message.
3809 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3811 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3812 command itself ended in error, the underlying message about the transport
3813 filter timeout was being overwritten with the pipe command error. Now the
3814 underlying error message should be appended to the second error message.
3816 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3818 PH/02 Reduce the amount of output that "make" produces by default. Full output
3819 can still be requested.
3821 PH/03 The warning log line about a condition test deferring for a "warn" verb
3822 was being output only once per connection, rather than after each
3823 occurrence (because it was using the same function as for successful
3824 "warn" verbs). This seems wrong, so I have changed it.
3826 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3827 it should not have, which might have caused a crash in the right
3828 circumstances, but probably never did.
3830 PH/04 Installed a modified version of Tony Finch's patch to make submission
3831 mode fix the return path as well as the Sender: header line, and to
3832 add a /name= option so that you can make the user's friendly name appear
3835 TF/03 Added the control = fakedefer ACL modifier.
3837 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3838 Mark Lowes for thorough testing.
3840 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3842 TK/03 Merged latest SRS patch from Miles Wilton.
3844 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3845 with the definition in sysexits.h (which is #included earlier).
3846 Fortunately, Exim does not actually use EX_OK. The code used to try to
3847 preserve the sysexits.h value, by assuming that macro definitions were
3848 scanned for macro replacements. I have been disabused of this notion,
3849 so now the code just undefines EX_OK before #including unistd.h.
3851 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3852 in the smtp transport. When a block could not be written in a single
3853 write() function, the timeout was being re-applied to each part-write.
3854 This seems wrong - if the receiver was accepting one byte at a time it
3855 would take for ever. The timeout is now adjusted when this happens. It
3856 doesn't have to be particularly precise.
3858 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3859 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3861 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3862 without a domain if the check to <postmaster@domain> fails.
3864 SC/01 Eximstats: added -xls and the ability to specify output files
3865 (patch written by Frank Heydlauf).
3867 SC/02 Eximstats: use FileHandles for outputting results.
3869 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3871 SC/04 Eximstats: fixed display of large numbers with -nvr option
3873 SC/05 Eximstats: fixed merging of reports with empty tables.
3875 SC/06 Eximstats: added the -include_original_destination flag
3877 SC/07 Eximstats: removed tabs and trailing whitespace.
3879 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3881 TK/06 MBOX spool code: Add real "From " MBOX separator line
3882 so the .eml file is really in mbox format (even though
3883 most programs do not really care). Patch from Alex Miller.
3885 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3886 The latter is generated from $received_to and is only set if the
3887 message has one envelope recipient. SA can use these headers,
3888 obviously out-of-the-box. Patch from Alex Miller.
3890 PH/08 The ${def test on a variable was returning false if the variable's
3891 value was "0", contrary to what the specification has always said!
3892 The result should be true unless the variable is empty.
3894 PH/09 The syntax error of a character other than { following "${if
3895 def:variable_name" (after optional whitespace) was not being diagnosed.
3896 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3897 accidental colon was present, for example, could give incorrect results.
3899 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3900 result is used (not including appendfile, where other changes are about
3903 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3904 This involved changing a lot of size variables from int to off_t. It
3905 should work with maildirs and everything.
3907 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3908 spamd dying while we are connected to it.
3910 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3913 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3914 function for BDB 4.3.
3916 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3918 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3919 the sources for PCRE have been re-arranged and more files are now
3922 PH/15 The code I had for printing potentially long long variables in PH/11
3923 above was not the best (it lost precision). The length of off_t variables
3924 is now inspected at build time, and an appropriate printing format (%ld
3925 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3926 to be "long long int" or "long int". This is needed for the internal
3927 formatting function string_vformat().
3929 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3930 the configuration file to be ":syslog", then the script "guesses" where
3931 the logs files are, rather than using the compiled in default. In our
3932 case the guess is not the same as the compiled default, so the script
3933 suddenly stopped working when I started to use syslog. The patch checks
3934 to see if log_file_path is "". If so, it attempts to read it from exim
3935 with no configuration file to get the compiled in version, before it
3936 falls back to the previous guessing code."
3938 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3939 implementing BATV in an Exim configuration. See NewStuff for the gory
3942 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3943 Makefile that are specific to HP-UX.
3945 PH/18 If the "use_postmaster" option was set for a recipient callout together
3946 with the "random" option, the postmaster address was used as the MAIL
3947 FROM address for the random test, but not for the subsequent recipient
3948 test. It is now used for both.
3950 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3951 patch removes a few documentation additions to RFC 3028, because the
3952 latest draft now contains them. It adds the new en;ascii-case comparator
3953 and a new error check for 8bit text in MIME parts. Comparator and
3954 require names are now matched exactly. I enabled the subaddress
3955 extension, but it is not well tested yet (read: it works for me)."
3957 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3958 rework some of the code of TK/09 above to avoid the hardwired use of
3959 "%lld" and "long long". Replaced the call to snprintf() with a call to
3962 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3963 records point to non-existent hosts", "retry timeout exceeded", and
3964 "retry time not reached for any host after a long failure period".
3966 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3967 experimental DomainKeys support:
3969 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3970 (2) On an error such as an illegally used "control", the wrong name for
3971 the control was given.
3973 These problems did NOT occur unless DomainKeys support was compiled.
3975 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3977 PH/24 Added ${if match_ip condition.
3979 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3980 databases so that it will be absolutely obvious if a crash occurs in the
3981 DB library. This is a regular occurrence (often caused by mis-matched
3984 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3985 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3986 picked up on a user's system that detects such things. There doesn't seem
3987 to be a gcc warning option for this - only an attribute that has to be
3988 put on the function's prototype. It seems that in Fedora Core 4 they have
3989 set this on a number of new functions. No doubt there will be more in due
3992 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3993 specify a transport. However, if an address that was verified by such a
3994 router was the subject of a callout, Exim crashed because it tried to
3995 read the rcpt_include_affixes from the non-existent transport. Now it
3996 just assumes that the setting of that option is false. This bug was
3997 introduced by 4.51/PH/31.
3999 PH/28 Changed -d+all to exclude +memory, because that information is very
4000 rarely of interest, but it makes the output a lot bigger. People tend to
4001 do -d+all out of habit.
4003 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4004 code in os-type was giving problems when libc.so lives in lib64, like on
4007 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4008 aren't the modern standard, and it seems that some systems' include files
4009 don't always have them. Exim was already checking for some of the newer
4010 ones like T_AAAA, and defining it itself. I've added checks for all the
4011 record types that Exim uses.
4013 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4014 not automatically generating a new one, as it is supposed to. This
4015 prevented TLS from working. If the file did exist, but contained invalid
4016 data, a new version was generated, as expected. It was only the case of a
4017 non-existent file that was broken.
4019 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4020 with a change in libdomainkeys > 0.64.
4022 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4023 from DNS. If the selector record carries the flag, it now has
4024 precedence over the domain-wide flag.
4026 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4028 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4029 the use of a port name that isn't defined in /etc/services) occurred, the
4030 message was deferred as in a normal delivery, and thus remained on the
4031 spool, instead of being failed because of the mua_wrapper setting. This
4032 is now fixed, and I tidied up some of the mua_wrapper messages at the
4035 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4036 the messages in a hash of arrays rather than using individual hashes.
4037 This is a bit cleaner and results in dramatic memory savings, albeit
4038 at a slight CPU cost.
4040 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4041 as requested by Marc Sherman.
4043 SC/10 Eximstats: added histograms for user specified patterns as requested
4046 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4048 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4049 fopen() in the content-scanning modules that did not already have it.
4055 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4056 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4058 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4060 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4062 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4063 to test Sieve filters that use "vacation".
4065 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4066 that changes the way the GnuTLS parameters are stored in the cache file.
4067 The new format can be generated externally. For backward compatibility,
4068 if the data in the cache doesn't make sense, Exim assumes it has read an
4069 old-format file, and it generates new data and writes a new file. This
4070 means that you can't go back to an older release without removing the
4073 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4074 work if there are any delivery delays because "one_time" forces the
4075 parent to be marked "delivered", so its unseen clone is never tried
4076 again. For this reason, Exim now forbids the simultaneous setting of
4079 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4080 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4081 entry if you want to know the details. The fix, however, neglected to
4082 consider the case when local delivery batching is involved. The test for
4083 "previously delivered" was not happening when checking to see if an
4084 address could be batched with a previous (undelivered) one; under
4085 certain circumstances this could lead to multiple deliveries to the same
4088 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4089 in its include files, and this causes problems building Exim.
4091 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4092 header_syntax) but Exim was just ignoring anything given after a slash.
4093 In particular, this caused confusion with an attempt to use "verify =
4094 reverse_host_lookup/defer_ok". An error is now given when options are
4095 supplied for verify items that do not have them. (Maybe reverse_host_
4096 lookup should have a defer_ok option, but that's a different point.)
4098 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4099 defined by RFC 821) to 2048, because there were problems with some AUTH
4100 commands, and RFC 1869 says the size should be increased for extended
4101 SMTP commands that take arguments.
4103 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4106 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4107 "unknown" error; now it says that the functionality isn't in the binary.
4109 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4110 an address' error message when a string expansion fails (syntax or
4111 whatever). Otherwise the password may appear in the log. Following change
4112 PH/42 below, there is no longer a chance of it appearing in a bounce
4115 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4117 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4118 of its addresses was ever tried. (Bugzilla bug #2.)
4120 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4121 the result incorrectly in the debug output. (It correctly added a newline
4122 to what was transported.)
4124 TF/01 Added $received_time.
4126 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4127 commented out examples of how to interface to a virus scanner and to
4128 SpamAssassin. Also added commented examples of av_scanner and
4129 spamd_address settings.
4131 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4132 and controls are allowed in which ACLs. There were a couple of minor
4133 errors. Some of the entries in the conditions table (which is a table of
4134 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4135 negation of where the condition IS allowed.
4137 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4139 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4140 header file does not have a version number, so I've had to invent a new
4141 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4142 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4143 radiusclient), but it was contributed by a Radius user.
4145 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4146 files or directories, for OpenSSL.
4148 PH/19 When an Exim process that is running as root has to create an Exim log
4149 file, it does so in a subprocess that runs as exim:exim so as to get the
4150 ownership right at creation (otherwise, other Exim processes might see
4151 the file with the wrong ownership). There was no test for failure of this
4152 fork() call, which would lead to the process getting stuck as it waited
4153 for a non-existent subprocess. Forks do occasionally fail when resources
4154 run out. I reviewed all the other calls to fork(); they all seem to check
4157 PH/20 When checking for unexpected SMTP input at connect time (before writing
4158 the banner), Exim was not dealing correctly with a non-positive return
4159 from the read() function. If the client had disconnected by this time,
4160 the result was a log entry for a synchronization error with an empty
4161 string after "input=" when read() returned zero. If read() returned -1
4162 (an event I could not check), uninitialized data bytes were printed.
4163 There were reports of junk text (parts of files, etc) appearing after
4166 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4168 PH/22 Added support for macro redefinition, and (re)definition in between
4169 driver and ACL definitions.
4171 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4172 forgetting to use the resulting value; it was using the unexpanded value.
4174 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4175 hadn't been configured. The fix is from Juergen Kreileder, who
4176 understands it better than I do:
4178 "Here's what I see happening with three configured cyrus_sasl
4179 authenticators configured (plain, login, cram-md5):
4181 On startup auth_cyrus_sasl_init() gets called for each of these.
4182 This means three calls to sasl_listmech() without a specified mech_list.
4183 => SASL tests which mechs of all available mechs actually work
4184 => three warnings about OTP not working
4185 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4187 With the patch, sasl_listmech() also gets called three times. But now
4188 SASL's mech_list option is set to the server_mech specified in the the
4189 authenticator. Or in other words, the answer from sasl_listmech()
4190 gets limited to just the mech you're testing for (which is different
4192 => the return list contains just 'plain' or 'login', 'cram-md5' or
4193 nothing depending on the value of ob->server_mech.
4195 I've just tested the patch: Authentication still works fine,
4196 unavailable mechs specified in the exim configuration are still
4197 caught, and the auth.log warnings about OTP are gone."
4199 PH/25 When debugging is enabled, the contents of the command line are added
4200 to the debugging output, even when log_selector=+arguments is not
4203 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4204 answer is "GNU", and only if the return is "GNU/something" is the answer
4207 PH/27 $acl_verify_message is now set immediately after the failure of a
4208 verification in an ACL, and so is available in subsequent modifiers. In
4209 particular, the message can be preserved by coding like this:
4211 warn !verify = sender
4212 set acl_m0 = $acl_verify_message
4214 Previously, $acl_verify_message was set only while expanding "message"
4215 and "log_message" when a very denied access.
4217 PH/28 Modified OS/os.c-Linux with
4219 -#ifndef OS_LOAD_AVERAGE
4220 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4222 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4223 nomenclature these days.)
4225 PH/29 Installed patch from the Sieve maintainer that adds the options
4226 sieve_useraddress and sieve_subaddress to the redirect router.
4228 PH/30 In these circumstances:
4229 . Two addresses routed to the same list of hosts;
4230 . First host does not offer TLS;
4231 . First host accepts first address;
4232 . First host gives temporary error to second address;
4233 . Second host offers TLS and a TLS session is established;
4234 . Second host accepts second address.
4235 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4236 and peerdn, if requested) that were in fact used only for the second
4239 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4240 attention to any local part prefix or suffix that was matched by the
4241 router that accepted the address. It now behaves in the same way as it
4242 does for delivery: the affixes are removed from the local part unless
4243 rcpt_include_affixes is set on the transport.
4245 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4246 timeout during the DATA phase of an incoming message.
4248 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4249 applied a patch sanctioned by the Sieve maintainer.
4251 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4252 the uid or gid is negative. A case of a negative gid caused this to be
4253 noticed. The fix allows for either to be negative.
4255 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4256 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4259 PH/36 The change PH/12 above was broken. Fixed it.
4261 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4262 the grounds that routing the same address twice would always produce the
4263 same answer. This might have been true once, but it is certainly no
4264 longer true now. Routing a child address may depend on the previous
4265 routing that produced that child. Some complicated redirection strategies
4266 went wrong when messages had multiple recipients, and made Exim's
4267 behaviour dependent on the order in which the addresses were given.
4269 I have moved the duplicate checking until after the routing is complete.
4270 Exim scans the addresses that are assigned to local and remote
4271 transports, and removes any duplicates. This means that more work will be
4272 done, as duplicates will always all be routed, but duplicates are
4273 presumably rare, so I don't expect this is of any significance.
4275 For deliveries to pipes, files, and autoreplies, the duplicate checking
4276 still happens during the routing process, since they are not going to be
4279 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4280 It corrects a timeout issue with spamd. This is Ian's comment: "The
4281 background is that sometimes spamd either never reads data from a
4282 connection it has accepted, or it never writes response data. The exiscan
4283 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4284 blindly assumes that writes won't block so it may never time out."
4286 PH/39 Allow G after quota size as well as K and M.
4288 PH/40 The value set for $authenticated_id in an authenticator may not contain
4289 binary zeroes or newlines because the value is written to log lines and
4290 to spool files. There was no check on this. Now the value is run through
4291 the string_printing() function so that such characters are converted to
4292 printable escape sequences.
4294 PH/41 $message_linecount is a new variable that contains the total number of
4295 lines in the message. Compare $body_linecount, which is the count for the
4298 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4299 bounce and delay warning messages, except in certain special cases, which
4302 (a) An SMTP error message from a remote host;
4303 (b) A message specified in a :fail: redirection;
4304 (c) A message specified in a "fail" command in a system filter;
4305 (d) A message specified in a FAIL return from the queryprogram router;
4306 (e) A message specified by the cannot_route_message router option.
4308 In these cases only, Exim does include the error details in bounce and
4309 warning messages. There are also a few cases where bland messages such
4310 as "unrouteable address" or "local delivery error" are given.
4312 PH/43 $value is now also set for the "else" part of a ${run expansion.
4314 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4315 being worked on, but at least Exim now implements the latest version to
4318 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4319 process to complete was treated as a delivery failure, a timeout while
4320 writing the message to the pipe was logged, but erroneously treated as a
4321 successful delivery. Such timeouts include transport filter timeouts. For
4322 consistency with the overall process timeout, these timeouts are now
4323 treated as errors, giving rise to delivery failures by default. However,
4324 there is now a new Boolean option for the pipe transport called
4325 timeout_defer, which, if set TRUE, converts the failures into defers for
4326 both kinds of timeout. A transport filter timeout is now identified in
4329 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4330 systems where "make" and "gmake" are different, calling "gmake" at top
4331 level broke things. I've arranged for the value of $(MAKE) to be passed
4332 from the Makefile to this script so that it can call the same version of
4336 A note about Exim versions 4.44 and 4.50
4337 ----------------------------------------
4339 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4340 changes of various kinds. As a consequence, a big documentation update was
4341 needed. This delayed the release for rather longer than seemed good, especially
4342 in the light of a couple of (minor) security issues. Therefore, the changes
4343 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4344 release. So 4.44 and 4.50 are in effect two different branches that both start
4347 I have left the 4.50 change log unchanged; it contains all the changes since
4348 4.43. The change log for 4.44 is below; many of its items are identical to
4349 those for 4.50. This seems to be the most sensible way to preserve the
4350 historical information.
4356 1. Minor wording change to the doc/README.SIEVE file.
4358 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4359 computation of the current number of files was incorrect.
4361 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4362 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4365 4. Give more explanation in the error message when the command for a transport
4366 filter fails to execute.
4368 5. There are several places where Exim runs a non-Exim command in a
4369 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4370 was being done only for the command run by the queryprogram router. It is
4371 now done for all such subprocesses. The other cases are: ${run, transport
4372 filters, and the commands run by the lmtp and pipe transports.
4374 6. Added CONFIGURE_GROUP build-time option.
4376 7. Some older OS have a limit of 256 on the maximum number of file
4377 descriptors. Exim was using setrlimit() to set 1000 as a large value
4378 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4379 systems. I've change it so that if it can't get 1000, it tries for 256.
4381 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4382 was an oversight, and furthermore, ever since the addition of extra
4383 controls (e.g. 4.43/32), the checks on when to allow different forms of
4384 "control" were broken. There should now be diagnostics for all cases when a
4385 control that does not make sense is encountered.
4387 9. Added the /retain_sender option to "control=submission".
4389 10. $recipients is now available in the predata ACL (oversight).
4391 11. Tidy the search cache before the fork to do a delivery from a message
4392 received from the command line. Otherwise the child will trigger a lookup
4393 failure and thereby defer the delivery if it tries to use (for example) a
4394 cached ldap connection that the parent has called unbind on.
4396 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4397 of $address_data from the recipient verification was clobbered by the
4398 sender verification.
4400 13. The value of address_data from a sender verification is now available in
4401 $sender_address_data in subsequent conditions in the ACL statement.
4403 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4405 15. Added a new option "connect=<time>" to callout options, to set a different
4408 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4409 was its contents. (It was OK if the option was not defined at all.)
4411 17. A "Completed" log line is now written for messages that are removed from
4412 the spool by the -Mrm option.
4414 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4415 information about exactly what failed.
4417 19. Added -dd to debug only the daemon process.
4419 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4420 handles timeouts, both on the server side and network timeouts. Renamed the
4421 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4423 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4424 It is now set to "smtps".
4426 22. $host_address is now set to the target address during the checking of
4427 ignore_target_hosts.
4429 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4430 being passed; this would have caused $sender_host_name to have been used if
4431 matching the list had actually called for a host name (not very likely,
4432 since this list is usually IP addresses). A host name is now passed as
4435 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4436 code that specifies a non-restarting handler (typically sigaction() in
4437 modern systems) in an attempt to fix a rare and obscure crash bug.
4439 25. Narrowed the window for a race in the daemon that could cause it to ignore
4440 SIGCHLD signals. This is not a major problem, because they are used only to
4441 wake it up if nothing else does.
4443 26. A malformed maildirsize file could cause Exim to calculate negative values
4444 for the mailbox size or file count. Odd effects could occur as a result.
4445 The maildirsize information is now recalculated if the size or filecount
4448 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4449 support for a long time. Removed HAVE_SYS_VFS_H.
4451 28. Installed the latest version of exipick from John Jetmore.
4453 29. In an address list, if the pattern was not a regular expression, an empty
4454 subject address (from a bounce message) matched only if the pattern was an
4455 empty string. Non-empty patterns were not even tested. This was the wrong
4456 because it is perfectly reasonable to use an empty address as part of a
4457 database query. An empty address is now tested by patterns that are
4458 lookups. However, all the other forms of pattern expect the subject to
4459 contain a local part and a domain, and therefore, for them, an empty
4460 address still always fails if the pattern is not itself empty.
4462 30. Exim went into a mad DNS loop when attempting to do a callout where the
4463 host was specified on an smtp transport, and looking it up yielded more
4464 than one IP address.
4466 31. Re-factored the code for checking spool and log partition space into a
4467 function that finds that data and another that does the check. The former
4468 is then used to implement four new variables: $spool_space, $log_space,
4469 $spool_inodes, and $log_inodes.
4471 32. The RFC2047 encoding function was originally intended for short strings
4472 such as real names; it was not keeping to the 75-character limit for
4473 encoded words that the RFC imposes. It now respects the limit, and
4474 generates multiple encoded words if necessary. To be on the safe side, I
4475 have increased the buffer size for the ${rfc2047: expansion operator from
4478 33. It is now permitted to omit both strings after an "if" condition; if the
4479 condition is true, the result is "true". As before, when the second string
4480 is omitted, a false condition yields an empty string. This makes it less
4481 cumbersome to write custom ACL and router conditions.
4483 34. Failure to deliver a bounce message always caused it to be frozen, even if
4484 there was an errors_to setting on the router. The errors_to setting is now
4487 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4488 canonical form (fully expanded) before being placed in
4489 $sender_host_address.
4491 36. The table in the code that translates DNS record types into text (T_A to
4492 "A" for instance) was missing entries for NS and CNAME. It is just possible
4493 that this could have caused confusion if both these types were looked up
4494 for the same domain, because the text type is used as part of Exim's
4495 per-process caching. But the chance of anyone hitting this buglet seems
4498 37. The dnsdb lookup has been extended in a number of ways.
4500 (1) There is a new type, "zns", which walks up the domain tree until it
4501 finds some nameserver records. It should be used with care.
4503 (2) There is a new type, "mxh", which is like "mx" except that it returns
4504 just the host names, not the priorities.
4506 (3) It is now possible to give a list of domains (or IP addresses) to be
4507 looked up. The behaviour when one of the lookups defers can be
4508 controlled by a keyword.
4510 (4) It is now possible to specify the separator character for use when
4511 multiple records are returned.
4513 38. The dnslists ACL condition has been extended: it is now possible to supply
4514 a list of IP addresses and/or domains to be looked up in a particular DNS
4517 39. Added log_selector=+queue_time_overall.
4519 40. When running the queue in the test harness, wait just a tad after forking a
4520 delivery process, to get repeatability of debugging output.
4522 41. Include certificate and key file names in error message when GnuTLS fails
4523 to set them up, because the GnuTLS error message doesn't include the name
4524 of the failing file when there is a problem reading it.
4526 42. Allow both -bf and -bF in the same test run.
4528 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4530 44. The "Exiscan patch" is now merged into the mainline Exim source.
4532 45. Sometimes the final signoff response after QUIT could fail to get
4533 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4534 < 0 before doing a fflush(). This bug looks as though it goes back to the
4535 introduction of TLS in release 3.20, but "sometimes" must have been rare
4536 because the tests only now provoked it.
4538 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4539 (this can affect the format of dates).
4541 47. exim_tidydb, when checking for the continued existence of a message for
4542 which it has found a message-specific retry record, was not finding
4543 messages that were in split spool directories. Consequently, it was
4544 deleting retry records that should have stayed in existence.
4546 48. Steve fixed some bugs in eximstats.
4548 49. The SPA authentication driver was not abandoning authentication and moving
4549 on to the next authenticator when an expansion was forced to fail,
4550 contradicting the general specification for all authenticators. Instead it
4551 was generating a temporary error. It now behaves as specified.
4553 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4554 (the order specifies the preference for clients). The order is now AES256,
4555 AES128, 3DES, ARCFOUR128.
4557 51. Small patch to Sieve code - explicitly set From: when generating an
4560 52. Exim crashed if a remote delivery caused a very long error message to be
4561 recorded - for instance if somebody sent an entire SpamAssassin report back
4562 as a large number of 550 error lines. This bug was coincidentally fixed by
4563 increasing the size of one of Exim's internal buffers (big_buffer) that
4564 happened as part of the Exiscan merge. However, to be on the safe side, I
4565 have made the code more robust (and fixed the comments that describe what
4568 53. Now that there can be additional text after "Completed" in log lines (if
4569 the queue_time_overall log selector is set), a one-byte patch to exigrep
4570 was needed to allow it to recognize "Completed" as not the last thing in
4573 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4574 patch that reportedly fixes this has been added. I am not expert enough to
4575 create a test for it. This is what the patch creator wrote:
4577 "I found a little strange behaviour of ldap code when working with
4578 Windows 2003 AD Domain, where users was placed in more than one
4579 Organization Units. When I tried to give exim partial DN, the exit code
4580 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4581 But simultaneously result of request was absolutely normal ldap result,
4582 so I produce this patch..."
4584 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4585 REFERENCE, so I have modified the code to exclude the patch when that macro
4588 55. Some experimental protocols are using DNS PTR records for new purposes. The
4589 keys for these records are domain names, not reversed IP addresses. The
4590 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4591 leaves it alone. Component reversal etc. now happens only for IP addresses.
4594 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4596 57. Double the size of the debug message buffer (to 2048) so that more of very
4597 long debug lines gets shown.
4599 58. The exicyclog utility now does better if the number of log files to keep
4600 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4602 59. Two changes related to the smtp_active_hostname option:
4604 (1) $smtp_active_hostname is now available as a variable.
4605 (2) The default for smtp_banner uses $smtp_active_hostname instead
4606 of $primary_hostname.
4608 60. The host_aton() function is supposed to be passed a string that is known
4609 to be a valid IP address. However, in the case of IPv6 addresses, it was
4610 not checking this. This is a hostage to fortune. Exim now panics and dies
4611 if the condition is not met. A case was found where this could be provoked
4612 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4613 components; fortuitously, this particular loophole had already been fixed
4614 by change 4.50/55 above.
4616 If there are any other similar loopholes, the new check in host_aton()
4617 itself should stop them being exploited. The report I received stated that
4618 data on the command line could provoke the exploit when Exim was running as
4619 exim, but did not say which command line option was involved. All I could
4620 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4621 running as the user.
4624 61. There was a buffer overflow vulnerability in the SPA authentication code
4625 (which came originally from the Samba project). I have added a test to the
4626 spa_base64_to_bits() function which I hope fixes it.
4629 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4630 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4632 63. The daemon start-up calls getloadavg() while still root for those OS that
4633 need the first call to be done as root, but it missed one case: when
4634 deliver_queue_load_max is set with deliver_drop_privilege. This is
4635 necessary for the benefit of the queue runner, because there is no re-exec
4636 when deliver_drop_privilege is set.
4638 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4639 This has been fixed.
4641 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4642 was in use, was not putting the data itself into the right store pool;
4643 consequently, it could be overwritten for a subsequent message in the same
4644 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4647 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4649 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4650 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4651 the function were treating the return as a boolean value, which happened to
4652 work because 0=false and not-0=true, but is not correct code.
4654 68. The host_aton() function was not handling scoped IPv6 addresses (those
4655 with, for example, "%eth0" on the end) correctly.
4657 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4658 negated items (that is, ~something) in unsigned ints. Some compilers
4659 apparently mutter when there is no cast.
4661 70. If an address verification called from an ACL failed, and did not produce a
4662 user-specific message (i.e. there was only a "system" message), nothing was
4663 put in $acl_verify_message. In this situation, it now puts the system
4666 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4667 session; change 4.31/43 added the unwanted input to the log line - except
4668 that it did not do this in the start of session case. It now does.
4670 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4671 This is wrong and can cause the other end to generate a synchronization
4672 error if it is another Exim or anything else that does the synchronization
4673 check. A QUIT command is no longer sent after a timeout.
4675 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4676 during host lookups.
4678 74. The defer_ok option of callout verification was not working if it was used
4679 when verifying addresses in header lines, that is, for this case:
4681 verify = header_sender/callout=defer_ok
4683 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4684 those file descriptors could be used for SMTP connections. If anything
4685 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4686 could be sent to the SMTP client, causing chaos. The daemon now opens
4687 stdin, stdout, and stderr to /dev/null when it puts itself into the
4690 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4691 log by default. The user can override this with suitable Perl magic.
4693 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4694 the log message when discard triggers, was not working for the DATA ACL or
4695 for the non-SMTP ACL.
4697 78. Error message wording change in sieve.c.
4699 79. If smtp_accept_max_per_host was set, the number of connections could be
4700 restricted to fewer than expected, because the daemon was trying to set up
4701 a new connection before checking whether the processes handling previous
4702 connections had finished. The check for completed processes is now done
4703 earlier. On busy systems, this bug wouldn't be noticed because something
4704 else would have woken the daemon, and it would have reaped the completed
4707 80. If a message was submitted locally by a user whose login name contained one
4708 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4709 It caused a spool format error. I have fixed the spool reading code. A
4710 related problem was that the "from" clause in the Received: line became
4711 illegal because of the space(s). It is now covered by ${quote_local_part.
4713 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4716 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4718 83. Patch from Sieve maintainer to fix unterminated string problem in
4719 "vacation" handling.
4721 84. Some minor changes to the Linux configuration files to help with other
4722 OS variants using glibc.
4724 85. One more patch for Sieve to update vacation handling to latest spec.
4727 ----------------------------------------------------
4728 See the note above about the 4.44 and 4.50 releases.
4729 ----------------------------------------------------
4735 1. Change 4.43/35 introduced a bug that caused file counts to be
4736 incorrectly computed when quota_filecount was set in an appendfile
4739 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4740 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4743 3. Give more explanation in the error message when the command for a transport
4744 filter fails to execute.
4746 4. There are several places where Exim runs a non-Exim command in a
4747 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4748 was being done only for the command run by the queryprogram router. It is
4749 now done for all such subprocesses. The other cases are: ${run, transport
4750 filters, and the commands run by the lmtp and pipe transports.
4752 5. Some older OS have a limit of 256 on the maximum number of file
4753 descriptors. Exim was using setrlimit() to set 1000 as a large value
4754 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4755 systems. I've change it so that if it can't get 1000, it tries for 256.
4757 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4758 was an oversight, and furthermore, ever since the addition of extra
4759 controls (e.g. 4.43/32), the checks on when to allow different forms of
4760 "control" were broken. There should now be diagnostics for all cases when a
4761 control that does not make sense is encountered.
4763 7. $recipients is now available in the predata ACL (oversight).
4765 8. Tidy the search cache before the fork to do a delivery from a message
4766 received from the command line. Otherwise the child will trigger a lookup
4767 failure and thereby defer the delivery if it tries to use (for example) a
4768 cached ldap connection that the parent has called unbind on.
4770 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4771 of $address_data from the recipient verification was clobbered by the
4772 sender verification.
4774 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4775 was its contents. (It was OK if the option was not defined at all.)
4777 11. A "Completed" log line is now written for messages that are removed from
4778 the spool by the -Mrm option.
4780 12. $host_address is now set to the target address during the checking of
4781 ignore_target_hosts.
4783 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4784 being passed; this would have caused $sender_host_name to have been used if
4785 matching the list had actually called for a host name (not very likely,
4786 since this list is usually IP addresses). A host name is now passed as
4789 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4790 code that specifies a non-restarting handler (typically sigaction() in
4791 modern systems) in an attempt to fix a rare and obscure crash bug.
4793 15. Narrowed the window for a race in the daemon that could cause it to ignore
4794 SIGCHLD signals. This is not a major problem, because they are used only to
4795 wake it up if nothing else does.
4797 16. A malformed maildirsize file could cause Exim to calculate negative values
4798 for the mailbox size or file count. Odd effects could occur as a result.
4799 The maildirsize information is now recalculated if the size or filecount
4802 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4803 support for a long time. Removed HAVE_SYS_VFS_H.
4805 18. Updated exipick to current release from John Jetmore.
4807 19. Allow an empty sender to be matched against a lookup in an address list.
4808 Previously the only cases considered were a regular expression, or an
4811 20. Exim went into a mad DNS lookup loop when doing a callout where the
4812 host was specified on the transport, if the DNS lookup yielded more than
4815 21. The RFC2047 encoding function was originally intended for short strings
4816 such as real names; it was not keeping to the 75-character limit for
4817 encoded words that the RFC imposes. It now respects the limit, and
4818 generates multiple encoded words if necessary. To be on the safe side, I
4819 have increased the buffer size for the ${rfc2047: expansion operator from
4822 22. Failure to deliver a bounce message always caused it to be frozen, even if
4823 there was an errors_to setting on the router. The errors_to setting is now
4826 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4827 canonical form (fully expanded) before being placed in
4828 $sender_host_address.
4830 24. Updated eximstats to version 1.33
4832 25. Include certificate and key file names in error message when GnuTLS fails
4833 to set them up, because the GnuTLS error message doesn't include the name
4834 of the failing file when there is a problem reading it.
4836 26. Expand error message when OpenSSL has problems setting up cert/key files.
4839 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4840 (this can affect the format of dates).
4842 28. exim_tidydb, when checking for the continued existence of a message for
4843 which it has found a message-specific retry record, was not finding
4844 messages that were in split spool directories. Consequently, it was
4845 deleting retry records that should have stayed in existence.
4847 29. eximstats updated to version 1.35
4848 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4849 1.35 - bugfix such that pie charts by volume are generated correctly
4851 30. The SPA authentication driver was not abandoning authentication and moving
4852 on to the next authenticator when an expansion was forced to fail,
4853 contradicting the general specification for all authenticators. Instead it
4854 was generating a temporary error. It now behaves as specified.
4856 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4857 (the order specifies the preference for clients). The order is now AES256,
4858 AES128, 3DES, ARCFOUR128.
4860 31. Small patch to Sieve code - explicitly set From: when generating an
4863 32. Exim crashed if a remote delivery caused a very long error message to be
4864 recorded - for instance if somebody sent an entire SpamAssassin report back
4865 as a large number of 550 error lines. This bug was coincidentally fixed by
4866 increasing the size of one of Exim's internal buffers (big_buffer) that
4867 happened as part of the Exiscan merge. However, to be on the safe side, I
4868 have made the code more robust (and fixed the comments that describe what
4871 33. Some experimental protocols are using DNS PTR records for new purposes. The
4872 keys for these records are domain names, not reversed IP addresses. The
4873 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4874 leaves it alone. Component reversal etc. now happens only for IP addresses.
4877 34. The host_aton() function is supposed to be passed a string that is known
4878 to be a valid IP address. However, in the case of IPv6 addresses, it was
4879 not checking this. This is a hostage to fortune. Exim now panics and dies
4880 if the condition is not met. A case was found where this could be provoked
4881 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4882 components; fortuitously, this particular loophole had already been fixed
4883 by change 4.50/55 or 4.44/33 above.
4885 If there are any other similar loopholes, the new check in host_aton()
4886 itself should stop them being exploited. The report I received stated that
4887 data on the command line could provoke the exploit when Exim was running as
4888 exim, but did not say which command line option was involved. All I could
4889 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4890 running as the user.
4893 35. There was a buffer overflow vulnerability in the SPA authentication code
4894 (which came originally from the Samba project). I have added a test to the
4895 spa_base64_to_bits() function which I hope fixes it.
4898 36. The daemon start-up calls getloadavg() while still root for those OS that
4899 need the first call to be done as root, but it missed one case: when
4900 deliver_queue_load_max is set with deliver_drop_privilege. This is
4901 necessary for the benefit of the queue runner, because there is no re-exec
4902 when deliver_drop_privilege is set.
4904 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4905 was in use, was not putting the data itself into the right store pool;
4906 consequently, it could be overwritten for a subsequent message in the same
4907 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4910 38. Sometimes the final signoff response after QUIT could fail to get
4911 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4912 < 0 before doing a fflush(). This bug looks as though it goes back to the
4913 introduction of TLS in release 3.20, but "sometimes" must have been rare
4914 because the tests only now provoked it.
4920 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4921 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4922 Now it returns an integer. A number of calls were still expecting a T/F
4923 return. Fortuitously, in all cases, the tests worked in OK situations,
4924 which is the norm. However, things would have gone wrong on any write
4925 failures on the smtp file descriptor. This function is used when sending
4926 messages over SMTP and also when doing verify callouts.
4928 2. When Exim is called to do synchronous delivery of a locally submitted
4929 message (the -odf or -odi options), it no longer closes stderr before doing
4932 3. Implemented the mua_wrapper option.
4934 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4936 5. Implemented the functions header_remove(), header_testname(),
4937 header_add_at_position(), and receive_remove_recipient(), and exported them
4940 6. If an ACL "warn" statement specified the addition of headers, Exim already
4941 inserted X-ACL-Warn: at the start if there was no header name. However, it
4942 was not making this test for the second and subsequent header lines if
4943 there were newlines in the string. This meant that an invalid header could
4944 be inserted if Exim was badly configured.
4946 7. Allow an ACL "warn" statement to add header lines at the start or after all
4947 the Received: headers, as well as at the end.
4949 8. Added the rcpt_4xx retry error code.
4951 9. Added postmaster_mailfrom=xxx to callout verification option.
4953 10. Added mailfrom=xxxx to the callout verification option, for verify=
4956 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4957 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4959 12. Inserted some casts to stop certain compilers warning when using pointer
4960 differences as field lengths or precisions in printf-type calls (mostly
4961 affecting debugging statements).
4963 13. Added optional readline() support for -be (dynamically loaded).
4965 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4966 same clock tick as a message's arrival, so that its received time was the
4967 same as the "first fail" time on the retry record, and that message
4968 remained on the queue past the ultimate address timeout, every queue runner
4969 would try a delivery (because it was past the ultimate address timeout) but
4970 after another failure, the ultimate address timeout, which should have then
4971 bounced the address, did not kick in. This was a "< instead of <=" error;
4972 in most cases the first failure would have been in the next clock tick
4973 after the received time, and all would be well.
4975 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4976 being recognized when the domain list was tested by the match_domain
4977 condition in an expansion string.
4979 16. Added the ${str2b64: operator.
4981 17. Exim was always calling setrlimit() to set a large limit for the number of
4982 processes, without checking whether the existing limit was already
4983 adequate. (It did check for the limit on file descriptors.) Furthermore,
4984 errors from getrlimit() and setrlimit() were being ignored. Now they are
4985 logged to the main and panic logs, but Exim does carry on, to try to do its
4986 job under whatever limits there are.
4988 18. Imported PCRE 5.0.
4990 19. Trivial typo in log message " temporarily refused connection" (the leading
4993 20. If the log selector return_path_on_delivery was set and an address was
4994 redirected to /dev/null, the delivery process crashed because it assumed
4995 that a return path would always be set for a "successful" delivery. In this
4996 case, the whole delivery is bypassed as an optimization, and therefore no
4999 21. Internal re-arrangement: the function for sending a challenge and reading
5000 a response while authentication was assuming a zero-terminated challenge
5001 string. It's now changed to take a pointer and a length, to allow for
5002 binary data in such strings.
5004 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5006 23. Exim was not respecting finduser_retries when seeking the login of the
5007 uid under which it was called; it was always trying 10 times. (The default
5008 setting of finduser_retries is zero.) Also, it was sleeping after the final
5009 failure, which is pointless.
5011 24. Implemented tls_on_connect_ports.
5013 25. Implemented acl_smtp_predata.
5015 26. If the domain in control=submission is set empty, Exim assumes that the
5016 authenticated id is a complete email address when it generates From: or
5017 Sender: header lines.
5019 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5020 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5021 chown and chgrp in /bin and hostname in /usr/bin.
5023 28. Exim was keeping the "process log" file open after each use, just as it
5024 does for the main log. This opens the possibility of it remaining open for
5025 long periods when the USR1 signal hits a daemon. Occasional processlog
5026 errors were reported, that could have been caused by this. Anyway, it seems
5027 much more sensible not to leave this file open at all, so that is what now
5030 29. The long-running daemon process does not normally write to the log once it
5031 has entered its main loop, and it closes the log before doing so. This is
5032 so that log files can straightforwardly be renamed and moved. However,
5033 there are a couple of unusual error situations where the daemon does write
5034 log entries, and I had neglected to close the log afterwards.
5036 30. The text of an SMTP error response that was received during a remote
5037 delivery was being truncated at 512 bytes. This is too short for some of
5038 the long messages that one sometimes sees. I've increased the limit to
5041 31. It is now possible to make retry rules that apply only when a message has a
5042 specific sender, in particular, an empty sender.
5044 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5045 it possible to be selective about when SMTP synchronization is enforced.
5047 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5049 32. Implemented hosts_connection_nolog.
5051 33. Added an ACL for QUIT.
5053 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5056 35. Added mailbox_size and mailbox_filecount to appendfile.
5058 36. Added control = no_multiline_responses to ACLs.
5060 37. There was a bug in the logic of the code that waits for the clock to tick
5061 in the case where the clock went backwards by a substantial amount such
5062 that the microsecond fraction of "now" was more than the microsecond
5063 fraction of "then" (but the whole seconds number was less).
5065 38. Added support for the libradius Radius client library this is found on
5066 FreeBSD (previously only the radiusclient library was supported).
5072 1. When certain lookups returned multiple values in the form name=value, the
5073 quoting of the values was not always being done properly. Specifically:
5074 (a) If the value started with a double quote, but contained no whitespace,
5076 (b) If the value contained whitespace other than a space character (i.e.
5077 tabs or newlines or carriage returns) it was not quoted.
5078 This fix has been applied to the mysql and pgsql lookups by writing a
5079 separate quoting function and calling it from the lookup code. The fix
5080 should probably also be applied to nisplus, ibase and oracle lookups, but
5081 since I cannot test any of those, I have not disturbed their existing code.
5083 2. A hit in the callout cache for a specific address caused a log line with no
5084 reason for rejecting RCPT. Now it says "Previous (cached) callout
5085 verification failure".
5087 3. There was an off-by-one bug in the queryprogram router. An over-long
5088 return line was truncated at 256 instead of 255 characters, thereby
5089 overflowing its buffer with the terminating zero. As well as fixing this, I
5090 have increased the buffer size to 1024 (and made a note to document this).
5092 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5093 when Exim is waiting for an SMTP response from a remote server, Exim
5094 restarts its select() call on the socket, thereby resetting its timeout.
5095 This is not a problem when such interrupts are rare. Somebody set up a cron
5096 job to run exiwhat every 2 minutes, which is less than the normal select()
5097 timeout (5 or 10 minutes). This meant that the select() timeout never
5098 kicked in because it was always reset. I have fixed this by comparing the
5099 time when an interrupt arrives with the time at the start of the first call
5100 to select(). If more time than the timeout has elapsed, the interrupt is
5101 treated as a timeout.
5103 5. Some internal re-factoring in preparation for the addition of Sieve
5104 extensions (by MH). In particular, the "personal" test is moved to a
5105 separate function, and given an option for scanning Cc: and Bcc: (which is
5106 not set for Exim filters).
5108 6. When Exim created an email address using the login of the caller as the
5109 local part (e.g. when creating a From: or Sender: header line), it was not
5110 quoting the local part when it contained special characters such as @.
5112 7. Installed new OpenBSD configuration files.
5114 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5115 try to make them clearer.
5117 9. Callout options, other than the timeout value, were being ignored when
5118 verifying sender addresses in header lines. For example, when using
5120 verify = header_sender/callout=no_cache
5122 the cache was (incorrectly) being used.
5124 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5125 only the Cygwin environment.
5127 11. When return_path_on_delivery was set as a log selector, if different remote
5128 addresses in the same message used different return paths and parallel
5129 remote delivery occurred, the wrong values would sometimes be logged.
5130 (Whenever a remote delivery process finished, the return path value from
5131 the most recently started remote delivery process was logged.)
5133 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5134 lines when AUTH and/or TLS are in use. This is the "received protocol"
5135 field. Exim used to use "asmtp" for authenticated SMTP, without any
5136 indication (in the protocol name) for TLS use. Now it follows the RFC and
5137 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5138 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5139 names appear in log lines as well as in Received: header lines.
5141 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5142 extensions, and comparison tests, and to fix some bugs.
5144 14. Changes to the "personal" filter test:
5146 (1) The test was buggy in that it was just doing the equivalent of
5147 "contains" tests on header lines. For example, if a user's address was
5148 anne@some.where, the "personal" test would incorrectly be true for
5150 To: susanne@some.where
5152 This test is now done by extracting each address from the header in turn,
5153 and checking the entire address. Other tests that are part of "personal"
5154 are now done using regular expressions (for example, to check local parts
5155 of addresses in From: header lines).
5157 (2) The list of non-personal local parts in From: addresses has been
5158 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5159 taken from the Sieve specification recommendations.
5161 (3) If the message contains any header line starting with "List-" it is
5162 treated as non-personal.
5164 (4) The test for "circular" in the Subject: header line has been removed
5165 because it now seems ill-conceived.
5167 15. Minor typos in src/EDITME comments corrected.
5169 16. Installed latest exipick from John Jetmore.
5171 17. If headers_add on a router specified a text string that was too long for
5172 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5173 of string_sprintf() is now avoided.
5175 18. $message_body_size was not set (it was always zero) when running the DATA
5176 ACL and the local_scan() function.
5178 19. For the "mail" command in an Exim filter, no default was being set for
5179 the once_repeat time, causing a random time value to be used if "once" was
5180 specified. (If the value happened to be <= 0, no repeat happened.) The
5181 default is now 0s, meaning "never repeat". The "vacation" command was OK
5182 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5183 (I found it when inspecting the code).
5185 20. There is now an overall timeout for performing a callout verification. It
5186 defaults to 4 times the callout timeout, which applies to individual SMTP
5187 commands during the callout. The overall timeout applies when there is more
5188 than one host that can be tried. The timeout is checked before trying the
5189 next host. This prevents very long delays if there are a large number of
5190 hosts and all are timing out (e.g. when the network connections are timing
5191 out). The value of the overall timeout can be changed by specifying an
5192 additional sub-option for "callout", called "maxwait". For example:
5194 verify = sender/callout=5s,maxwait=20s
5196 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5197 to the end before writing, but this should make it even safer).
5199 22. Exim was forgetting that it had advertised PIPELINING for the second and
5200 subsequent messages on an SMTP connection. It was also not resetting its
5201 memory on STARTTLS and an internal HELO.
5203 23. When Exim logs an SMTP synchronization error within a session, it now
5204 records whether PIPELINING has been advertised or not.
5206 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5207 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5208 rather than long int.
5210 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5212 26. Added the never_mail option to autoreply.
5218 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5219 crash if the getsockname() call failed; this can happen if a connection is
5220 closed very soon after it is established. The problem was simply in the
5221 order in which certain operations were done, causing Exim to try to write
5222 to the SMTP stream before it had set up the file descriptor. The bug has
5223 been fixed by making things happen in the correct order.
5229 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5230 before the connection was closed, thus losing the rejection response.
5232 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5233 some early Solaris releases, but causes trouble in current releases where
5234 socklen_t is defined.
5236 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5239 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5242 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5243 that releasing the top of it at the end releases what was used for sub-
5244 expansions (unless the block got too big). However, discard this block if
5245 the first thing is a variable or header, so that we can use its block when
5246 it is dynamic (useful for very large $message_headers, for example).
5248 6. Lookups now cache *every* query, not just the most recent. A new, separate
5249 store pool is used for this. It can be recovered when all lookup caches are
5250 flushed. Lookups now release memory at the end of their result strings.
5251 This has involved some general refactoring of the lookup sources.
5253 7. Some code has been added to the store_xxx() functions to reduce the amount
5254 of flapping under certain conditions.
5256 8. log_incoming_interface used to affect only the <= reception log lines. Now
5257 it causes the local interface and port to be added to several more SMTP log
5258 lines, for example "SMTP connection from", and rejection lines.
5260 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5262 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5264 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5265 could be overwritten at the end of the current message (or the start of a
5266 new message if it was set in a HELO ACL). The value is now preserved for
5267 the duration of the SMTP connection.
5269 12. If a transport had a headers_rewrite setting, and a matching header line
5270 contained an unqualified address, that address was qualified, even if it
5271 did not match any rewriting rules. The underlying bug was that the values
5272 of the flags that permit the existence of unqualified sender and recipient
5273 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5274 non-local messages, and by -bnq for local messages) were not being
5275 preserved with the message after it was received.
5277 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5278 "next input=" as part of the text comprising the host identity instead of
5279 the correct text. The code was using the same buffer for two different
5280 strings. However, depending on which order the printing function evaluated
5281 its arguments, the bug did not always show up. Under Linux, for example, my
5282 test suite worked just fine.
5284 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5285 doesn't work with some older versions of Perl. It has been changed to "my",
5286 which in any case is probably the better facility to use.
5288 15. A really picky compiler found some instances of statements for creating
5289 error messages that either had too many or two few arguments for the format
5292 16. The size of the buffer for calls to the DNS resolver has been increased
5293 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5294 for addresses that have a lot of PTR records. This alleviates a problem; it
5295 does not fully solve it.
5297 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5298 buffer now truncates the list and logs the incident, which is the same
5299 action as happens when Exim is looking up a host name and its aliases.
5300 Previously in this situation something unpredictable would happen;
5301 sometimes it was "internal error: store_reset failed".
5303 18. If a server dropped the connection unexpectedly when an Exim client was
5304 using GnuTLS and trying to read a response, the client delivery process
5305 crashed while trying to generate an error log message.
5307 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5308 string, for example:
5310 warn message = H1: something\nH2: something
5312 the text was added as a single header line from Exim's point of view
5313 though it ended up OK in the delivered message. However, searching for the
5314 second and subsequent header lines using $h_h2: did not work. This has been
5315 fixed. Similarly, if a system filter added multiple headers in this way,
5316 the routers could not see them.
5318 20. Expanded the error message when iplsearch is called with an invalid key to
5319 suggest using net-iplsearch in a host list.
5321 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5322 ACLs are no longer actually imposed (and a message to that effect is
5325 22. If a "gecos" field in a passwd entry contained escaped characters, in
5326 particular, if it contained a \" sequence, Exim got it wrong when building
5327 a From: or a Sender: header from that name. A second bug also caused
5328 incorrect handling when an unquoted " was present following a character
5329 that needed quoting.
5331 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5332 was not being matched caselessly.
5334 24. Arranged for all hyphens in the exim.8 source to be escaped with
5337 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5338 properly. Recipient callout cache records were still being keyed to include
5339 the sender, even when use_sender was set false. This led to far more
5340 callouts that were necessary. The sender is no longer included in the key
5341 when use_sender is false.
5343 26. Added "control = submission" modifier to ACLs.
5345 27. Added the ${base62d: operator to decode base 62 numbers.
5347 28. dnsdb lookups can now access SRV records.
5349 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5350 the configuration file.
5352 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5353 (-v) mode. This makes the output for a verbose queue run more intelligible.
5355 31. Added a use_postmaster feature to recipient callouts.
5357 32. Added the $body_zerocount variable, containing the number of binary zero
5358 bytes in the message body.
5360 33. The time of last modification of the "new" subdirectory is now used as the
5361 "mailbox time last read" when there is a quota error for a maildir
5364 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5366 35. Added +ignore_unknown as a special item in host lists.
5368 36. Code for decoding IPv6 addresses in host lists is now included, even if
5369 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5370 address was recognized as an IP address, but was then not correctly decoded
5371 into binary, causing unexpected and incorrect effects when compared with
5378 1. Very minor rewording of debugging text in manualroute to say "list of
5379 hosts" instead of "hostlist".
5381 2. If verify=header_syntax was set, and a header line with an unqualified
5382 address (no domain) and a large number of spaces between the end of the
5383 name and the colon was received, the reception process suffered a buffer
5384 overflow, and (when I tested it) crashed. This was caused by some obsolete
5385 code that should have been removed. The fix is to remove it!
5387 3. When running in the test harness, delay a bit after writing a bounce
5388 message to get a bit more predictability in the log output.
5390 4. Added a call to search_tidyup() just before forking a reception process. In
5391 theory, someone could use a lookup in the expansion of smtp_accept_max_
5392 per_host which, without the tidyup, could leave open a database connection.
5394 5. Added the variables $recipient_data and $sender_data which get set from a
5395 lookup success in an ACL "recipients" or "senders" condition, or a router
5396 "senders" option, similar to $domain_data and $local_part_data.
5398 6. Moved the writing of debug_print from before to after the "senders" test
5401 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5402 problems for message scanning, either using a data ACL, or using
5403 local_scan() because the Received: header was not generated till after they
5404 were called (in order to set the time as the time of reception completion).
5405 I have revised the way this works. The header is now generated after the
5406 body is received, but before the ACL or local_scan() are called. After they
5407 are run, the timestamp in the header is updated.
5413 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5414 before starting a queue runner without re-exec. This happened only when
5415 deliver_drop_privilege was set or when the Exim user was set to root. The
5416 effect of the bug was that timeouts during subsequent deliveries caused
5417 crashes instead of being properly handled. The handler is now left at its
5418 default (and expected) setting.
5420 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5421 message, again when deliver_drop_privilege is set or Exim is run as root.
5422 The bug described in (1) was not present in this case, but the tidying up
5423 of the other signals was missing. I have made the two cases consistent.
5425 3. The ignore_target_hosts setting on a manualroute router was being ignored
5426 for hosts that were looked up using the /MX notation.
5428 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5431 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5432 operated on the sender address. After changing the $sender_address to <>
5433 for the sender address verify, Exim was re-instated it as the original
5434 (before rewriting) address, but remembering that it had rewritten it, so it
5435 wasn't rewriting it again. This bug also had the effect of breaking the
5436 sender address verification caching when the sender address was rewritten.
5438 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5439 This has been changed so that if the ip literal address matches
5440 ignore_target_hosts, the router declines.
5442 7. Added expansion conditions match_domain, match_address, and match_local_
5443 part (NOT match_host).
5445 8. The placeholder for the Received: header didn't have a length field set.
5447 9. Added code to Exim itself and to exim_lock to test for a specific race
5448 condition that could lead to file corruption when using MBX delivery. The
5449 issue is with the lockfile that is created in /tmp. If this file is removed
5450 after a process has opened it but before that process has acquired a lock,
5451 there is the potential for a second process to recreate the file and also
5452 acquire a lock. This could lead to two Exim processes writing to the file
5453 at the same time. The added code performs the same test as UW imapd; it
5454 checks after acquiring the lock that its file descriptor still refers to
5455 the same named file.
5457 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5458 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5461 11. Added the smtp_active_hostname option. If used, this will typically be made
5462 to depend on the incoming interface address. Because $interface_address is
5463 not set up until the daemon has forked a reception process, error responses
5464 that can happen earlier (such as "too many connections") no longer contain
5467 12. If an expansion in a condition on a "warn" statement fails because a lookup
5468 defers, the "warn" statement is abandoned, and the next ACL statement is
5469 processed. Previously this caused the whole ACL to be aborted.
5471 13. Added the iplsearch lookup type.
5473 14. Added ident_timeout as a log selector.
5475 15. Added tls_certificate_verified as a log selector.
5477 16. Added a global option tls_require_ciphers (compare the smtp transport
5478 option of the same name). This controls incoming TLS connections.
5480 17. I finally figured out how to make tls_require_ciphers do a similar thing
5481 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5482 before starting the TLS session.
5484 18. Tabs are now shown as \t in -bP output.
5486 19. If the log selector return_path_on_delivery was set, Exim crashed when
5487 bouncing a message because it had too many Received: header lines.
5489 20. If two routers both had headers_remove settings, and the first one included
5490 a superfluous trailing colon, the final name in the first list and the
5491 first name in the second list were incorrectly joined into one item (with a
5492 colon in the middle).
5498 1. Added -C and -D options to the exinext utility, mainly to make it easier
5499 to include in the automated testing, but these could be helpful when
5500 multiple configurations are in use.
5502 2. The exinext utility was not formatting the output nicely when there was
5503 an alternate port involved in the retry record key, nor when there was a
5504 message id as well (for retries that were specific to a specific message
5505 and a specific host). It was also confused by IPv6 addresses, because of
5506 the additional colons they contain. I have fixed the IPv4 problem, and
5507 patched it up to do a reasonable job for IPv6.
5509 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5510 delivery, the log line now contains "pipelined" if PIPELINING was used.
5512 4. An SMTP transport process used to panic and die if the bind() call to set
5513 an explicit outgoing interface failed. This has been changed; it is now
5514 treated in the same way as a connect() failure.
5516 5. A reference to $sender_host_name in the part of a conditional expansion
5517 that was being skipped was still causing a DNS lookup. This no longer
5520 6. The def: expansion condition was not recognizing references to header lines
5521 that used bh_ and bheader_.
5523 7. Added the _cache feature to named lists.
5525 8. The code for checking quota_filecount in the appendfile transport was
5526 allowing one more file than it should have been.
5528 9. For compatibility with Sendmail, the command line option
5536 and sets the incoming protocol and host name (for trusted callers). The
5537 host name and its colon can be omitted when only the protocol is to be set.
5538 Note the Exim already has two private options, -pd and -ps, that refer to
5539 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5540 "s", but I don't think that's a major issue.
5542 10. A number of refactoring changes to the code, none of which should affect
5545 (a) The number of logging options was getting close to filling up the
5546 32-bit word that was used as a bit map. I have split them into two classes:
5547 those that are passed in the argument to log_write(), and those that are
5548 only ever tested independently outside of that function. These are now in
5549 separate 32-bit words, so there is plenty of room for expansion again.
5550 There is no change in the user interface or the logging behaviour.
5552 (b) When building, for example, log lines, the code previously used a
5553 macro that called string_cat() twice, in order to add two strings. This is
5554 not really sufficiently general. Furthermore, there was one instance where
5555 it was actually wrong because one of the argument was used twice, and in
5556 one call a function was used. (As it happened, calling the function twice
5557 did not affect the overall behaviour.) The macro has been replaced by a
5558 function that can join an arbitrary number of extra strings onto a growing
5561 (c) The code for expansion conditions now uses a table and a binary chop
5562 instead of a serial search (which was left over from when there were very
5563 few conditions). Also, it now recognizes conditions like "pam" even when
5564 the relevant support is not compiled in: a suitably worded error message is
5565 given if an attempt is made to use such a condition.
5567 11. Added ${time_interval:xxxxx}.
5569 12. A bug was causing one of the ddress fields not to be passed back correctly
5570 from remote delivery subprocesses. The field in question was not being
5571 subsequently used, so this caused to problems in practice.
5573 13. Added new log selectors queue_time and deliver_time.
5575 14. Might have fixed a bug in maildirsizefile handling that threw up
5576 "unexpected character" debug warnings, and recalculated the data
5577 unnecessarily. In any case, I expanded the warning message to give more
5580 15. Added the message "Restricted characters in address" to the statements in
5581 the default ACL that block characters like @ and % in local parts.
5583 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5584 Three changes have been made:
5586 (a) There was a serious bug; a negative response to MAIL caused the whole
5587 recipient domain to be cached as invalid, thereby blocking all messages
5588 to all local parts at the same domain, from all senders. This bug has
5589 been fixed. The domain is no longer cached after a negative response to
5590 MAIL if the sender used is not empty.
5592 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5595 (c) A new callout option, "use_sender" has been added for people who want
5596 the modified behaviour.
5602 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5605 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5606 indeed breaks things for older releases.
5608 3. Added additional logging to the case where there is a problem reading data
5609 from a filter that is running in a subprocess using a pipe, in order to
5610 try to track down a specific problem.
5612 4. Testing facility fudge: when running in the test harness and attempting
5613 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5614 getting "No route to host". Convert this to a timeout.
5616 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5619 6. Some OS don't have socklen_t but use size_t instead. This affects the
5620 fifth argument of getsockopt() amongst other things. This is now
5621 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5622 can be set for individual OS. I have set it for SunOS5, OSF1, and
5623 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5624 some earlier ones do not.
5626 7. Change 4.30/15 was not doing the test caselessly.
5628 8. The standard form for an IPv6 address literal was being rejected by address
5629 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5630 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5631 this, as well as the form without the "IPv6" on the front (but only when
5632 address literals are enabled, of course).
5634 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5636 10. Exim crashed if a message with an empty sender address specified by -f
5637 encountered a router with an errors_to setting. This could be provoked only
5638 by a command such as
5642 where an empty string was supplied; "<>" did not hit this bug.
5644 11. Installed PCRE release 4.5.
5646 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5647 remained set. It is now erased.
5649 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5650 times from message ids (which are base 36 rather than the normal 62).
5652 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5653 were being counted as actual protocol errors, and logged if the log
5654 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5655 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5656 and DATA following a set of rejected RCPTs do not count as protocol errors.
5657 In other words, Exim assumes they were pipelined, though this may not
5658 actually be the case. Of course, in all cases the client gets an
5659 appropriate error code.
5661 15. If a lookup fails in an ACL condition, a message about the failure may
5662 be available; it is used if testing the ACL cannot continue, because most
5663 such messages specify what the cause of the deferral is. However, some
5664 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5665 that caused an old message to be retained and used if a later statement
5666 caused a defer, replacing the real cause of the deferral.
5668 16. If an IP address had so many PTR records that the DNS lookup buffer
5669 was not large enough to hold them, Exim could crash while trying to process
5670 the truncated data. It now detects and logs this case.
5672 17. Further to 4.21/58, another change has been made: if (and only if) the
5673 first line of a message (the first header line) ends with CRLF, a bare LF
5674 in a subsequent header line has a space inserted after it, so as not to
5675 terminate the header.
5677 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5678 unnecessarily, used atoi() instead of strtol(), and didn't check the
5679 termination when getting file sizes from file names by regex.
5681 19. Completely re-implemented the support for maildirsize files, in the light
5682 of a number of problems with the previous contributed implementation
5683 (4.30/29). In particular:
5685 . If the quota is zero, the maildirsize file is maintained, but no quota is
5688 . If the maildir directory does not exist, it is created before any attempt
5689 to write a maildirsize file.
5691 . The quota value in the file is just a cache; if the quota is changed in
5692 the transport, the new value overrides.
5694 . A regular expression is available for excluding directories from the
5697 20. The autoreply transport checks the characters in options that define the
5698 message's headers; it allows continued headers, but it was checking with
5699 isspace() after an embedded newline instead of explicitly looking for a
5702 21. If all the "regular" hosts to which an address was routed had passed their
5703 expiry times, and had not reached their retry times, the address was
5704 bounced, even if fallback hosts were defined. Now Exim should go on to try
5707 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5708 equivalent code in the SMTP transport. Some hosts send humungous responses
5709 to HELO/EHLO, more than 1024 it seems.
5711 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5712 gives compiler warnings in some environments. I've now done it "properly",
5715 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5716 (because of problems with the built-in one) was declared to return uschar *
5717 instead of char *, causing compiler failure.
5719 25. Fixed a file descriptor leak when processing alias/forward files.
5721 26. Fixed a minor format string issue in dbfn.c.
5723 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5725 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5726 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5729 29. When a sender address is verified, it is cached, to save repeating the test
5730 when there is more than one recipient in a message. However, when the
5731 verification involves a callout, it is possible for different callout
5732 options to be set for different recipients. It is too complicated to keep
5733 track of this in the cache, so now Exim always runs a verification when a
5734 callout is required, relying on the callout cache for the optimization.
5735 The overhead is duplication of the address routing, but this should not be
5738 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5739 to be verified with callout=postmaster, and the main callout worked but the
5740 postmaster check failed, the verification correctly failed. However, if a
5741 subsequent RCPT command asked for sender verification *without* the
5742 postmaster check, incorrect caching caused this verification also to fail,
5745 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5746 it was not caching the DNS options (qualify_single, search_parents) that
5747 were used when the lookup failed. A subsequent lookup with different
5748 options therefore always gave the same answer, though there were cases
5749 where it should not have. (Example: a "domains = !$mx_any" option on a
5750 dnslookup router: the "domains" option is always processed without any
5751 widening, but the router might have qualify_single set.) Now Exim uses the
5752 cached value only when the same options are set.
5754 32. Added John Jetmore's "exipick" utility to the distribution.
5756 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5757 than a timeout (e.g. a certificate is required, and is not provided), an
5758 Exim server now closes the connection immediately. Previously it waited for
5759 the client to close - but if the client is SSL, it seems that they each
5760 wait for each other, leading to a delay before one of them times out.
5762 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5763 maintained 0.8.x compatibility because I don't think many are using it, and
5764 it is clearly obsolete.
5766 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5769 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5770 client certificate was expired. A simple patch fixes this, though I don't
5771 understand the full logic of why the verify callback is called multiple
5774 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5775 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5776 which causes problems with some clients (such as the Certicom SSL Plus
5777 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5778 disables the coutermeasure allowing Eudora to connect."
5780 38. Exim was not checking that a write() to a log file succeeded. This could
5781 lead to Bad Things if a log got too big, in particular if it hit a file
5782 size limit. Exim now panics and dies if it cannot write to a log file, just
5783 as it does if it cannot open a log file.
5785 39. Modified OS/Makefile-Linux so that it now contains
5787 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5789 The two -D definitions ensure that Exim is compiled with large file
5790 support, which makes it possible to handle log files that are bigger than
5793 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5794 instance) a domain was checked against a named list that involved a lookup,
5795 causing $domain_data to be set, then another domain was checked against the
5796 same list, then the first domain was re-checked, the value of $domain_data
5797 after the final check could be wrong. In particular, if the second check
5798 failed, it could be set empty. This bug probably also applied to
5801 41. The strip_trailing_dot option was not being applied to the address given
5802 with the -f command-line option.
5804 42. The code for reading a message's header from the spool was incrementing
5805 $received_count, but never initializing it. This meant that the value was
5806 incorrect (doubled) while delivering a message in the same process in which
5807 it was received. In the most common configuration of Exim, this never
5808 happens - a fresh exec is done - but it can happen when
5809 deliver_drop_privilege is set.
5811 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5812 it now includes up to 150 characters of the unexpected data in the log
5815 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5816 and building data strings. The size of both of these buffers was 10 000
5817 bytes - far larger than anybody would *ever* want, thought I. Needless to
5818 say, somebody hit the limit. I have increased the maximum line length to
5819 20 000 and the maximum data length of concatenated lines to 100 000. I have
5820 also fixed two bugs, because there was no checking on these buffers. Tsk,
5821 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5822 buffer is too small.
5824 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5825 lsearch lookups. Now it does.
5827 46. When parsing a route_list item in a manualroute router, a fixed-length
5828 buffer was used for the list of hosts. I made this 1024 bytes long,
5829 thinking that nobody would ever have a list of hosts that long. Wrong.
5830 Somebody had a whole pile of complicated expansion conditions, and the
5831 string was silently truncated, leading to an expansion error. It turns out
5832 that it is easier to change to an unlimited length (owing to other changes
5833 that have happened since this code was originally written) than to build
5834 structure for giving a limitation error. The length of the item that
5835 expands into the list of hosts is now unlimited.
5837 47. The lsearch lookup could not handle data where the length of text line was
5838 more than 4095 characters. Such lines were truncated, leading to shortened
5839 data being returned. It should now handle lines of any length.
5841 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5842 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5845 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5846 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5847 no longer happens in comment lines. A list of replacements is now placed
5848 at the head of all of the source files, except those whose only change is
5849 to replace PERL_COMMAND in the very first #! line.
5851 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5852 messages on the queue, with a bottom-up merge sort, using code contributed
5853 by Michael Haardt. This should make operations like -bp somewhat faster on
5854 large queues. It won't affect queue runners, except when queue_run_in_order
5857 51. Installed eximstats 1.31 in the distribution.
5859 52. Added support for SRV lookups to the dnslookup router.
5861 53. If an ACL referred to $message_body or $message_body_end, the value was not
5862 reset for any messages that followed in the same SMTP session.
5864 54. The store-handling optimization for building very long strings was not
5865 differentiating between the different store pools. I don't think this
5866 actually made any difference in practice, but I've tidied it.
5868 55. While running the routers to verify a sender address, $sender_address
5869 was still set to the sender address. This is wrong, because when routing to
5870 send a bounce to the sender, it would be empty. Therefore, I have changed
5871 it so that, while verifying a sender address, $sender_address is set to <>.
5872 (There is no change to what happens when verifying a recipient address.)
5874 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5875 target A or AAAA records (if not already returned) without resetting the
5876 qualify_single or search_parents options of the DNS resolver. These are
5877 inappropriate in this case because the targets of MX and SRV records must
5878 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5879 target that, when qualified, matched something in the local domain. These
5880 two options are now turned off when doing these lookups.
5882 57. It seems that at least some releases of Reiserfs (which does not have the
5883 concept of a fixed number of inodes) returns zero and not -1 for the
5884 number of available inodes. This interacted badly with check_spool_inodes,
5885 which assumed that -1 was the "no such thing" setting. What I have done is
5886 to check that the total number of inodes is greater than zero before doing
5887 the test of how many are available.
5889 58. When a "warn" ACL statement has a log_message modifier, the message is
5890 remembered, and not repeated. This is to avoid a lot of repetition when a
5891 message has many recipients that cause the same warning to be written.
5892 However, Exim was preserving the list of already written lines for an
5893 entire SMTP session, which doesn't seem right. The memory is now reset if a
5894 new message is started.
5896 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5897 debugging output unless log_rewrite was also set.
5899 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5900 of (int)(handle) when we know that handle contains (void *)(-1).
5902 61. The Exim daemon panic-logs an error return when it closes the incoming
5903 connection. However "connection reset by peer" seems to be common, and
5904 isn't really an error worthy of noting specially, so that particular error
5907 62. When Exim is trying to find all the local interfaces, it used to panic and
5908 die if the ioctl to get the interface flags failed. However, it seems that
5909 on at least one OS (Solaris 9) it is possible to have an interface that is
5910 included in the list of interfaces, but for which you get a failure error
5911 for this call. This happens when the interface is not "plumbed" into a
5912 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5913 failure of the "get flags" call assumes that the interface is down.
5915 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5916 makes life easier for people who are doing arithmetic on fields extracted
5917 from dates, where you often get leading zeros that should not be
5918 interpreted as octal.
5920 64. Added qualify_domain to the redirect router, to override the global
5923 65. If a pathologically long header line contained very many addresses (the
5924 report of this problem mentioned 10 000) and each of them was rewritten,
5925 Exim could use up a very large amount of memory. (It kept on making new
5926 copies of the header line as it rewrote, and never released the old ones.)
5927 At the expense of a bit more processing, the header rewriting function has
5928 been changed so that it no longer eats memory in this way.
5930 66. The generation of the Received: header has been moved from the time that a
5931 message starts to be received, to the time that it finishes. The timestamp
5932 in the Received: header should now be very close to that of the <= log
5933 line. There are two side-effects of this change:
5935 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5936 logged header lines no longer include the local Received: line, because
5937 it has not yet been created. The same applies to a copy of the message
5938 that is returned to a non-SMTP sender when a message is rejected.
5940 (b) When a filter file is tested using -bf, no additional Received: header
5941 is added to the test message. After some thought, I decided that this
5944 This change does not affect the value of $received_for. It is still set
5945 after address rewriting, but before local_scan() is called.
5947 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5949 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5950 gave an unhelpful panic error message, and a defer error. I have managed to
5951 change this behaviour so that it now rejects any supplied certificate,
5952 which seems right, as the list of acceptable certificates is empty.
5954 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5955 gave an unhelpful defer error. I have not managed to make this reject any
5956 supplied certificates, but the error message it gives is "no certificate
5957 supplied", which is not helpful.
5959 70. exigrep's output now also includes lines that are not associated with any
5960 message, but which match the given pattern. Implemented by a patch from
5961 Martin Sluka, which also tidied up the Perl a bit.
5963 71. Recipient callout verification, like sender verification, was using <> in
5964 the MAIL FROM command. This isn't really the right thing, since the actual
5965 sender may affect whether the remote host accepts the recipient or not. I
5966 have changed it to use the actual sender in the callout; this means that
5967 the cache record is now keyed on a recipient/sender pair, not just the
5968 recipient address. There doesn't seem to be a real danger of callout loops,
5969 since a callout by the remote host to check the sender would use <>.
5970 [SEE ABOVE: changed after hitting problems.]
5972 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5973 temporary errors. However, in the case of such a code being given after
5974 the end of a data transmission (i.e. after ".") Exim was failing to write
5975 a retry record for the message. (Yes, there was some broken host that was
5976 actually sending 8xx at this point.)
5978 73. An unknown lookup type in a host list could cause Exim to panic-die when
5979 the list was checked. (An example that provoked this was putting <; in the
5980 middle of a list instead of at the start.) If this happened during a DATA
5981 ACL check, a -D file could be left lying around. This kind of configuration
5982 error no longer causes Exim to die; instead it causes a defer error. The
5983 incident is still logged to the main and panic logs.
5985 74. Buglet left over from Exim 3 conversion. The message "too many messages
5986 in one connection" was written to the rejectlog but not the mainlog, except
5987 when address rewriting (yes!) was being logged.
5989 75. Added write_rejectlog option.
5991 76. When a system filter was run not as root (that is, when system_filter_user
5992 was set), the values of the $n variables were not being returned to the
5993 main process; thus, they were not subsequently available in the $sn
5996 77. Added +return_path_on_delivery log selector.
5998 78. A connection timeout was being treated differently from recipients deferred
5999 when testing hosts_max_try with a message that was older than the host's
6000 retry timeout. (The host should not be counted, thus allowing all hosts to
6001 be tried at least once before bouncing.) This may have been the cause of an
6002 occasionally reported bug whereby a message would remain on the queue
6003 longer than the retry timeout, but would be bounced if a delivery was
6004 forced. I say "may" because I never totally pinned down the problem;
6005 setting up timeout/retry tests is difficult. See also the next item.
6007 79. The ultimate address timeout was not being applied to errors that involved
6008 a combination of host plus message (for example, a timeout on a MAIL
6009 command). When an address resolved to a number of possible hosts, and they
6010 were not all tried for each delivery (e.g. because of hosts_max_try), a
6011 message could remain on the queue longer than the retry timeout.
6013 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6016 81. Fixed an obscure SMTP outgoing bug which required at least the following
6017 conditions: (a) there was another message waiting for the same server;
6018 (b) the server returned 5xx to all RCPT commands in the first message so
6019 that the message was not completed; (c) the server dropped the connection
6020 or gave a negative response to the RSET that Exim sends to abort the
6021 transaction. The observed case was a dropped connection after DATA that had
6022 been sent in pipelining mode. That is, the server had advertised PIPELINING
6023 but was not implementing it correctly. The effect of the bug was incorrect
6024 behaviour, such as trying another host, and this could lead to a crash.
6030 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6031 and daemon.c were passed as pointers to ints; they should have been
6032 pointers to socklen_t variables (which are typically unsigned ints).
6034 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6037 3. Fixed a really odd bug that affected only the testing scheme; patching a
6038 certain fixed string in the binary changed the value of another string that
6039 happened to be identical to the end of the original first string.
6041 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6042 name", it returns that address as the IP address. On some operating
6043 systems (e.g. Solaris), it also passes back the IP address string as the
6044 "host name". However, on others (e.g. Linux), it passes back an empty
6045 string. Exim wasn't checking for this, and was changing the host name to an
6046 empty string, assuming it had been canonicalized.
6048 5. Although rare, it is permitted to have more than one PTR record for a given
6049 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6050 all the names associated with an address, because they do in Solaris.
6051 However, it seems that they do not in Linux for data that comes from the
6052 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6053 I found this out when I moved to a new Linux workstation and tried to run
6054 the Exim test suite.
6056 To get round this problem I have changed the code so that it now does its
6057 own call to the DNS to look up PTR records when searching for a host name.
6058 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6059 addresses that are only in /etc/hosts are still found.
6061 This behaviour is, however, controlled by an option called host_lookup_
6062 order, which defaults to "bydns:byaddr". If people want to use the other
6063 order, or indeed, just use one or the other means of lookup, they can
6064 specify it in this variable.
6066 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6067 some operating systems, this comes back from gethostbyaddr() as an empty
6068 string, and this is what Exim used to test for. However, it seems that in
6069 other systems, "." is yielded. Exim now tests for this case too.
6071 7. The values of check_spool_space and check_log_space are now held internally
6072 as a number of kilobytes instead of an absolute number of bytes. If a
6073 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6074 kilobyte. This means that much larger values can be stored.
6076 8. Exim monitor: an attempt to get the action menu when not actually pointing
6077 at a message produces an empty menu entitled "No message selected". This
6078 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6079 no entries in it ("Shell widget menu has zero width and/or height"). So I
6080 have added a single, blank menu entry in this case.
6082 9. Added ${quote_local_part.
6084 10. MIME decoding is now applied to the contents of Subject: header lines when
6087 11. Now that a reference to $sender_host_address automatically causes a reverse
6088 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6089 host lookup before query-style lookups in lists that might use this
6090 variable. This has therefore been abolished, and the "net-" prefix is no
6091 longer necessary for query-style lookups.
6093 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6094 have been a typo for LFLAGS, so it has been changed.
6096 13. The install script calls Exim with "-C /dev/null" in order to find the
6097 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6098 to be output. However, since Exim outputs its version number before the
6099 error, it didn't break the script. It just looked ugly. I fixed this by
6100 always allowing "-C /dev/null" if the caller is root.
6102 14. Ignore overlarge ACL variable number when reading spool file - insurance
6103 against a later release with more variables having written the file.
6105 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6106 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6107 this, as well as the form without the "IPv6" on the front.
6109 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6110 OS/Makefile-Darwin file.
6112 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6113 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6115 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6116 variable. However, some OS use ints for inodes. I've added cast to long int
6117 to get rid of the compiler warning.
6119 19. I had forgotten to lock out "/../" in configuration file names when
6120 ALT_CONFIG_PREFIX was set.
6122 20. Routers used for verification do not need to specify transports. However,
6123 if such a router generated a host list, and callout was configured, Exim
6124 crashed, because it could not find a port number from the (non-existent)
6125 transport. It now assumes port 25 in this circumstance.
6127 21. Added the -t option to exigrep.
6129 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6130 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6131 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6132 them is set, LOOKUP_LSEARCH is forced.
6134 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6136 24. Added sender and host information to the "rejected by local_scan()" log
6137 line; previously there was no indication of these.
6139 25. Added .include_if_exists.
6141 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6142 Linux. It turns out that not all file systems support this. Apparently some
6143 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6144 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6145 not supported on the file descriptor, is now ignored when Exim is trying to
6146 sync a directory. This applies only to Linux.
6148 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6150 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6151 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6154 29. Added support for maildirsize files from supplied patch (modified a bit).
6156 30. The use of :fail: followed by an empty string could lead Exim to respond to
6157 sender verification failures with (e.g.):
6159 550 Verification failed for <xxx>
6160 550 Sender verify failed
6162 where the first response line was missing the '-' that indicates it is not
6163 the final line of the response.
6165 31. The loop for finding the name of the user that called Exim had a hardwired
6166 limit of 10; it now uses the value of finduser_retries, which is used for
6167 all other user lookups.
6169 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6172 33. Exim was neglecting to zero errno before one call of strtol() when
6173 expanding a string and expecting an integer value. On some systems this
6174 resulted in spurious "integer overflow" errors. Also, it was casting the
6175 result into an int without checking.
6177 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6178 did not work. The code looks as if it has *never* worked, though it appears
6179 to have been documented since at least release 1.62. I have made it work.
6181 35. The "timeout_DNS" error in retry rules, also documented since at least
6182 1.62, also never worked. As it isn't clear exactly what this means, and
6183 clearly it isn't a major issue, I have abolished the feature by treating it
6184 as "timeout", and writing a warning to the main and panic logs.
6186 36. The display of retry rules for -brt wasn't always showing the error code
6189 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6190 timeout_connect_A, timeout_connect_MX.
6192 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6193 to the empty sender.
6195 39. The daemon was not analysing the content of -oX till after it had closed
6196 stderr and disconnected from the controlling terminal. This meant that any
6197 syntax errors were only noted on the panic log, and the return code from
6198 the command was 0. By re-arranging the code a little, I've made the
6199 decoding happen first, so such errors now appear on stderr, and the return
6200 code is 1. However, the actual setting up of the sockets still happens in
6201 the disconnected process, so errors there are still only recorded on the
6204 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6205 connections (as happens on some IP stacks) was logged at start up time as
6206 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6207 it from "IPv6 and IPv4", which means that two separate sockets are being
6210 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6211 says whether AF_INET or AF_INET6 was passed as an argument.
6213 42. Exiwhat output was messed up when time zones were included in log
6216 43. Exiwhat now gives more information about the daemon's listening ports,
6217 and whether -tls-on-connect was used.
6219 44. The "port" option of the smtp transport is now expanded.
6221 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6222 silently ignored. Now an error message is written to the main and panic
6225 46. There's a new ACL modifier called "logwrite" which writes to a log file
6226 as soon as it is encountered.
6228 47. Added $local_user_uid and $local_user_gid at routing time.
6230 48. Exim crashed when trying to verify a sender address that was being
6233 49. Exim was recognizing only a space character after ".include". It now also
6234 recognizes a tab character.
6236 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6237 extracting the relevant information from the specification. The man page no
6238 longer contains scrambled data for the -d option, and I've added a section
6239 at the front about calling Exim under different names.
6241 51. Added "extra_headers" argument to the "mail" command in filter files.
6243 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6246 53. Installed eximstats 1.29.
6248 54. Added transport_filter_timeout as a generic transport option.
6250 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6251 Cc: header lines. This was required by RFC 822, but it not required by RFC
6254 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6255 incoming messages that did not have them. Now it does so only if the
6256 message originates locally, that is, if there is no associated remote host
6257 address. When Resent- header lines are present, this applies to the Resent-
6258 lines rather than the non-Resent- lines.
6260 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6261 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6263 58. Messages for configuration errors now include the name of the main
6264 configuration file - useful now that there may be more than one file in a
6265 list (.included file names were always shown).
6267 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6268 for those rare installations that do not start the daemon as root or run it
6269 setuid root. I've cut out the call to initgroups() if the daemon is not
6272 60. The Exim user and group can now be bound into the binary as text strings
6273 that are looked up at the start of Exim's processing.
6275 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6277 62. Added $mailstore_basename variable.
6279 63. Installed patch to sieve.c from Michael Haardt.
6281 64. When Exim failed to open the panic log after failing to open the main log,
6282 the original message it was trying to log was written to stderr and debug
6283 output, but if they were not available (the usual case in production), it
6284 was lost. Now it is written to syslog before the two lines that record the
6285 failures to open the logs.
6287 65. Users' Exim filters run in subprocesses under the user's uid. It is
6288 possible for a "deliver" command or an alias in a "personal" command to
6289 provoke an address rewrite. If logging of address rewriting is configured,
6290 this fails because the process is not running as root or exim. There may be
6291 a better way of dealing with this, but for the moment (because 4.30 needs
6292 to be released), I have disabled address rewrite logging when running a
6293 filter in a non-root, non-exim process.
6299 1. The buildconfig auxiliary program wasn't quoting the value set for
6300 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6301 not defined. This bug was masked in 4.22 by the effect that was fixed in
6304 2. Some messages that were rejected after a message id was allocated were
6305 shown as "incomplete" by exigrep. It no longer does this for messages that
6306 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6308 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6309 have allow_domain_literals set, the ID did not get logged in the <= line.
6310 Domain literals are now always recognized in Message-ID: header lines.
6312 4. The first argument for a ${extract expansion item is the key name or field
6313 number. Leading and trailing spaces in this item were not being ignored,
6314 causing some misleading effects.
6316 5. When deliver_drop_privilege was set, single queue runner processes started
6317 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6318 same command in the process it spins off) were not dropping privilege.
6320 6. When the daemon running as "exim" started a queue runner, it always
6321 re-executed Exim in the spun-off process. This is a waste of effort when
6322 deliver_drop_privilege is set. The new process now just calls the
6323 queue-runner function directly.
6329 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6332 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6333 ignored. Though the use of -oP was forcing the writing of a pid file, it
6334 was always written to the default place.
6336 3. If the message "no IP address found for host xxxx" is generated during
6337 incoming verification, it is now followed by identification of the incoming
6338 connection (so you can more easily find what provoked it).
6340 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6342 5. Added some features to "harden" Exim a bit more against certain attacks:
6344 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6345 be put in Local/Makefile. This is like the never_users runtime option,
6346 but it cannot be overridden. The default setting is "root".
6348 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6349 prefix string with which any file named in a -C command line option
6352 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6353 is retained for -C and -D only if the caller of Exim is root. Without
6354 it, the exim user may also use -C and -D and retain privilege.
6356 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6357 command line option is disabled.
6359 6. Macro names set by the -D option must start with an upper case letter, just
6360 like macro names defined in the configuration file.
6362 7. Added "dereference=" facility to LDAP.
6364 8. Two instances of the typo "uknown" in the source files are fixed.
6366 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6367 the Configure-Makefile script screwed up while processing it.
6369 10. Incorporated PCRE 4.4.
6371 11. The SMTP synchronization check was not operating right at the start of an
6372 SMTP session. For example, it could not catch a HELO sent before the client
6373 waited for the greeting. There is now a check for outstanding input at the
6374 point when the greeting is written. Because of the duplex, asynchronous
6375 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6376 way, but not yet received, when the check is performed.
6378 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6379 on TCP/IP sockets, because this apparently causes some broken clients to
6382 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6383 unchanged) from the Cygwin maintainer.
6385 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6386 was supported instead of testing for mbx. Effectively a typo.
6388 15. The spa authenticator server code was not checking that the input it
6389 received was valid base64.
6391 16. The debug output line for the "set" modifier in ACLs was not showing the
6392 name of the variable that was being set.
6394 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6396 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6397 lookup on its own. Something else was needed to trigger the lookup. For
6398 example, a match in host_lookup or the need for a host name in a host list.
6399 Now, if $sender_host_name is referenced and the host name has not yet been
6400 looked up, a lookup is performed. If the lookup fails, the variable remains
6401 empty, and $host_lookup_failed is set to "1".
6403 19. Added "eqi" as a case-independent comparison operator.
6405 20. The saslauthd authentication condition could segfault if neither service
6406 nor realm was specified.
6408 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6409 error message that was logged was misleading, and incoming SMTP
6410 connections were dropped. The message is now more accurate, and temporary
6411 errors are given to SMTP connections.
6413 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6414 (except RSET) with a 421 error, until QUIT is received. However, it was
6415 failing to send a response to QUIT.
6417 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6418 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6419 ACL is now run after the helo_try_verify_hosts code.
6421 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6422 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6423 case-independent, so other case variants are also recognized). Apparently
6424 some systems use these upper case variants.
6426 25. If more than two messages were waiting for the same host, and a transport
6427 filter was specified for the transport, Exim sent two messages over the
6428 same TCP/IP connection, and then failed with "socket operation on non-
6429 socket" when it tried to send the third.
6431 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6433 27. The extern definition of crypt16() in expand.c was not being excluded when
6434 the OS had its own crypt16() function.
6436 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6437 as a preferred synonym for return_size_limit, both as an option and as an
6440 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6442 30. Changed the default configuration ACL to relax the local part checking rule
6443 for addresses that are not in any local domains. For these addresses,
6444 slashes and pipe symbols are allowed within local parts, but the sequence
6445 /../ is explicitly forbidden.
6447 31. SPA server authentication was not clearing the challenge buffer before
6450 32. log_message in a "warn" ACL statement was writing to the reject log as
6451 well as to the main log, which contradicts the documentation and doesn't
6452 seem right (because no rejection is happening). So I have stopped it.
6454 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6455 However, I am unable to do any testing of this.
6457 34. Fixed an infelicity in the appendfile transport. When checking directories
6458 for a mailbox, to see if any needed to be created, it was accidentally
6459 using path names with one or more superfluous leading slashes; tracing
6460 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6462 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6463 contents are added to the log line that is written for every discarded
6464 recipient. (Previously a log_message setting was ignored.)
6466 36. The ${quote: operator now quotes the string if it is empty.
6468 37. The install script runs exim in order to find its version number. If for
6469 some reason other than non-existence or emptiness, which it checks, it
6470 could not run './exim', it was installing it with an empty version number,
6471 i.e. as "exim-". This error state is now caught, and the installation is
6474 38. An argument was missing from the function that creates an error message
6475 when Exim fails to connect to the socket for saslauthd authentication.
6476 This could cause Exim to crash, or give a corrupted message.
6478 39. Added isip, isip4, and isip6 to ${if conditions.
6480 40. The ACL variables $acl_xx are now saved with the message, and can be
6481 accessed later in routers, transports, and filters.
6483 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6484 strings in the file are not string-expanded.
6486 42. If a MAIL command specified a SIZE value that was too large to fit into an
6487 int variable, the check against message_size_limit failed. Such values are
6488 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6489 day this will have to be increased, but I don't think I want to be around
6490 when emails are that large.
6497 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6498 iconv() is not standard in FreeBSD.
6500 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6501 IPv6 enabled. The observed symptom was a segmentation fault on return from
6502 the function os_common_find_running_interfaces() in src/os.c.
6504 3. In the check_special_case() function in daemon.c I had used "errno" as an
6505 argument name, which causes warnings on some systems. This was basically a
6506 typo, since it was named "eno" in the comments!
6508 4. The code that waits for the clock to tick (at a resolution of some fraction
6509 of a second) so as to ensure message-id uniqueness was always waiting for
6510 at least one whole tick, when it could have waited for less. [This is
6511 almost certainly not relevant at current processor speeds, where it is
6512 unlikely to ever wait at all. But we try to future-proof.]
6514 5. The function that sleeps for a time interval that includes fractions of a
6515 second contained a race. It did not block SIGALRM between setting the
6516 timer, and suspending (a couple of lines later). If the interval was short
6517 and the sigsuspend() was delayed until after it had expired, the suspension
6518 never ended. On busy systems this could lead to processes getting stuck for
6521 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6522 process, before it forks any delivery processes. The open lookup caching
6523 mechanism meant that the open file or database connection was passed into
6524 the delivery process. The problem was that delivery processes always tidy
6525 up cached lookup data. This could cause a problem for the next delivery
6526 process started by the queue runner, because the external queue runner
6527 process does not know about the closure. So the next delivery process
6528 still has data in the lookup cache. In the case of a file lookup, there was
6529 no problem because closing a file descriptor in a subprocess doesn't affect
6530 the parent. However, if the lookup was caching a connection to a database,
6531 the connection was closed, and the second delivery process was likely to
6532 see errors such as "PGSQL: query failed: server closed the connection
6533 unexpectedly". The problem has been fixed by closing all cached lookups
6534 in a queue runner before running a delivery process.
6536 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6537 seem to have the "const" qualifier which it has on other OS. I've
6540 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6541 *of the same type* (client or server) with the same public name that an
6542 error should be diagnosed.
6544 9. When Exim looked up a host name for an IP address, but failed to find the
6545 original IP address when looking up the host name (a safety check), it
6546 output the message "<ip address> does not match any IP for NULL", which was
6547 confusing, to say the least. The bug was that the host name should have
6548 appeared instead of "NULL".
6550 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6551 user that is built into the binary, and the -C or -D options is used, root
6552 privilege is dropped before the configuration file is read. In addition,
6553 logging is switched to stderr instead of the normal log files. If the
6554 configuration then re-defines the Exim user, the unprivileged environment
6555 is probably not what is expected, so Exim logs a panic warning message (but
6558 However, if deliver_drop_privilege is set, the unprivileged state may well
6559 be exactly what is intended, so the warning has been cut out in that case,
6560 and Exim is allowed to try to write to its normal log files.
6566 1. smtp_return_error_details was not giving details for temporary sender
6567 or receiver verification errors.
6569 2. Diagnose a configuration error if two authenticators have the same public
6572 3. Exim used not to create the message log file for a message until the first
6573 delivery attempt. This could be confusing when incoming messages were held
6574 for policy or load reasons. The message log file is now created at the time
6575 the message is received, and an initial "Received" line is written to it.
6577 4. The automatically generated man page for command line options had a minor
6578 bug that caused no ill effects; however, a more serious problem was that
6579 the procedure for building the man page automatically didn't always
6580 operate. Consequently, release 4.20 contains an out-of-date version. This
6581 shouldn't happen again.
6583 5. When building Exim with embedded Perl support, the script that builds the
6584 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6585 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6587 6. The freeze_tell option was not being used for messages that were frozen on
6588 arrival, either by an ACL or by local_scan().
6590 7. Added the smtp_incomplete_transaction log selector.
6592 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6593 was accepting AUTH without a new EHLO.
6595 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6596 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6599 10. Logging of TCP/IP connections (when configured) now happens in the main
6600 daemon process instead of the child process, so that the TCP/IP connection
6601 count is more accurate (but it can never be perfect).
6603 11. The use of "drop" in a nested ACL was not being handled correctly in the
6604 outer ACL. Now, if condition failure induced by the nested "drop" causes
6605 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6606 or "require"), the connection is dropped.
6608 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6609 that yield "discard" can now be used with an "accept" or a "discard" verb,
6610 but an error is generated for any others (because I can't see a useful way
6611 to define what should happen).
6613 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6614 are now processed in the same way as lines in the Exim configuration file.
6615 In particular, continuation lines are supported.
6617 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6619 15. Added -ti meaning -t -i.
6621 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6622 domains, and warn by logging if others are found.
6624 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6625 returned from GIFCONF when Exim is trying to find the list of interfaces on
6626 a host. The code in os.c has been modified to copy each ifreq to an aligned
6627 structure in all cases.
6629 Also, in some cases, the returned ifreq's were being copied to a 'struct
6630 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6631 means the last couple of bytes of an IPv6 address could be chopped if the
6632 ifreq contained only a normal sockaddr (14 bytes storage).
6634 18. Named domain lists were not supported in the hosts_treat_as_local option.
6635 An entry such as +xxxx was not recognized, and was treated as a literal
6638 19. Ensure that header lines added by a DATA ACL are included in the reject log
6639 if the ACL subsequently rejects the message.
6641 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6642 MD5 (which is deprecated).
6644 21. When testing a filter file using -bf, Exim was writing a message when it
6645 took the sender from a "From " line in the message, but it was not doing so
6646 when it took $return_path from a Return-Path: header line. It now does.
6648 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6649 with a valid header line field name (a series of printing characters
6650 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6652 23. Changed "disc" in the source to "disk" to conform to the documentation and
6653 the book and for uniformity.
6655 24. Ignore Sendmail's -Ooption=value command line item.
6657 25. When execve() failed while trying to run a command in a pipe transport,
6658 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6659 could be confused with a return value of 69 from the command itself. This
6660 has been changed to 127, the value the shell returns if it is asked to run
6661 a non-existent command. The wording for the related log line suggests a
6662 non-existent command as the problem.
6664 26. If received_header_text expands to an empty string, do not add a Received:
6665 header line to the message. (Well, it adds a token one on the spool, but
6666 marks it "old" so that it doesn't get used or transmitted.)
6668 27. Installed eximstats 1.28 (addition of -nt option).
6670 28. There was no check for failure on the call to getsockname() in the daemon
6671 code. This can fail if there is a shortage of resources on the system, with
6672 ENOMEM, for example. A temporary error is now given on failure.
6674 29. Contrary to the C standard, it seems that in some environments, the
6675 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6676 program. Exim now does this explicitly; it affects the formatting of
6677 timestamps using strftime().
6679 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6680 complaints. I've now initialized all the variables, to avoid this.
6682 32. Header lines added by a system filter were not being "seen" during
6683 transport-time rewrites.
6685 33. The info_callback() function passed to OpenSSL is set up with type void
6686 (*)(SSL *, int, int), as described somewhere. However, when calling the
6687 function (actually a macro) that sets it up, the type void(*)() is
6688 expected. I've put in a cast to prevent warnings from picky compilers.
6690 34. If a DNS black list lookup found a CNAME record, but there were no A
6691 records associated with the domain it pointed at, Exim crashed.
6693 35. If a DNS black list lookup returned more than one A record, Exim ignored
6694 all but the first. It now scans all returned addresses if a particular IP
6695 value is being sought. In this situation, the contents of the
6696 $dnslist_value variable are a list of all the addresses, separated by a
6699 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6700 to accept a host name and all its aliases if the forward lookup for any of
6701 them yielded the IP address of the incoming connection. Now it accepts only
6702 those names whose forward lookup yields the correct IP address. Any other
6703 names are discarded. This closes a loophole whereby a rogue DNS
6704 administrator could create reverse DNS records to break through a
6705 wildcarded host restriction in an ACL.
6707 37. If a user filter or a system filter that ran in a subprocess used any of
6708 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6709 the wrong values were passed to the pipe command ($thisaddress had the
6710 value of $0, $0 had the value of $1, etc). This bug was introduced by
6711 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6713 38. Improved the line breaking for long SMTP error messages from ACLs.
6714 Previously, if there was no break point between 40 and 75 characters, Exim
6715 left the rest of the message alone. Two changes have been made: (a) I've
6716 reduced the minimum length to 35 characters; (b) if it can't find a break
6717 point between 35 and 75 characters, it looks ahead and uses the first one
6718 that it finds. This may give the occasional overlong line, but at least the
6719 remaining text gets split now.
6721 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6722 file descriptors might be low, and that setting 1000 would always raise it.
6723 It turns out that in some environments, the limit is already over 1000 and
6724 that lowering it causes trouble. So now Exim takes care not to decrease it.
6726 40. When delivering a message, the value of $return_path is set to $sender_
6727 address at the start of routing (routers may change the value). By an
6728 oversight, this default was not being set up when an address was tested by
6729 -bt or -bv, which affected the outcome if any router or filter referred to
6732 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6733 log only when "message" or "log_message" are set. However, if one of the
6734 conditions was an address verification, or a call to a nested ACL, the
6735 messages generated by the underlying test were being passed through. This
6736 no longer happens. The underlying message is available in $acl_verify_
6737 message for both "message" and "log_message" expansions, so it can be
6738 passed through if needed.
6740 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6741 new expansion $bh_ to give the encoded byte string without charset
6742 translation. Translation happens only if iconv() is available; HAVE_ICONV
6743 indicates this at build time. HEADERS_CHARSET gives the charset to
6744 translate to; headers_charset can change it in the configuration, and
6745 "headers charset" can change it in an individual filter file.
6747 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6748 that creates RFC 2047 encoded "words" labels them as that charset instead
6749 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6750 expansion operator; (ii) when Exim creates a From: line for a local
6751 message; (iii) when a header line is rewritten to include a "phrase" part.
6753 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6754 buggy, causing it to skip the first lines of messages whose message ID
6755 ended in 'D'. This would not have bitten before Exim release 4.14, because
6756 message IDs were unlikely to end in 'D' before then. The effect was to have
6757 incorrect size information for certain domains.
6759 45. #include "config.h" was missing at the start of the crypt16.c module. This
6760 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6763 46. If there was a timeout during a "random" callout check, Exim treated it as
6764 a failure of the random address, and carried on sending RSET and the real
6765 address. If the delay was just some slowness somewhere, the response to the
6766 original RCPT would be taken as a response to RSET and so on, causing
6767 mayhem of various kinds.
6769 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6770 when I implemented it. It didn't allow for the fact that some option values
6771 may legitimately be negative (e.g. size_addition), and it didn't even do
6772 the right test for positive values.
6774 48. Domain names in DNS records are case-independent. Exim always looks them up
6775 in lower case. Some resolvers return domain names in exactly the case they
6776 appear in the zone file, that is, they may contain uppercase letters. Not
6777 all resolvers do this - some return always lower case. Exim was treating a
6778 change of case by a resolver as a change of domain, similar to a widening
6779 of a domain abbreviation. This triggered its re-routing code and so it was
6780 trying to route what was effectively the same domain again. This normally
6781 caused routing to fail (because the router wouldn't handle the domain
6782 twice). Now Exim checks for this case specially, and just changes the
6783 casing of the domain that it ultimately uses when it transmits the message
6786 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6789 50. If a filter generated a file delivery with a non-absolute name (possible if
6790 no home directory exists for the router), the forbid_file option was not
6793 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6794 the existing equality matching.
6796 52. Exim was using ints instead of ino_t variables in some places where it was
6797 dealing with inode numbers.
6799 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6800 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6801 called TMPDIR, and if it finds it is different, it changes its value.
6803 54. The smtp_printf() function is now made available to local_scan() so
6804 additional output lines can be written before returning. There is also an
6805 smtp_fflush() function to enable the detection of a dropped connection.
6806 The variables smtp_input and smtp_batched_input are exported to
6809 55. Changed the default runtime configuration: the message "Unknown user"
6810 has been removed from the ACL, and instead placed on the localuser router,
6811 using the cannot_route_message feature. This means that any verification
6812 failures that generate their own messages won't get overridden. Similarly,
6813 the "Unrouteable address" message that was in the ACL for unverifiable
6814 relay addresses has also been removed.
6816 56. Added hosts_avoid_esmtp to the smtp transport.
6818 57. The exicyclog script was not checking for the esoteric option
6819 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6820 will work only if exicyclog is run under the appropriate euid.
6822 58. Following a discussion on the list, the rules by which Exim recognises line
6823 endings on incoming messages have been changed. The -dropcr and drop_cr
6824 options are now no-ops, retained only for backwards compatibility. The
6825 following line terminators are recognized: LF CRLF CR. However, special
6826 processing applies to CR:
6828 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6829 nor a local message in the state where . is a terminator.
6831 (ii) If a bare CR is encountered in a header line, an extra space is added
6832 after the line terminator so as not to end the header. The reasoning
6833 behind this is that bare CRs in header lines are most likely either
6834 to be mistakes, or people trying to play silly games.
6836 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6837 was being incorrectly given as 18 bytes larger than it should have been.
6838 This is a VOB (very old bug).
6840 60. This may never have affected anything current, but just in case it has:
6841 When the local host is found other than at the start of a list of hosts,
6842 the local host, those with the same MX, and any that follow, are discarded.
6843 When the list in question was part of a longer list of hosts, the following
6844 hosts (not currently being processed) were also being discarded. This no
6845 longer happens. I'm not sure if this situation could ever has previously
6848 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6851 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6852 header. This is something that is recommended in a new Internet Draft, and
6853 is something that is documented as being done by Sendmail. There are two
6854 possible values. For messages generated by the autoreply transport, Exim
6857 Auto-Submitted: auto-replied
6859 whereas for all other generated messages (e.g. bounces) it adds
6861 Auto-Submitted: auto-generated
6863 63. The "personal" condition in filters now includes a test for the
6864 Auto-Submitted: header. If it contains the string "auto-" the message it
6865 not considered personal.
6867 64. Added rcpt_include_affixes as a generic transport option.
6869 65. Added queue_only_override (default true).
6871 66. Added the syslog_duplication option.
6873 67. If what should have been the first header line of a message consisted of
6874 a space followed by a colon, Exim was mis-interpreting it as a header line.
6875 It isn't of course - it is syntactically invalid and should therefore be
6876 treated as the start of the message body. The misbehaviour could have
6877 caused a number of strange effects, including loss of data in subsequent
6878 header lines, and spool format errors.
6880 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6881 client host had authenticated. This control can now be exercised by an ACL
6882 for more flexibility.
6884 69. By default, callouts do not happen when testing with -bh. There is now a
6885 variant, -bhc, which does actually run the callout code, including
6886 consulting and updating the callout cache.
6888 70. Added support for saslauthd authentication, courtesy of Alexander
6891 71. If statvfs() failed on the spool or log directories while checking their
6892 size for availability, Exim confusingly gave the error "space shortage".
6893 Furthermore, in debugging mode it crashed with a floating point exception.
6894 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6895 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6896 serious problem, Exim now writes to the main and panic logs when this
6897 happens, with details of the failure. It then refuses to accept the
6898 incoming message, giving the message "spool directory problem" or "log
6899 directory problem" with a 421 code for SMTP messages.
6901 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6902 0, 1, and 2 exist, because some OS complain for execs without them (see
6903 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6904 descriptors. However, the code omitted to check that the open succeeded,
6905 causing mysterious errors if for some reason the permissions on /dev/null
6906 got screwed. Now Exim writes a message to the main and panic logs, and
6907 bombs out if it can't open /dev/null.
6909 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6910 interact so that it is all more flexible. It is supposed to remain
6911 backwards compatible. Also added extra_local_interfaces.
6913 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6914 to bomb out with an assertion failure - to the client this appears as a
6915 connection drop. This problem occurs in the part of the code that was taken
6916 from the Samba project. Fortunately, the assertion is in a very simple
6917 function, so I have fixed this by reproducing the function inline in the
6918 one place where it is called, and arranging for authentication to fail
6919 instead of killing the process with assert().
6921 75. The SPA client code was not working when the server requested OEM rather
6922 than Unicode encoding.
6924 76. Added code to make require_files with a specific uid setting more usable in
6925 the case where statting the file as root fails - usually a non-root-mounted
6926 NFS file system. When this happens and the failure is EACCES, Exim now
6927 forks a subprocess and does the per-uid checking as the relevant uid.
6929 77. Added process_log_path.
6931 78. If log_file_path was not explicitly set, a setting of check_log_space or
6932 check_log_inodes was ignored.
6934 79. If a space check for the spool or log partitions fails, the incident is now
6935 logged. Of course, in the latter case the data may get lost...
6937 80. Added the %p formatting code to string_format() so that it can be used to
6938 print addresses in debug_print(). Adjusted all the address printing in the
6939 debugging in store.c to use %p rather than %d.
6941 81. There was a concern that a line of code in smtp_in.c could overflow a
6942 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6943 initially expressed, the concern was not well-founded, because trailing
6944 spaces are removed early. However, if the trailing spaces were followed by
6945 a NULL, they did not get removed, so the overflow was possible. Two fixes
6948 (a) I re-wrote the offending code in a cleaner fashion.
6949 (b) If an incoming SMTP command contains a NULL character, it is rejected
6952 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6953 runs initgroups(), so that if the Exim user is in any additional groups,
6954 they will be used during message reception.
6960 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob