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 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
53 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
54 by the template makefile "src/EDITME".
60 JH/01 Rework error string handling in TLS interface so that the caller in
61 more cases is responsible for logging. This permits library-sourced
62 string to be attached to addresses during delivery, and collapses
63 pairs of long lines into single ones.
65 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
66 during configuration. Wildcards are allowed and expanded.
68 JH/02 Rework error string handling in DKIM to pass more info back to callers.
69 This permits better logging.
71 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
72 do not close it down and have the child transport start it up again on
73 the passed-on TCP connection. Instead, proxy the child (and any
74 subsequent ones) for TLS via a unix-domain socket channel. Logging is
75 affected: the continued delivery log lines do not have any DNSSEC, TLS
76 Certificate or OCSP information. TLS cipher information is still logged.
78 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
79 identical IP addresses on different listening ports. Will also affect
82 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
83 add noisy ifdef guards to special-case this sillyness.
84 Patch from Bernd Kuhls.
86 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
87 than 255 are no longer allowed.
89 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
90 Disable the session-cache too, which might reduce our load. Since we
91 currrectly use a new context for every connection, both as server and
92 client, there is no benefit for these.
93 GnuTLS appears to not support tickets server-side by default (we don't
94 call gnutls_session_ticket_enable_server()) but client side is enabled
95 by default on recent versions (3.1.3 +) unless the PFS priority string
98 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
99 <https://reproducible-builds.org/specs/source-date-epoch/>.
101 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
102 the check for any unsuccessful recipients did not notice the limit, and
103 erroneously found still-pending ones.
105 JH/08 Pipeline CHUNKING command and data together, on kernels that support
106 MSG_MORE. Only in-clear (not on TLS connections).
108 JH/09 Avoid using a temporary file during transport using dkim. Unless a
109 transport-filter is involved we can buffer the headers in memory for
110 creating the signature, and read the spool data file once for the
111 signature and again for transmission.
113 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
114 4.77 as the kernel support then wasn't solid, having issues in 64bit
115 mode. Now, it's been long enough. Add support for FreeBSD also.
117 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
118 case where the routing stage had gathered several addresses to send to
119 a host before calling the transport for the first, we previously failed
120 to close down TLS in the old transport process before passing the TCP
121 connection to the new process. The new one sent a STARTTLS command
122 which naturally failed, giving a failed delivery and bloating the retry
123 database. Investigation and fix prototype from Wolfgang Breyha.
125 JH/12 Fix check on SMTP command input synchronisation. Previously there were
126 false-negatives in the check that the sender had not preempted a response
127 or prompt from Exim (running as a server), due to that code's lack of
128 awareness of the SMTP input buffering.
130 PP/04 Add commandline_checks_require_admin option.
131 Exim drops privileges sanely, various checks such as -be aren't a
132 security problem, as long as you trust local users with access to their
133 own account. When invoked by services which pass untrusted data to
134 Exim, this might be an issue. Set this option in main configuration
135 AND make fixes to the calling application, such as using `--` to stop
138 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
139 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
140 responses to those, into a single TLS record each way (this usually means
141 a single packet). As a side issue, smtp_enforce_sync now works on TLS
144 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
145 affects you only if you're dancing at the edge of the param size limits.
146 If you are, and this message makes sense to you, then: raise the
147 configured limit or use OpenSSL 1.1. Nothing we can do for older
150 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
151 cmdline element to get the documented default one. Previously it was
154 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
155 are unsuable for non-message-specific reasons - eg. connection timeout,
156 banner-time rejection.
158 JH/16 Fix logging of delivery remote port, when specified by router, under
161 PP/06 Repair manualroute's ability to take options in any order, even if one
162 is the name of a transport.
165 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
167 JH/17 Change the list-building routines interface to use the expanding-string
168 triplet model, for better allocation and copying behaviour.
170 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
171 Previously it was constructed the first time a possibly-matching string
172 was met in the configuration file input during startup; now it is done
175 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
176 compatible one, to avoid the (poorly documented) possibility of a config
177 file in the working directory redirecting the DB files, possibly correpting
178 some existing file. CVE-2017-10140 assigned for BDB.
180 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
181 cache-hot. Previously, although the result was properly cached, the
182 initial verify call returned a defer.
184 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
185 the main verify for receipient in uncached-mode.
187 JH/22 Retire historical build files to an "unsupported" subdir. These are
188 defined as "ones for which we have no current evidence of testing".
190 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
191 if present. Previously it was ignored.
193 JH/24 Start using specified-initialisers in C structure init coding. This is
194 a C99 feature (it's 2017, so now considered safe).
196 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
197 if was a fixed-sized field and bitmask ops via macros; it is now more
200 PP/07 GitHub PR 56: Apply MariaDB build fix.
201 Patch provided by Jaroslav Škarvada.
203 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
204 during Coverity cleanups [4.87 JH/47]
205 Diagnosis and fix provided by Michael Fischer v. Mollard.
207 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
208 the right size to place the terminating semicolon on its own folded
209 line, the header hash was calculated to an incorrect value thanks to
210 the (relaxed) space the fold became.
212 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
213 and confused the parent.
215 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
216 which could crash as a result. This could lead to undeliverable messages.
218 JH/28 Logging: "next input sent too soon" now shows where input was truncated
221 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
222 matters on fast-turnover and PID-randomising systems, which were getting
223 out-of-order delivery.
225 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
226 a possibly-overlapping copy. The symptom was that "Remote host closed
227 connection in response to HELO" was logged instead of the actual 4xx
230 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
231 Previously only that bufferd was discarded, resulting in SYMTP command
234 JH/32 DKIM: when a message has multiple signatures matching an identity given
235 in dkim_verify_signers, run the dkim acl once for each. Previously only
236 one run was done. Bug 2189.
238 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
239 "panic the current process" to "deliberately defer". The panic log is
240 still written with the problem list name; the mail and reject logs now
241 get a temp-reject line for the message that was being handled, saying
242 something like "domains check lookup or other defer". The SMTP 451
243 message is still "Temporary local problem".
245 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
246 A crafted sequence of BDAT commands could result in in-use memory beeing
247 freed. CVE-2017-16943.
249 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
250 from SMTP input. Previously it was always done; now only done for DATA
251 and not BDAT commands. CVE-2017-16944.
253 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
254 to the message (such as an overlong header line). Previously this was
255 not done and we did not exit BDAT mode. Followon from the previous item
256 though a different problem.
262 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
263 than -2003 did; needs libidn2 in addition to libidn.
265 JH/02 The path option on a pipe transport is now expanded before use.
267 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
268 Patch provided by "Björn", documentation fix added too.
270 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
271 missing a wire-to-host endian conversion.
273 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
274 close after a BDAT command line could be taken as a following command,
275 giving a synch failure. Fix by only checking for synch immediately
276 before acknowledging the chunk.
278 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
279 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
280 macro. Patches provided by Josh Soref.
282 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
283 Previously we did not; the RFC seems ambiguous and VRFY is not listed
284 by IANA as a service extension. However, John Klensin suggests that we
287 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
288 the dkim code may be unix-mode line endings rather than smtp wire-format
289 CRLF, so prepend a CR to any bare LF.
291 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
292 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
294 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
295 the first were themselves being wrongly included in the feed into dkim
296 processing; with most chunk sizes in use this resulted in an incorrect
297 body hash calculated value.
299 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
300 DKIM signature block, for verification. Although advised against by
301 standards it is specifically not ruled illegal.
303 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
305 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
306 missing a body hash (the bh= tag).
308 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
309 It seems that HAProxy sends the Proxy Protocol information in clear and
310 only then does a TLS startup, so do the same.
312 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
313 TCP connections (such as for Spamd) unless the daemon successfully set
314 Fast Open mode on its listening sockets. This fixes breakage seen on
315 too-old kernels or those not configured for Fast Open, at the cost of
316 requiring both directions being enabled for TFO, and TFO never being used
317 by non-daemon-related Exim processes.
319 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
320 endings, at least on the first header line. Try to canonify any that get
321 past that check, despite the cost.
323 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
324 now limited to an arbitrary five deep, while parsing addresses with the
325 strip_excess_angle_brackets option enabled.
327 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
328 instead leave the unprompted TLS handshake in socket buffer for the
329 TLS library to consume.
331 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
333 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
335 JH/16 Drop variables when they go out of scope. Memory management drops a whole
336 region in one operation, for speed, and this leaves assigned pointers
337 dangling. Add checks run only under the testsuite which checks all
338 variables at a store-reset and panics on a dangling pointer; add code
339 explicitly nulling out all the variables discovered. Fixes one known
340 bug: a transport crash, where a dangling pointer for $sending_ip_address
341 originally assigned in a verify callout, is re-used.
343 PP/06 Drop '.' from @INC in various Perl scripts.
345 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
347 PP/08 Reduce a number of compilation warnings under clang; building with
348 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
349 should be warning-free.
351 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
353 HS/01 Fix portability problems introduced by PP/08 for platforms where
354 realloc(NULL) is not equivalent to malloc() [SunOS et al].
356 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
357 chunk. This allows us to accept broken chunked messages. We need a more
358 general solution here.
360 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
361 already-broken messages in the queue.
363 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
365 JH/19 Fix reference counting bug in routing-generated-address tracking.
371 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
372 supports it and a size is available (ie. the sending peer gave us one).
374 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
375 years of being deprecated). The replacements are the ACLs
376 acl_smtp_mime and acl_not_smtp_mime.
378 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
379 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
380 or even an in-clear connection were permitted. Now, if the host lookup
381 was dnssec and dane was requested then the host is only used if the
382 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
383 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
384 if one fails this test.
385 This means that a poorly-configured remote DNS will make it incommunicado;
386 but it protects against a DNS-interception attack on it.
388 JH/04 Bug 1810: make continued-use of an open smtp transport connection
389 non-noisy when a race steals the message being considered.
391 JH/05 If main configuration option tls_certificate is unset, generate a
392 self-signed certificate for inbound TLS connections.
394 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
395 in rewrites and routers.
397 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
398 and logged a warning sing 4.83; now they are a configuration file error.
400 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
401 (lacking @domain). Apply the same qualification processing as RCPT.
403 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
405 JH/10 Support ${sha256:} applied to a string (as well as the previous
408 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
409 a cutthrough deliver is pending, as we always want to make a connection.
410 This also avoids re-routing the message when later placing the cutthrough
411 connection after a verify cache hit.
412 Do not update it with the verify result either.
414 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
415 when routing results in more than one destination address.
417 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
418 signing (which inhibits the cutthrough capability). Previously only
419 the presence of an option was tested; now an expansion evaluating as
420 empty is permissible (obviously it should depend only on data available
421 when the cutthrough connection is made).
423 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
424 the relevant preceding SMTP command did not note the pipelining mode.
426 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
427 Previously they were not counted.
429 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
430 as one having no matching records. Previously we deferred the message
431 that needed the lookup.
433 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
434 distinguished as "(=".
436 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
437 for missing MX records. Previously it only worked for missing A records.
439 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
441 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
442 after the data-go-ahead and data-ack. Patch from Jason Betts.
444 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
445 even for a "none" policy. Patch from Tony Meyer.
447 JH/22 Fix continued use of a connection for further deliveries. If a port was
448 specified by a router, it must also match for the delivery to be
451 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
452 When one of the recipients of a message was unsuitable for the connection
453 (has no matching addresses), we lost track of needing to mark it
454 deferred. As a result mail would be lost.
456 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
458 JH/25 Decoding ACL controls is now done using a binary search; the source code
459 takes up less space and should be simpler to maintain. Merge the ACL
460 condition decode tables also, with similar effect.
462 JH/26 Fix problem with one_time used on a redirect router which returned the
463 parent address unchanged. A retry would see the parent address marked as
464 delivered, so not attempt the (identical) child. As a result mail would
467 JH/27 Fix a possible security hole, wherein a process operating with the Exim
468 UID can gain a root shell. Credit to http://www.halfdog.net/ for
469 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
472 JH/28 Enable {spool,log} filesystem space and inode checks as default.
473 Main config options check_{log,spool}_{inodes,space} are now
474 100 inodes, 10MB unless set otherwise in the configuration.
476 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
477 Previously it only applied to the main-section connection policy
480 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
482 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
483 by me. Added RFC7919 DH primes as an alternative.
485 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
486 are not in the system include path.
488 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
489 GnuTLS, when a session startup failed (eg because the client disconnected)
490 Exim did stdio operations after fclose. This was exposed by a recent
491 change which nulled out the file handle after the fclose.
493 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
494 signed directly by the cert-signing cert, rather than an intermediate
495 OCSP-signing cert. This is the model used by LetsEncrypt.
497 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
499 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
500 an incoming connection.
502 HS/02 Bug 1802: Do not half-close the connection after sending a request
505 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
506 fallback to "prime256v1".
508 JH/34 SECURITY: Use proper copy of DATA command in error message.
509 Could leak key material. Remotely exploitable. CVE-2016-9963.
515 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
516 and 3.4.4 - once the server is enabled to respond to an OCSP request
517 it does even when not requested, resulting in a stapling non-aware
518 client dropping the TLS connection.
520 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
521 support variable-length bit vectors. No functional change.
523 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
524 The I= interface field on outgoing lines is now after the H= remote
525 host field, same as incoming lines. There is a separate
526 outgoing_interface log selector which allows you to disable the
529 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
530 If not running log_selector +smtp_connection the mainlog would be held
531 open indefinitely after a "too many connections" event, including to a
532 deleted file after a log rotate. Leave the per net connection logging
533 leaving it open for efficiency as that will be quickly detected by the
534 check on the next write.
536 HS/01 Bug 1671: Fix post transport crash.
537 Processing the wait-<transport> messages could crash the delivery
538 process if the message IDs didn't exist for some reason. When
539 using 'split_spool_directory=yes' the construction of the spool
540 file name failed already, exposing the same netto behaviour.
542 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
543 mime_regex ACL conditions.
545 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
546 to DSN fail messages (bounces): remote IP, remote greeting, remote response
547 to HELO, local diagnostic string.
549 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
550 log line to debug. Even when configured with a tls authenticator many
551 client connections are expected to not authenticate in this way, so
552 an authenticate fail is not an error.
554 HS/02 Add the Exim version string to the process info. This way exiwhat
555 gives some more detail about the running daemon.
557 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
558 matter for fast-change records such as DNSBLs.
560 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
561 retry record, even if constant. There may be multiple transports with
562 different interface settings and the retry behaviour needs to be kept
565 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
567 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
569 JH/10 Bug 840: fix log_defer_output option of pipe transport
571 JH/11 Bug 830: use same host for all RCPTS of a message, even under
572 hosts_randomize. This matters a lot when combined with mua_wrapper.
574 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
575 ${quote_pgsql:<string>} operator.
577 JH/13 Bug 1708: avoid misaligned access in cached lookup.
579 JH/14 Change header file name for freeradius-client. Relevant if compiling
580 with Radius support; from the Gentoo tree and checked under Fedora.
582 JH/15 Bug 1712: Introduce $prdr_requested flag variable
584 JH/16 Bug 1714: Permit an empty string as expansion result for transport
585 option transport_filter, meaning no filtering.
587 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
589 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
590 defaults to "*" (all hosts). The variable is now available when not built
591 with TLS, default unset, mainly to enable keeping the testsuite sane.
592 If a server certificate is not supplied (via tls_certificate) an error is
593 logged, and clients will find TLS connections fail on startup. Presumably
594 they will retry in-clear.
595 Packagers of Exim are strongly encouraged to create a server certificate
596 at installation time.
598 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
599 with the $config_file variable.
601 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
602 in transport context, after the attempt, and per-recipient. The latter type
603 is per host attempted. The event data is the error message, and the errno
604 information encodes the lookup type (A vs. MX) used for the (first) host,
605 and the trailing two digits of the smtp 4xx response.
607 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
608 to write to mainlog (or rejectlog, paniclog) in the window between file
609 creation and permissions/ownership being changed. Particularly affects
610 installations where exicyclog is run as root, rather than exim user;
611 result is that the running daemon panics and dies.
613 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
615 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
616 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
617 "pri" and "weight". Note that the previous implicit priority given by the
618 list order is no longer honoured.
620 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
623 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
624 by defining SUPPORT_SOCKS.
626 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
627 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
628 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
629 variables are renamed to proxy_{local,external}_{address,port}.
631 JH/27 Move Internationalisation support from Experimental to mainline, enabled
632 for a build by defining SUPPORT_I18N
634 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
635 of the query string, and make ${quote_redis:} do that quoting.
637 JH/29 Move Events support from Experimental to mainline, enabled by default
638 and removable for a build by defining DISABLE_EVENT.
640 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
642 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
643 cached by the daemon.
645 JH/32 Move Redis support from Experimental to mainline, enabled for a build
646 by defining LOOKUP_REDIS. The libhiredis library is required.
648 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
649 keys are given for lookup.
651 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
652 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
653 only supported when built with TLS support. The PolarSSL SHA routines
654 are still used when the TLS library is too old for convenient support.
656 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
657 openssl_options), for security. OpenSSL forces this from version 1.1.0
658 server-side so match that on older versions.
660 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
661 allocation for $value could be released as the expansion processing
662 concluded, but leaving the global pointer active for it.
664 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
665 and to use the domains and local_parts ACL conditions.
667 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
668 incorrectly not doubled on cutthrough transmission, hence seen as a
669 body-termination at the receiving system - resulting in truncated mails.
670 Commonly the sender saw a TCP-level error, and retransmitted the message
671 via the normal store-and-forward channel. This could result in duplicates
672 received - but deduplicating mailstores were liable to retain only the
673 initial truncated version.
675 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
677 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
679 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
680 we're in there, support oversigning also; bug 1309.
682 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
684 HS/04 Add support for keep_environment and add_environment options.
686 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
687 either intentional arithmetic overflow during PRNG, or testing config-
690 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
691 delivery resulted in actual delivery. Cancel cutthrough before DATA
694 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
695 rejects a recipient: pass the reject to the originator.
697 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
698 Many were false-positives and ignorable, but it's worth fixing the
701 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
702 for the new environment-manipulation done at startup. Move the routines
703 from being local to tls.c to being global via the os.c file.
705 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
706 an extract embedded as result-arg for a map, the first arg for extract
707 is unavailable so we cannot tell if this is a numbered or keyed
708 extraction. Accept either.
714 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
717 JH/02 The smtp transport option "multi_domain" is now expanded.
719 JH/03 The smtp transport now requests PRDR by default, if the server offers
722 JH/04 Certificate name checking on server certificates, when exim is a client,
723 is now done by default. The transport option tls_verify_cert_hostnames
724 can be used to disable this per-host. The build option
725 EXPERIMENTAL_CERTNAMES is withdrawn.
727 JH/05 The value of the tls_verify_certificates smtp transport and main options
728 default to the word "system" to access the system default CA bundle.
729 For GnuTLS, only version 3.0.20 or later.
731 JH/06 Verification of the server certificate for a TLS connection is now tried
732 (but not required) by default. The verification status is now logged by
733 default, for both outbound TLS and client-certificate supplying inbound
736 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
739 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
740 Status Notification (bounce) messages are now MIME format per RFC 3464.
741 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
742 under the control of the dsn_advertise_hosts option, and routers may
743 have a dsn_lasthop option.
745 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
746 default, modifiable by a malware= option. The list separator for
747 the options can now be changed in the usual way. Bug 68.
749 JH/10 The smtp_receive_timeout main option is now expanded before use.
751 JH/11 The incoming_interface log option now also enables logging of the
752 local interface on delivery outgoing connections.
754 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
755 if the interface and destination host and port all match.
757 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
760 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
761 Patch from Andrew Lewis.
763 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
764 now supports optional time-restrictions, weighting, and priority
765 modifiers per server. Patch originally by <rommer@active.by>.
767 JH/16 The spamd_address main option now supports a mixed list of local
768 and remote servers. Remote servers can be IPv6 addresses, and
769 specify a port-range.
771 JH/17 Bug 68: The spamd_address main option now supports an optional
772 timeout value per server.
774 JH/18 Bug 1581: Router and transport options headers_add/remove can
775 now have the list separator specified.
777 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
780 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
783 JH/21 Support for the A6 type of dns record is withdrawn.
785 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
786 rather than the verbs used.
788 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
789 from 255 to 1024 chars.
791 JH/24 Verification callouts now attempt to use TLS by default.
793 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
794 are generic router options now. The defaults didn't change.
796 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
797 Original patch from Alexander Shikoff, worked over by JH.
799 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
800 files if ZCAT_COMMAND is not executable.
802 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
804 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
806 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
807 Normally benign, it bites when the pair was led to by a CNAME;
808 modern usage is to not canonicalize the domain to a CNAME target
809 (and we were inconsistent anyway for A-only vs AAAA+A).
811 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
813 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
814 when evaluating $sender_host_dnssec.
816 JH/31 Check the HELO verification lookup for DNSSEC, adding new
817 $sender_helo_dnssec variable.
819 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
821 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
823 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
825 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
826 documented as working, but never had. Support all but $spam_report.
828 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
829 added for tls authenticator.
831 HS/03 Add perl_taintmode main config option
837 TL/01 When running the test suite, the README says that variables such as
838 no_msglog_check are global and can be placed anywhere in a specific
839 test's script, however it was observed that placement needed to be near
840 the beginning for it to behave that way. Changed the runtest perl
841 script to read through the entire script once to detect and set these
842 variables, reset to the beginning of the script, and then run through
843 the script parsing/test process like normal.
845 TL/02 The BSD's have an arc4random API. One of the functions to induce
846 adding randomness was arc4random_stir(), but it has been removed in
847 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
848 function when detected.
850 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
851 cause callback expansion.
853 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
854 syntax errors in an expansion can be treated as a string instead of
855 logging or causing an error, due to the internal use of bool_lax
856 instead of bool when processing it.
858 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
859 server certificates when making smtp deliveries.
861 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
863 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
865 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
867 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
868 Merged patch from Sebastian Wiedenroth.
870 JH/05 Fix results-pipe from transport process. Several recipients, combined
871 with certificate use, exposed issues where response data items split
872 over buffer boundaries were not parsed properly. This eventually
873 resulted in duplicates being sent. This issue only became common enough
874 to notice due to the introduction of connection certificate information,
875 the item size being so much larger. Found and fixed by Wolfgang Breyha.
877 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
878 size buffer was used, resulting in syntax errors when an expansion
881 JH/07 Add support for directories of certificates when compiled with a GnuTLS
882 version 3.3.6 or later.
884 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
885 is EXPERIMENTAL_EVENT, the main-configuration and transport options
886 both become "event_action", the variables become $event_name, $event_data
887 and $event_defer_errno. There is a new variable $verify_mode, usable in
888 routers, transports and related events. The tls:cert event is now also
889 raised for inbound connections, if the main configuration event_action
892 TL/06 In test suite, disable OCSP for old versions of openssl which contained
893 early OCSP support, but no stapling (appears to be less than 1.0.0).
895 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
896 server certificate names available under the smtp transport option
897 "tls_verify_cert_hostname" now do not permit multi-component wildcard
900 JH/10 Time-related extraction expansions from certificates now use the main
901 option "timezone" setting for output formatting, and are consistent
902 between OpenSSL and GnuTLS compilations. Bug 1541.
904 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
905 encoded parameter in the incoming message. Bug 1558.
907 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
908 include certificate info, eximon was claiming there were spoolfile
911 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
913 JH/14 Log delivery-related information more consistently, using the sequence
914 "H=<name> [<ip>]" wherever possible.
916 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
917 are problematic for Debian distribution, omit them from the release
920 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
922 JH/16 Fix string representation of time values on 64bit time_t architectures.
925 JH/17 Fix a null-indirection in certextract expansions when a nondefault
926 output list separator was used.
931 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
932 checkers that were complaining about end of non-void function with no
935 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
936 This was a regression introduced in 4.83 by another bugfix.
938 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
940 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
941 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
947 TF/01 Correctly close the server side of TLS when forking for delivery.
949 When a message was received over SMTP with TLS, Exim failed to clear up
950 the incoming connection properly after forking off the child process to
951 deliver the message. In some situations the subsequent outgoing
952 delivery connection happened to have the same fd number as the incoming
953 connection previously had. Exim would try to use TLS and fail, logging
954 a "Bad file descriptor" error.
956 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
957 utilities have not been installed.
959 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
960 temporary space as the ACL may create new global variables.
962 TL/01 LDAP support uses per connection or global context settings, depending
963 upon the detected version of the libraries at build time.
965 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
966 to extract and use the src ip:port in logging and expansions as if it
967 were a direct connection from the outside internet. PPv2 support was
968 updated based on HAProxy spec change in May 2014.
970 JH/02 Add ${listextract {number}{list}{success}{fail}}.
972 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
973 Properly escape header and check for NULL return.
975 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
978 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
980 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
981 characters in header names, implemented as a verify condition.
982 Contributed by Michael Fischer v. Mollard.
984 TL/05 Rename SPF condition results err_perm and err_temp to standardized
985 results permerror and temperror. Previous values are deprecated but
986 still accepted. In a future release, err_perm and err_temp will be
987 completely removed, which will be a backward incompatibility if the
988 ACL tests for either of these two old results. Patch contributed by
989 user bes-internal on the mailing list.
991 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
993 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
994 selectors, in both main and reject logs.
996 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
999 JH/07 Add malware type "sock" for talking to simple daemon.
1001 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1003 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1004 routers/transports under cutthrough routing.
1006 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1007 numbers. Touch up "bool" conditional to keep the same definition.
1009 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1011 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1013 JH/12 Expand items in router/transport headers_add or headers_remove lists
1014 individually rather than the list as a whole. Bug 1452.
1016 Required for reasonable handling of multiple headers_ options when
1017 they may be empty; requires that headers_remove items with embedded
1018 colons must have them doubled (or the list-separator changed).
1020 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1021 view the policy declared in the DMARC record. Currently, $dmarc_status
1022 is a combined value of both the record presence and the result of the
1025 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1027 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1028 dnslookup router and the smtp transport (applying to the forward
1031 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1032 of ldap servers used for a specific lookup. Patch provided by Heiko
1035 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1036 New variable $lookup_dnssec_authenticated for observability.
1038 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1039 Patch submitted by Lars Timman.
1041 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1043 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1044 Requires trusted mode and valid format message id, aborts otherwise.
1045 Patch contributed by Heiko Schlichting.
1047 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1048 certextract with support for various fields. Bug 1358.
1050 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1051 is requested by default, modifiable by smtp transport option
1054 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1055 operate on certificate variables to give certificate fingerprints
1056 Also new ${sha256:cert_variable}.
1058 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1060 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1061 Christian Aistleitner.
1063 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1065 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1066 file. Patch from Wolfgang Breyha.
1068 JH/25 Expand the coverage of the delivery $host and $host_address to
1069 client authenticators run in verify callout. Bug 1476.
1071 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1072 align with daemon_smtp_ports. Bug 72.
1074 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1075 support and error reporting did not work properly.
1077 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1078 and is readable. Patch from Andrew Colin Kissa.
1080 TL/14 Enhance documentation of ${run expansion and how it parses the
1081 commandline after expansion, particularly in the case when an
1082 unquoted variable expansion results in an empty value.
1084 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1086 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1087 by renaming away from it. Addresses GH issue 15, reported by
1090 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1092 TL/15 SECURITY: prevent double expansion in math comparison functions
1093 (can expand unsanitized data). Not remotely exploitable.
1100 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1102 PP/02 Make -n do something, by making it not do something.
1103 When combined with -bP, the name of an option is not output.
1105 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1108 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1109 $sender_host_name and config options to manage this, and basic check
1112 PP/05 DSCP support for outbound connections and control modifier for inbound.
1114 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1115 (Only plugin which currently uses this is kerberos4, which nobody should
1116 be using, but we should make it available and other future plugins might
1117 conceivably use it, even though it would break NAT; stuff *should* be
1118 using channel bindings instead).
1120 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1121 name; added for Sendmail compatibility; requires admin caller.
1122 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1123 just ignore it); requires trusted caller.
1124 Also parse but ignore: -Ac -Am -X<logfile>
1127 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1129 TL/02 Add +smtp_confirmation as a default logging option.
1131 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1132 Patch by Magnus Holmgren from 2007-02-20.
1134 TL/04 Bugzilla 1281 - Spec typo.
1135 Bugzilla 1283 - Spec typo.
1136 Bugzilla 1290 - Spec grammar fixes.
1138 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1140 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1142 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1143 mailing list by by Dmitry Isaikin.
1145 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1147 JH/02 Support "G" suffix to numbers in ${if comparisons.
1149 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1151 NM/01 Bugzilla 1197 - Spec typo
1152 Bugzilla 1196 - Spec examples corrections
1154 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1156 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1157 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1160 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1161 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1163 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1164 "acl {{name}{arg}...}", and optional args on acl condition
1167 JH/05 Permit multiple router/transport headers_add/remove lines.
1169 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1171 JH/07 Avoid using a waiting database for a single-message-only transport.
1172 Performance patch from Paul Fisher. Bugzilla 1262.
1174 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1177 JH/09 Add $headers_added variable, with content from use of ACL modifier
1178 add_header (but not yet added to the message). Bugzilla 199.
1180 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1181 Pulled from Bugzilla 817 by Wolfgang Breyha.
1183 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1185 (nb: this is the same fix as in Exim 4.80.1)
1187 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1190 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1193 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1195 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1196 advertise SMTP AUTH mechanism to us, instead of a generic
1197 protocol violation error. Also, make Exim more robust to bad
1198 data from the Dovecot auth socket.
1200 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1202 When a queue runner is handling a message, Exim first routes the
1203 recipient addresses, during which it prunes them based on the retry
1204 hints database. After that it attempts to deliver the message to
1205 any remaining recipients. It then updates the hints database using
1208 So if a recipient address works intermittently, it can get repeatedly
1209 deferred at routing time. The retry hints record remains fresh so the
1210 address never reaches the final cutoff time.
1212 This is a fairly common occurrence when a user is bumping up against
1213 their storage quota. Exim had some logic in its local delivery code
1214 to deal with this. However it did not apply to per-recipient defers
1215 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1217 This change adds a proper retry rule check during routing so that the
1218 final cutoff time is checked against the message's age. We only do
1219 this check if there is an address retry record and there is not a
1220 domain retry record; this implies that previous attempts to handle
1221 the address had the retry_use_local_parts option turned on. We use
1222 this as an approximation for the destination being like a local
1223 delivery, as in LMTP.
1225 I suspect this new check makes the old local delivery cutoff check
1226 redundant, but I have not verified this so I left the code in place.
1228 TF/02 Correct gecos expansion when From: is a prefix of the username.
1230 Test 0254 submits a message to Exim with the header
1234 When I ran the test suite under the user fanf2, Exim expanded
1235 the header to contain my full name, whereas it should have added
1236 a Resent-Sender: header. It erroneously treats any prefix of the
1237 username as equal to the username.
1239 This change corrects that bug.
1241 GF/01 DCC debug and logging tidyup
1242 Error conditions log to paniclog rather than rejectlog.
1243 Debug lines prefixed by "DCC: " to remove any ambiguity.
1245 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1247 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1248 Bug spotted by Jeremy Harris; was flawed since initial commit.
1249 Would have resulted in OCSP responses post-SNI triggering an Exim
1250 NULL dereference and crash.
1252 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1254 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1255 Bug detection, analysis and fix by Samuel Thibault.
1256 Bugzilla 1331, Debian bug #698092.
1258 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1260 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1261 Server implementation by Todd Lyons, client by JH.
1262 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1263 config variable "prdr_enable" controls whether the server
1264 advertises the facility. If the client requests PRDR a new
1265 acl_data_smtp_prdr ACL is called once for each recipient, after
1266 the body content is received and before the acl_smtp_data ACL.
1267 The client is controlled by both of: a hosts_try_prdr option
1268 on the smtp transport, and the server advertisement.
1269 Default client logging of deliveries and rejections involving
1270 PRDR are flagged with the string "PRDR".
1272 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1273 fclose(). Diagnosis by Todd Lyons.
1275 PP/17 Update configure.default to handle IPv6 localhost better.
1276 Patch by Alain Williams (plus minor tweaks).
1279 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1280 This is now consistent with GnuTLS, and is now documented: the
1281 previous undocumented portable approach to treating the option as
1282 unset was to force an expansion failure. That still works, and
1283 an empty string is now equivalent.
1285 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1286 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1287 not performing validation itself.
1289 PP/20 Added force_command boolean option to pipe transport.
1290 Patch from Nick Koston, of cPanel Inc.
1292 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1295 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1297 PP/21 Fix eximon continuous updating with timestamped log-files.
1298 Broken in a format-string cleanup in 4.80, missed when I repaired the
1299 other false fix of the same issue.
1300 Report and fix from Heiko Schlichting.
1303 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1304 Report from Prashanth Katuri.
1306 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1307 It's SecureTransport, so affects any MacOS clients which use the
1308 system-integrated TLS libraries, including email clients.
1310 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1311 using a MIME ACL for non-SMTP local injection.
1312 Report and assistance in diagnosis by Warren Baker.
1314 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1316 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1318 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1319 last id that failed so it may be referenced in subsequent ACL's.
1321 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1324 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1325 ldap library initialization, allowing self-signed CA's to be
1326 used. Also properly sets require_cert option later in code by
1327 using NULL (global ldap config) instead of ldap handle (per
1328 session). Bug diagnosis and testing by alxgomz.
1330 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1331 the src/util/ subdirectory.
1333 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1334 renamed to Transport Post Delivery Action by Jeremy Harris, as
1337 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1338 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1339 redis_servers = needs to be configured which will be used by the redis
1340 lookup. Patch from Warren Baker, of The Packet Hub.
1342 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1344 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1345 hostname or reverse DNS when processing a host list. Used suggestions
1346 from multiple comments on this bug.
1348 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1350 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1351 Missed a few lines, added it to make the runtest require no keyboard
1354 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1355 contains upper case chars. Make router use caseful_local_part.
1357 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1358 support when GnuTLS has been built with p11-kit.
1364 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1366 This, or similar/improved, will also be change PP/11 of 4.82.
1372 PP/01 Handle short writes when writing local log-files.
1373 In practice, only affects FreeBSD (8 onwards).
1374 Bugzilla 1053, with thanks to Dmitry Isaikin.
1376 NM/01 Bugzilla 949 - Documentation tweak
1378 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1381 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1383 PP/02 Implemented gsasl authenticator.
1385 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1387 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1388 `pkg-config foo` for cflags/libs.
1390 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1391 with rest of GSASL and with heimdal_gssapi.
1393 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1394 `pkg-config foo` for cflags/libs for the TLS implementation.
1396 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1397 properties get this fed in as external SSF. A number of robustness
1398 and debugging improvements to the cyrus_sasl authenticator.
1400 PP/08 cyrus_sasl server now expands the server_realm option.
1402 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1403 Patch by Jeremy Harris.
1405 PP/10 Added dbmjz lookup type.
1407 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1409 PP/12 MAIL args handles TAB as well as SP, for better interop with
1410 non-compliant senders.
1411 Analysis and variant patch by Todd Lyons.
1413 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1414 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1415 Patch from Dirk Mueller <dmueller@suse.com>
1417 PP/13 tls_peerdn now print-escaped for spool files.
1418 Observed some $tls_peerdn in wild which contained \n, which resulted
1419 in spool file corruption.
1421 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1422 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1423 or write after TLS renegotiation, which otherwise led to messages
1426 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1427 as a tracking header (ie: a signed header comes before the signature).
1428 Patch from Wolfgang Breyha.
1430 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1431 comma-sep list; embedded commas doubled.
1433 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1435 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1437 Report and patch from Dmitry Banschikov.
1439 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1440 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1441 failures. We appear to now support TLS1.1+ with Exim.
1443 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1444 lets Exim select keys and certificates based upon TLS SNI from client.
1445 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1446 before an outbound SMTP session. New log_selector, +tls_sni.
1448 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1449 NULL dereference. Report and patch from Alun Jones.
1451 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1452 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1453 Not seeing resolver debug output on NetBSD, but suspect this is a
1454 resolver implementation change.
1456 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1457 Left warnings. Added "eximon gdb" invocation mode.
1459 PP/21 Defaulting "accept_8bitmime" to true, not false.
1461 PP/22 Added -bw for inetd wait mode support.
1463 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1464 locate the relevant includes and libraries. Made this the default.
1466 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1467 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1469 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1470 This may cause build issues on older platforms.
1472 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1473 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1474 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1475 Added SNI support via GnuTLS too.
1476 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1478 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1480 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1481 Applied second patch from Janne, implementing suggestion to default
1482 multiple-strings-in-record handling to match SPF spec.
1484 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1486 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1487 read-only, out of scope).
1488 Patch from Wolfgang Breyha, report from Stuart Northfield.
1490 PP/29 Fix three issues highlighted by clang analyser static analysis.
1491 Only crash-plausible issue would require the Cambridge-specific
1492 iplookup router and a misconfiguration.
1493 Report from Marcin Mirosław.
1495 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1497 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1498 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1499 As part of this, removing so much warning spew let me fix some minor
1500 real issues in debug logging.
1502 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1503 assignment on my part. Fixed.
1505 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1506 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1507 Janne Snabb (who went above and beyond: thank you).
1509 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1510 string otherwise requires a connection and a bunch more work and it's
1511 relatively easy to get wrong. Should also expose TLS library linkage
1514 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1515 64-bit ${eval} (JH/03).
1517 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1518 GNU libc to support some of the 64-bit stuff, should not lead to
1519 conflicts. Defined before os.h is pulled in, so if a given platform
1520 needs to override this, it can.
1522 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1523 protection layer was required, which is not implemented.
1524 Bugzilla 1254, patch from Wolfgang Breyha.
1526 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1527 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1528 tls_dhparam take prime identifiers. Also unbreak combination of
1529 OpenSSL+DH_params+TLSSNI.
1531 PP/39 Disable SSLv2 by default in OpenSSL support.
1537 PP/01 Solaris build fix for Oracle's LDAP libraries.
1538 Bugzilla 1109, patch from Stephen Usher.
1540 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1542 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1545 TF/02 Fix a couple more cases where we did not log the error message
1546 when unlink() failed. See also change 4.74-TF/03.
1548 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1549 lock up or crash if it happened to be inside a call to libc when it
1550 got a SIGUSR1 from exiwhat.
1552 The SIGUSR1 handler appends the current process status to the process
1553 log which is later printed by exiwhat. It used to use the general
1554 purpose logging code to do this, but several functions it calls are
1555 not safe for signals.
1557 The new output code in the SIGUSR1 handler is specific to the process
1558 log, and simple enough that it's easy to inspect for signal safety.
1559 Removing some special cases also simplifies the general logging code.
1560 Removing the spurious timestamps from the process log simplifies
1563 TF/04 Improved ratelimit ACL condition.
1565 The /noupdate option has been deprecated in favour of /readonly which
1566 has clearer semantics. The /leaky, /strict, and /readonly update modes
1567 are mutually exclusive. The update mode is no longer included in the
1568 database key; it just determines when the database is updated. (This
1569 means that when you upgrade Exim will forget old rate measurements.)
1571 Exim now checks that the per_* options are used with an update mode that
1572 makes sense for the current ACL. For example, when Exim is processing a
1573 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1574 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1575 must specify per_mail/readonly. If you omit the update mode it defaults to
1576 /leaky where that makes sense (as before) or /readonly where required.
1578 The /noupdate option is now undocumented but still supported for
1579 backwards compatibility. It is equivalent to /readonly except that in
1580 ACLs where /readonly is required you may specify /leaky/noupdate or
1581 /strict/noupdate which are treated the same as /readonly.
1583 A useful new feature is the /count= option. This is a generalization
1584 of the per_byte option, so that you can measure the throughput of other
1585 aggregate values. For example, the per_byte option is now equivalent
1586 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1588 The per_rcpt option has been generalized using the /count= mechanism
1589 (though it's more complicated than the per_byte equivalence). When it is
1590 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1591 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1592 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1593 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1594 Note that using per_rcpt with a non-readonly update mode in more than
1595 one ACL will cause the recipients to be double-counted. (The per_mail
1596 and per_byte options don't have this problem.)
1598 The handling of very low rates has changed slightly. If the computed rate
1599 is less than the event's count (usually one) then this event is the first
1600 after a long gap. In this case the rate is set to the same as this event's
1601 count, so that the first message of a spam run is counted properly.
1603 The major new feature is a mechanism for counting the rate of unique
1604 events. The new per_addr option counts the number of different
1605 recipients that someone has sent messages to in the last time period. It
1606 behaves like per_rcpt if all the recipient addresses are different, but
1607 duplicate recipient addresses do not increase the measured rate. Like
1608 the /count= option this is a general mechanism, so the per_addr option
1609 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1610 example, measure the rate that a client uses different sender addresses
1611 with the options per_mail/unique=$sender_address. There are further
1612 details in the main documentation.
1614 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1616 TF/06 Removed a few PCRE remnants.
1618 TF/07 Automatically extract Exim's version number from tags in the git
1619 repository when doing development or release builds.
1621 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1622 Bugzilla 879. Patch from Paul Fisher.
1624 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1625 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1628 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1630 PP/05 Variable $av_failed, true if the AV scanner deferred.
1631 Bugzilla 1078. Patch from John Horne.
1633 PP/06 Stop make process more reliably on build failure.
1634 Bugzilla 1087. Patch from Heiko Schlittermann.
1636 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1637 Bugzilla 1089. Patch from Heiko Schlittermann.
1639 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1640 Bugzilla 1131. Patch from Holger Weiß.
1642 PP/09 Handle IPv6 addresses with SPF.
1643 Bugzilla 860. Patch from Wolfgang Breyha.
1645 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1647 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1650 PP/11 match_* no longer expand right-hand-side by default.
1651 New compile-time build option, EXPAND_LISTMATCH_RHS.
1652 New expansion conditions, "inlist", "inlisti".
1654 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1656 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1658 PP/14 fix log_write() format string regression from TF/03.
1659 Bugzilla 1152. Patch from Dmitry Isaikin.
1665 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1667 PP/02 Harmonised TLS library version reporting; only show if debugging.
1668 Layout now matches that introduced for other libraries in 4.74 PP/03.
1670 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1672 PP/04 New "dns_use_edns0" global option.
1674 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1677 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1678 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1680 TK/01 Updated PolarSSL code to 0.14.2.
1681 Bugzilla 1097. Patch from Andreas Metzler.
1683 PP/07 Catch divide-by-zero in ${eval:...}.
1684 Fixes bugzilla 1102.
1686 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1689 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1690 format-string attack -- SECURITY: remote arbitrary code execution.
1692 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1693 time unintentionally subject to list matching rules, letting the header
1694 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1695 arbitrary string expansion). This allowed for information disclosure.
1697 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1698 INT_MIN/-1 -- value coerced to INT_MAX.
1704 NM/01 Workaround for PCRE version dependency in version reporting
1707 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1708 This fixes portability to compilers other than gcc, notably
1709 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1711 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1712 makefiles for portability to HP-UX and POSIX correctness.
1714 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1715 Also via indented variable definition in the Makefile.
1716 (Debugging by Oliver Heesakkers).
1718 PP/02 Restore caching of spamd results with expanded spamd_address.
1719 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1721 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1722 Improves build reliability. Fix from: Frank Elsner
1724 NM/02 Fix wide character breakage in the rfc2047 coding
1725 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1727 NM/03 Allow underscore in dnslist lookups
1728 Fixes bug 1026. Patch from Graeme Fowler
1730 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1731 Code patches from Adam Ciarcinski of NetBSD.
1733 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1736 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1737 is logged, to avoid truncation. Patch from John Horne.
1739 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1740 Patch from Jakob Hirsch.
1742 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1743 SQL string expansion failure details.
1744 Patch from Andrey Oktyabrski.
1746 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1747 Patch from Simon Arlott.
1749 PP/09 New lookups functionality failed to compile on old gcc which rejects
1750 extern declarations in function scope.
1751 Patch from Oliver Fleischmann
1753 PP/10 Use sig_atomic_t for flags set from signal handlers.
1754 Check getgroups() return and improve debugging.
1755 Fixed developed for diagnosis in bug 927 (which turned out to be
1758 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1759 Patch from Mark Zealey.
1761 PP/12 Bugzilla 1056: Improved spamd server selection.
1762 Patch from Mark Zealey.
1764 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1765 Based on patch from Heiko Schlittermann.
1767 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1768 Patch from Uwe Doering, sign-off by Michael Haardt.
1770 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1771 variable declaration deep within a block. Bug and patch from
1774 PP/15 lookups-Makefile IRIX compatibility coercion.
1776 PP/16 Make DISABLE_DKIM build knob functional.
1778 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1779 Patch by Simon Arlott
1781 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1782 variable argument macros. Our copy now differs from upstream.
1788 TF/01 Failure to get a lock on a hints database can have serious
1789 consequences so log it to the panic log.
1791 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1792 controlled using the smtp_confirmation log selector.
1794 TF/03 Include the error message when we fail to unlink a spool file.
1796 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1797 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1798 for maintaining out-of-tree patches for some time.
1800 PP/01 Bugzilla 139: Documentation and portability issues.
1801 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1802 Handle per-OS dynamic-module compilation flags.
1804 PP/02 Let /dev/null have normal permissions.
1805 The 4.73 fixes were a little too stringent and complained about the
1806 permissions on /dev/null. Exempt it from some checks.
1807 Reported by Andreas M. Kirchwitz.
1809 PP/03 Report version information for many libraries, including
1810 Exim version information for dynamically loaded libraries. Created
1811 version.h, now support a version extension string for distributors
1812 who patch heavily. Dynamic module ABI change.
1814 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1815 privilege escalation vulnerability whereby the Exim run-time user
1816 can cause root to append content of the attacker's choosing to
1819 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1822 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1823 If dropping privileges for untrusted macros, we disabled normal logging
1824 on the basis that it would fail; for the Exim run-time user, this is not
1825 the case, and it resulted in successful deliveries going unlogged.
1826 Fixed. Reported by Andreas Metzler.
1832 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1833 only prepend for the Resent-* case. Fixes regression introduced in
1834 Exim 4.70 by NM/22 for Bugzilla 607.
1836 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1837 increasing numbers of administrators be bitten by this.
1839 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1841 PP/03 Bugzilla 994: added openssl_options main configuration option.
1843 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1845 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1847 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1849 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1850 without a peer certificate, leading to a segfault because of an
1851 assumption that peers always have certificates. Be a little more
1852 paranoid. Problem reported by Martin Tscholak.
1854 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1855 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1856 NB: ClamAV planning to remove STREAM in "middle of 2010".
1857 CL also introduces -bmalware, various -d+acl logging additions and
1858 more caution in buffer sizes.
1860 PP/09 Implemented reverse_ip expansion operator.
1862 PP/10 Bugzilla 937: provide a "debug" ACL control.
1864 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1866 PP/12 Bugzilla 973: Implement --version.
1868 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1870 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1872 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1874 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1875 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1876 ignore trailing whitespace.
1878 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1880 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1883 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1884 Notification from Dr Andrew Aitchison.
1886 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1887 ExtendedDetectionInfo response format.
1888 Notification from John Horne.
1890 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1893 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1894 XSL and documented dependency on system catalogs, with examples of how
1897 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1900 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1901 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1902 configuration file which is writeable by the Exim user or group.
1904 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1905 of configuration files to cover files specified with the -C option if
1906 they are going to be used with root privileges, not just the default
1909 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1910 option (effectively making it always true).
1912 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1913 files to be used while preserving root privileges.
1915 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1916 that rogue child processes cannot use them.
1918 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1919 run-time user, instead of root.
1921 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1922 Exim run-time user without dropping privileges.
1924 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1925 result string, instead of calling string_vformat() twice with the same
1928 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1929 for other users. Others should always drop root privileges if they use
1930 -C on the command line, even for a whitelisted configure file.
1932 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1934 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1940 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1941 $data_path, and $header_path variables; fixed documentation bugs and
1944 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1945 exipick to access non-standard spools, including the "frozen" queue
1948 NM/01 Bugzilla 965: Support mysql stored procedures.
1949 Patch from Alain Williams
1951 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1953 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1954 Patch from Andreas Metzler
1956 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1957 Patch from Kirill Miazine
1959 NM/05 Bugzilla 671: Added umask to procmail example.
1961 JJ/03 installed exipick 20100323.0, fixing doc bug
1963 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1964 directory. Notification and patch from Dan Rosenberg.
1966 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1968 TK/02 Improve log output when DKIM signing operation fails.
1970 MH/01 Treat the transport option dkim_domain as a colon separated
1971 list, not as a single string, and sign the message with each element,
1972 omitting multiple occurences of the same signer.
1974 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1975 Bugzilla 985, 986. Patch by Simon Arlott
1977 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1978 Patch by Simon Arlott
1980 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1981 MBX locking. Notification from Dan Rosenberg.
1987 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1989 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1991 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1993 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1995 NM/04 Bugzilla 915: Flag broken perl installation during build.
2001 TK/01 Added patch by Johannes Berg that expands the main option
2002 "spamd_address" if it starts with a dollar sign.
2004 TK/02 Write list of recipients to X-Envelope-Sender header when building
2005 the mbox-format spool file for content scanning (suggested by Jakob
2008 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2009 (http://www.dcc-servers.net/) support via dccifd. Activated by
2010 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2012 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2013 by Mark Daniel Reidel <mr@df.eu>.
2015 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2016 When building exim an external PCRE library is now needed -
2017 PCRE is a system library on the majority of modern systems.
2018 See entry on PCRE_LIBS in EDITME file.
2020 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2021 conversation. Added nologin parameter to request.
2022 Patch contributed by Kirill Miazine.
2024 TF/01 Do not log submission mode rewrites if they do not change the address.
2026 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2028 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2029 log files in place. Contributed by Roberto Lima.
2031 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2033 TF/03 Bugzilla 615: When checking the local_parts router precondition
2034 after a local_part_suffix or local_part_prefix option, Exim now
2035 does not use the address's named list lookup cache, since this
2036 contains cached lookups for the whole local part.
2038 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2039 Robert Millan. Documentation is in experimental-spec.txt.
2041 TF/04 Bugzilla 668: Fix parallel build (make -j).
2043 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2045 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2046 Patch provided by Jan Srzednicki.
2048 TF/05 Leading white space used to be stripped from $spam_report which
2049 wrecked the formatting. Now it is preserved.
2051 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2052 that they are available at delivery time.
2054 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2056 TF/08 TLS error reporting now respects the incoming_interface and
2057 incoming_port log selectors.
2059 TF/09 Produce a more useful error message if an SMTP transport's hosts
2060 setting expands to an empty string.
2062 NM/06 Bugzilla 744: EXPN did not work under TLS.
2063 Patch provided by Phil Pennock.
2065 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2066 Patch provided by Richard Godbee.
2068 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2069 acl_smtp_notquit, added index entry.
2071 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2072 Patch provided by Eugene Bujak.
2074 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2075 accept(). Patch provided by Maxim Dounin.
2077 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2078 Patch provided by Phil Pennock.
2080 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2082 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2083 Patch provided by Brad "anomie" Jorsch.
2085 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2086 Patch provided by Dean Brooks.
2088 TK/05 Add native DKIM support (does not depend on external libraries).
2090 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2091 Patch provided by Graeme Fowler.
2093 NM/16 Bugzilla 851: Documentation example syntax fix.
2095 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2097 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2100 NM/19 Bugzilla 745: TLS version reporting.
2101 Patch provided by Phil Pennock.
2103 NM/20 Bugzilla 167: bool: condition support.
2104 Patch provided by Phil Pennock.
2106 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2107 clients. Patch provided by Phil Pennock.
2109 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2110 Patch provided by Brad "anomie" Jorsch.
2112 NM/23 Bugzilla 687: Fix misparses in eximstats.
2113 Patch provided by Heiko Schlittermann.
2115 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2116 Patch provided by Heiko Schlittermann.
2118 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2119 plus update to original patch.
2121 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2123 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2124 Patch provided by David Brownlee.
2126 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2128 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2130 NM/30 Bugzilla 888: TLS documentation bugfixes.
2132 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2134 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2135 Unlike the original bugzilla I have changed all shell scripts in src tree.
2137 NM/33 Bugzilla 898: Transport filter timeout fix.
2138 Patch by Todd Rinaldo.
2140 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2141 Patch by Serge Demonchaux.
2143 NM/35 Bugzilla 39: Base64 decode bug fixes.
2144 Patch by Jakob Hirsch.
2146 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2148 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2150 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2152 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2158 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2159 ALT-N's libdkim that I have put here:
2160 http://duncanthrax.net/exim-experimental/
2162 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2163 were called 'true' and it seems that C99 defines that as a reserved
2164 keyword to be used with 'bool' variable types. That means you could
2165 not include C99-style headers which use bools without triggering
2166 build errors in sieve.c.
2168 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2169 as mailq or other aliases. Changed the --help handling significantly
2170 to do whats expected. exim_usage() emits usage/help information.
2172 SC/01 Added the -bylocaldomain option to eximstats.
2174 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2176 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2178 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2184 PH/01 Another patch from the Sieve maintainer.
2186 PH/02 When an IPv6 address is converted to a string for single-key lookup
2187 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2188 dots are used instead of colons so that keys in lsearch files need not
2189 contain colons. This was done some time before quoting was made available
2190 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2191 (notated using the quote facility) so as to distinguish them from IPv4
2192 keys. This meant that lookups for IP addresses in host lists did not work
2193 for iplsearch lookups.
2195 This has been fixed by arranging for IPv6 addresses to be expressed with
2196 colons if the lookup type is iplsearch. This is not incompatible, because
2197 previously such lookups could never work.
2199 The situation is now rather anomalous, since one *can* have colons in
2200 ordinary lsearch keys. However, making the change in all cases is
2201 incompatible and would probably break a number of configurations.
2203 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2206 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2207 conversion specification without a maximum field width, thereby enabling
2208 a rogue spamd server to cause a buffer overflow. While nobody in their
2209 right mind would setup Exim to query an untrusted spamd server, an
2210 attacker that gains access to a server running spamd could potentially
2211 exploit this vulnerability to run arbitrary code as the Exim user.
2213 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2214 $primary_hostname instead of what libspf2 thinks the hosts name is.
2216 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2217 a directory entry by the name of the lookup key. Previously, if a
2218 symlink pointed to a non-existing file or a file in a directory that
2219 Exim lacked permissions to read, a lookup for a key matching that
2220 symlink would fail. Now it is enough that a matching directory entry
2221 exists, symlink or not. (Bugzilla 503.)
2223 PH/03 The body_linecount and body_zerocount variables are now exported in the
2226 PH/04 Added the $dnslist_matched variable.
2228 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2229 This means they are set thereafter only if the connection becomes
2232 PH/06 Added the client_condition to authenticators so that some can be skipped
2233 by clients under certain conditions.
2235 PH/07 The error message for a badly-placed control=no_multiline_responses left
2236 "_responses" off the end of the name.
2238 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2240 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2241 (without spaces) instead of just copying the configuration text.
2243 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2245 PH/11 Added $max_received_linelength.
2247 PH/12 Added +ignore_defer and +include_defer to host lists.
2249 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2250 way in which PCRE > 7.0 is built.
2252 PH/14 Implemented queue_only_load_latch.
2254 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2255 MAIL command. The effect was to mangle the value on 64-bit systems.
2257 PH/16 Another patch from the Sieve maintainer.
2259 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2261 PH/18 If a system quota error occurred while trying to create the file for
2262 a maildir delivery, the message "Mailbox is full" was not appended to the
2263 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2264 only to a quota excession during the actual writing of the file.
2266 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2267 characters?) which causes problems in log lines. The DN values are now
2268 passed through string_printing() before being added to log lines.
2270 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2271 and InterBase are left for another time.)
2273 PH/21 Added message_body_newlines option.
2275 PH/22 Guard against possible overflow in moan_check_errorcopy().
2277 PH/23 POSIX allows open() to be a macro; guard against that.
2279 PH/24 If the recipient of an error message contained an @ in the local part
2280 (suitably quoted, of course), incorrect values were put in $domain and
2281 $local_part during the evaluation of errors_copy.
2287 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2288 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2291 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2292 issue a MAIL command.
2294 PH/02 In an ACL statement such as
2296 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2298 if a client was not listed at all, or was listed with a value other than
2299 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2300 the condition was not true (as it should be), so access was not denied.
2301 The bug was that the ! inversion was incorrectly passed on to the second
2302 item. This has been fixed.
2304 PH/03 Added additional dnslists conditions == and =& which are different from
2305 = and & when the dns lookup returns more than one IP address.
2307 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2308 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2310 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2311 FSYNC, which compiles an option called disable_fsync that allows for
2312 bypassing fsync(). The documentation is heavily laced with warnings.
2314 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2316 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2317 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2318 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2319 including adding "make clean"; (3) Added -fPIC when compiling the test
2320 dynamically loaded module, to get rid of a warning.
2322 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2323 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2324 = 0s. The bug is otherwise harmless.
2326 PH/07 There was a bug in the dovecot authenticator such that the value of
2327 $auth1 could be overwritten, and so not correctly preserved, after a
2328 successful authentication. This usually meant that the value preserved by
2329 the server_setid option was incorrect.
2331 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2333 PH/09 Installed PCRE release 7.0.
2335 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2336 run for batched SMTP input. It is now run at the start of every message
2337 in the batch. While fixing this I discovered that the process information
2338 (output by running exiwhat) was not always getting set for -bs and -bS
2339 input. This is fixed, and it now also says "batched" for BSMTP.
2341 PH/11 Added control=no_pipelining.
2343 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2344 patch, slightly modified), and move the expansion of helo_data till after
2345 the connection is made in the smtp transport (so it can use these
2348 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2350 PH/14 Added log_selector = +pid.
2352 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2354 PH/16 Add ${if forany and ${if forall.
2356 PH/17 Added dsn_from option to vary the From: line in DSNs.
2358 PH/18 Flush SMTP output before performing a callout, unless control =
2359 no_callout_flush is set.
2361 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2362 was true (the default) a successful delivery failed to delete the retry
2363 item, thus causing premature timeout of the address. The bug is now
2366 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2368 PH/21 Long custom messages for fakedefer and fakereject are now split up
2369 into multiline responses in the same way that messages for "deny" and
2370 other ACL rejections are.
2372 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2373 with slight modification.
2375 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2376 draft, changing the syntax and factoring some duplicate code".
2378 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2379 for deliveries of the second and subsequent messages over the same SMTP
2382 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2383 ${reduce, with only minor "tidies".
2385 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2387 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2388 expansion side effects.
2390 PH/27 When a message times out after an over-quota error from an Exim-imposed
2391 quota, the bounce message says "mailbox is full". This message was not
2392 being given when it was a system quota that was exceeded. It now should
2395 MH/03 Made $recipients available in local_scan(). local_scan() already has
2396 better access to the recipient list through recipients_list[], but
2397 $recipients can be useful in postmaster-provided expansion strings.
2399 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2400 in the case of a MAIL command with additional options following the
2401 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2402 were accidentally chopped off.
2404 PH/29 SMTP synchronization checks are implemented when a command is read -
2405 there is a check that no more input is waiting when there shouldn't be
2406 any. However, for some commands, a delay in an ACL can mean that it is
2407 some time before the response is written. In this time, more input might
2408 arrive, invalidly. So now there are extra checks after an ACL has run for
2409 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2410 pipelining has not been advertised.
2412 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2414 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2415 returned in the SMTP response when smtp_return_error_details was false.
2416 This has been fixed.
2418 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2419 instead of the C I/O that was originally supplied, because problems were
2420 reported on Solaris.
2422 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2423 Exim which did not show up earlier: it was assuming that a call to
2424 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2425 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2426 was a macro that became an assignment, so it seemed to work. This has
2427 changed to a proper function call with a void return, hence the compile
2428 error. Exim's code has been fixed.
2430 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2433 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2435 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2437 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2438 because it turns out to be too big or there is a timeout) there is no
2439 "Completed" line in the log. When some messages of this type were
2440 selected by exigrep, they were listed as "not completed". Others were
2441 picked up by some special patterns. I have improved the selection
2442 criteria to be more general.
2444 PH/38 The host_find_failed option in the manualroute router can now be set
2445 to "ignore", to completely ignore a host whose IP address cannot be
2446 found. If all hosts are ignored, the behaviour is controlled by the new
2447 host_all_ignored option.
2449 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2450 homing or because of multiple MX records with /mx) generated more than
2451 one IP address, and the following item turned out to be the local host,
2452 all the secondary addresses of the first item were incorrectly removed
2453 from the list, along with the local host and any following hosts (which
2454 is what is supposed to happen).
2456 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2457 whoever called Exim into the -H file. In the case of the daemon it was
2458 behaving confusingly. When first started, it used values for whoever
2459 started the daemon, but after a SIGHUP it used the Exim user (because it
2460 calls itself on a restart). I have changed the code so that it now always
2463 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2464 message are rejected with the same error (e.g. no authentication or bad
2465 sender address), and a DATA command is nevertheless sent (as can happen
2466 with PIPELINING or a stupid MUA), the error message that was given to the
2467 RCPT commands is included in the rejection of the DATA command. This is
2468 intended to be helpful for MUAs that show only the final error to their
2471 PH/42 Another patch from the Sieve maintainer.
2473 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2474 Eximstats - Fixed some broken HTML links and added missing column headers
2476 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2477 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2479 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2481 PH/43 Yet another patch from the Sieve maintainer.
2483 PH/44 I found a way to check for a TCP/IP connection going away before sending
2484 the response to the final '.' that terminates a message, but only in the
2485 case where the client has not sent further data following the '.'
2486 (unfortunately, this is allowed). However, in many cases there won't be
2487 any further data because there won't be any more messages to send. A call
2488 to select() can be used: if it shows that the input is "ready", there is
2489 either input waiting, or the socket has been closed. An attempt to read
2490 the next input character can distinguish the two cases. Previously, Exim
2491 would have sent an OK response which the client would never have see.
2492 This could lead to message repetition. This fix should cure that, at
2493 least in a lot of common cases.
2495 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2496 advertised in response to EHLO.
2502 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2503 fixed by 4.65/MH/01 (is this a record?) are fixed:
2505 (i) An empty string was always treated as zero by the numeric comparison
2506 operators. This behaviour has been restored.
2508 (ii) It is documented that the numeric comparison operators always treat
2509 their arguments as decimal numbers. This was broken in that numbers
2510 starting with 0 were being interpreted as octal.
2512 While fixing these problems I realized that there was another issue that
2513 hadn't been noticed. Values of message_size_limit (both the global option
2514 and the transport option) were treated as octal if they started with 0.
2515 The documentation was vague. These values are now always treated as
2516 decimal, and I will make that clear in the documentation.
2522 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2523 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2526 MH/01 Don't check that the operands of numeric comparison operators are
2527 integers when their expansion is in "skipping" mode (fixes bug
2528 introduced by 4.64-PH/07).
2530 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2531 child addresses, Exim now panics and dies. Previously, because the count
2532 is held in a short int, deliveries were likely to be lost. As such a
2533 large number of recipients for a single message is ridiculous
2534 (performance will be very, very poor), I have chosen to impose a limit
2535 rather than extend the field.
2541 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2542 leftover -K file (the existence of which was triggered by #402).
2543 While we were at it, introduced process PID as part of the -K
2544 filename. This should rule out race conditions when creating
2547 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2548 processing considerably. Previous code took too long for large mails,
2549 triggering a timeout which in turn triggers #401.
2551 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2552 in the DK code in transports.c. sendfile() is not really portable,
2553 hence the _LINUX specificness.
2555 TF/01 In the add_headers option to the mail command in an Exim filter,
2556 there was a bug that Exim would claim a syntax error in any
2557 header after the first one which had an odd number of characters
2560 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2561 callout verification, Exim cached a "reject" for the entire domain. This
2562 is correct for most verifications, but it is not correct for a recipient
2563 verification with use_sender or use_postmaster set, because in that case
2564 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2565 case of MAIL FROM:<> rejection from other early rejections (e.g.
2566 rejection of HELO). When verifying a recipient using a non-null MAIL
2567 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2568 Whatever the result of the callout, the value of the domain cache is
2569 left unchanged (for any other kind of callout, getting as far as trying
2570 RCPT means that the domain itself is ok).
2572 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2575 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2576 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2577 session, the daemon ignores ECONNECT errors and logs others; it now
2578 ignores EPIPE as well.
2580 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2581 (quoted-printable decoding).
2583 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2584 later the small subsequent patch to fix an introduced bug.
2586 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2588 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2590 PH/08 An error is now given if message_size_limit is specified negative.
2592 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2593 to be given (somewhat) arbitrary names.
2595 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2598 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2599 miscellaneous code fixes
2601 PH/10 Added the log_reject_target ACL modifier to specify where to log
2604 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2605 hostname. This is wrong, because it relates to the incoming message (and
2606 probably the interface on which it is arriving) and not to the outgoing
2607 callout (which could be using a different interface). This has been
2608 changed to use the value of the helo_data option from the smtp transport
2609 instead - this is what is used when a message is actually being sent. If
2610 there is no remote transport (possible with a router that sets up host
2611 addresses), $smtp_active_hostname is used.
2613 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2614 tweaks were necessary in order to get it to work (see also 21 below):
2615 (a) The code assumed that strncpy() returns a negative number on buffer
2616 overflow, which isn't the case. Replaced with Exim's string_format()
2618 (b) There were several signed/unsigned issues. I just did the minimum
2619 hacking in of casts. There is scope for a larger refactoring.
2620 (c) The code used strcasecmp() which is not a standard C function.
2621 Replaced with Exim's strcmpic() function.
2622 (d) The code set only $1; it now sets $auth1 as well.
2623 (e) A simple test gave the error "authentication client didn't specify
2624 service in request". It would seem that Dovecot has changed its
2625 interface. Fortunately there's a specification; I followed it and
2626 changed what the client sends and it appears to be working now.
2628 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2631 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2632 address A is aliased to B and C, where B exists and C does not. Without
2633 -v the output is "A verified" because verification stops after a
2634 successful redirection if more than one address is generated. However,
2635 with -v the child addresses are also verified. Exim was outputting "A
2636 failed to verify" and then showing the successful verification for C,
2637 with its parentage. It now outputs "B failed to verify", showing B's
2638 parentage before showing the successful verification of C.
2640 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2641 look up a TXT record in a specific list after matching in a combined
2644 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2645 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2646 they consult the DNS. I had assumed they would set it the way they
2647 wanted; and indeed my experiments on Linux seem to show that in some
2648 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2649 To be on the safe side, however, I have now made the interface to
2650 host_find_byname() similar to host_find_bydns(), with an argument
2651 containing the DNS resolver options. The host_find_byname() function now
2652 sets these options at its start, just as host_find_bydns() does. The smtp
2653 transport options dns_qualify_single and dns_search_parents are passed to
2654 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2655 of host_find_byname() use the default settings of RES_DEFNAMES
2656 (qualify_single) but not RES_DNSRCH (search_parents).
2658 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2659 spool_read_header() do less string testing, by means of a preliminary
2660 switch on the second character of optional "-foo" lines. (This is
2661 overdue, caused by the large number of possibilities that now exist.
2662 Originally there were few.) While I was there, I also converted the
2663 str(n)cmp tests so they don't re-test the leading "-" and the first
2664 character, in the hope this might squeeze out yet more improvement.
2666 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2667 flag allowing group syntax was set by the header_syntax check but not
2668 turned off, possible causing trouble later; (2) The flag was not being
2669 set at all for the header_verify test, causing "group"-style headers to
2670 be rejected. I have now set it in this case, and also caused header_
2671 verify to ignore an empty address taken from a group. While doing this, I
2672 came across some other cases where the code for allowing group syntax
2673 while scanning a header line wasn't quite right (mostly, not resetting
2674 the flag correctly in the right place). These bugs could have caused
2675 trouble for malformed header lines. I hope it is now all correct.
2677 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2678 with the "reply" argument non-NULL. The code, however (which originally
2679 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2680 but it didn't always do it. This confused somebody who was copying the
2681 code for some other use. I have removed all the tests.
2683 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2684 feature that was used to support insecure browsers during the U.S. crypto
2685 embargo. It requires special client support, and Exim is probably the
2686 only MTA that supported it -- and would never use it because real RSA is
2687 always available. This code has been removed, because it had the bad
2688 effect of slowing Exim down by computing (never used) parameters for the
2689 RSA_EXPORT functionality.
2691 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2692 authenticator to fail if there's a tab character in the incoming data
2693 (there should never be unless someone is messing about, as it's supposed
2694 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2695 if the connection is using TLS or if the remote IP is the same as the
2696 local IP, and the "valid-client-cert option" if a client certificate has
2699 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2700 authenticators. This can be used for authorization after authentication
2701 succeeds. (In the case of plaintext, it servers for both authentication
2704 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2705 if any retry times were supplied.
2707 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2708 connection, where there is no client IP address to check. In this
2709 situation, the verify now always succeeds.
2711 PH/25 Applied John Jetmore's -Mset patch.
2713 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2715 PH/27 In a string expansion for a processed (not raw) header when multiple
2716 headers of the same name were present, leading whitespace was being
2717 removed from all of them, but trailing whitespace was being removed only
2718 from the last one. Now trailing whitespace is removed from each header
2719 before concatenation. Completely empty headers in a concatenation (as
2720 before) are ignored.
2722 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2723 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2725 PH/29 [Removed. This was a change that I later backed out, and forgot to
2726 correct the ChangeLog entry (that I had efficiently created) before
2727 committing the later change.]
2729 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2730 address errors (4xx response to RCPT) over the same connection as other
2731 messages routed to the same hosts. Such deliveries are always "forced",
2732 so retry times are not inspected. This resulted in far too many retries
2733 for the affected addresses. The effect occurred only when there were more
2734 hosts than the hosts_max_try setting in the smtp transport when it had
2735 the 4xx errors. Those hosts that it had tried were not added to the list
2736 of hosts for which the message was waiting, so if all were tried, there
2737 was no problem. Two fixes have been applied:
2739 (i) If there are any address or message errors in an SMTP delivery, none
2740 of the hosts (tried or untried) are now added to the list of hosts
2741 for which the message is waiting, so the message should not be a
2742 candidate for sending over the same connection that was used for a
2743 successful delivery of some other message. This seems entirely
2744 reasonable: after all the message is NOT "waiting for some host".
2745 This is so "obvious" that I'm not sure why it wasn't done
2746 previously. Hope I haven't missed anything, but it can't do any
2747 harm, as the worst effect is to miss an optimization.
2749 (ii) If, despite (i), such a delivery is accidentally attempted, the
2750 routing retry time is respected, so at least it doesn't keep
2751 hammering the server.
2753 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2754 in ${readsocket because some servers need this prod.
2756 PH/32 Added some extra debug output when updating a wait-xxx database.
2758 PH/33 The hint "could be header name not terminated by colon", which has been
2759 given for certain expansion errors for a long time, was not being given
2760 for the ${if def:h_colon_omitted{... case.
2762 PH/34 The spec says: "With one important exception, whenever a domain list is
2763 being scanned, $domain contains the subject domain." There was at least
2764 one case where this was not true.
2766 PH/35 The error "getsockname() failed: connection reset by peer" was being
2767 written to the panic log as well as the main log, but it isn't really
2768 panic-worthy as it just means the connection died rather early on. I have
2769 removed the panic log writing for the ECONNRESET error when getsockname()
2772 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2773 runs only) independently of the message's sender address. This meant
2774 that, if the 4xx error was in fact related to the sender, a different
2775 message to the same recipient with a different sender could confuse
2776 things. In particular, this can happen when sending to a greylisting
2777 server, but other circumstances could also provoke similar problems.
2778 I have changed the default so that the retry time for these errors is now
2779 based a combination of the sender and recipient addresses. This change
2780 can be overridden by setting address_retry_include_sender=false in the
2783 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2784 remote server are returned as part of bounce messages. This was not
2785 happening for LMTP over a pipe (the lmtp transport), but now it is the
2786 same for both kinds of LMTP.
2788 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2789 in header lines that were in fact CNAMEs. This is no longer the case.
2791 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2792 and queue runs started by the daemon processed all messages. This has
2793 been fixed so that -R and -S can now usefully be given with -q<time>.
2795 PH/40 Import PCRE release 6.7 (fixes some bugs).
2797 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2799 PH/42 Give an error if -q is specified more than once.
2801 PH/43 Renamed the variables $interface_address and $interface_port as
2802 $received_ip_address and $received_port, to make it clear that these
2803 values apply to message reception, and not to the outgoing interface when
2804 a message is delivered. (The old names remain recognized, of course.)
2806 PH/44 There was no timeout on the connect() call when using a Unix domain
2807 socket in the ${readsocket expansion. There now is.
2809 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2810 be meaningful with "accept".
2812 SC/01 Eximstats V1.43
2813 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2815 SC/02 Eximstats V1.44
2816 Use a glob alias rather than an array ref in the generated
2817 parser. This improves both readability and performance.
2819 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2820 Collect SpamAssassin and rejection statistics.
2821 Don't display local sender or destination tables unless
2822 there is data to show.
2823 Added average volumes into the top table text output.
2825 SC/04 Eximstats V1.46
2826 Collect data on the number of addresses (recipients)
2827 as well as the number of messages.
2829 SC/05 Eximstats V1.47
2830 Added 'Message too big' to the list of mail rejection
2831 reasons (thanks to Marco Gaiarin).
2833 SC/06 Eximstats V1.48
2834 Mainlog lines which have GMT offsets and are too short to
2835 have a flag are now skipped.
2837 SC/07 Eximstats V1.49 (Alain Williams)
2838 Added the -emptyok flag.
2840 SC/08 Eximstats V1.50
2841 Fixes for obtaining the IP address from reject messages.
2843 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2844 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2845 whitespace changes from 4.64-PH/27
2847 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2850 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2851 are found, allow negative numbers in numeric criteria)
2853 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2855 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2856 to match changes made in 4.64-PH/43
2858 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2860 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2861 "domain" is actually an IP address, and force a failure. This locks out
2862 those revolvers/nameservers that support "A-for-A" lookups, in
2863 contravention of the specifications.
2865 PH/48 When a host name was looked up from an IP address, and the subsequent
2866 forward lookup of the name timed out, the host name was left in
2867 $sender_host_name, contrary to the specification.
2869 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2870 restricted to single-key lookups, Exim was not diagnosing an error if
2871 * or *@ was used with a query-style lookup.
2873 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2875 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2876 long ago, but noone interested enough thought of it. Let's just say that
2877 the "1.1" means that there are some new functions that weren't there at
2878 some point in the past.
2880 PH/51 Error processing for expansion failure of helo_data from an smtp
2881 transport during callout processing was broken.
2883 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2884 tested/used via the -bh/-bhc/-bs options.
2886 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2887 bug, fixed in subsequent PCRE releases).
2889 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2890 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2892 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2898 SC/01 Use a glob alias rather than an array ref in eximstats generated
2899 parser. This improves both readability and performance.
2901 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2902 Don't display local sender or destination tables in eximstats unless
2903 there is data to show.
2904 Added average volumes into the eximstats top table text output.
2906 SC/03 Collect data on the number of addresses (recipients) as well
2907 as the number of messages in eximstats.
2909 TF/01 Correct an error in the documentation for the redirect router. Exim
2910 does (usually) call initgroups() when daemonizing.
2912 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2913 with consistent privilege compared to when running as a daemon.
2915 TF/03 Note in the spec that $authenticated_id is not set for local
2916 submissions from trusted users.
2918 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2919 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2921 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2922 by adding some example configuration directives to the default
2923 configuration file. A little bit of work is required to uncomment the
2924 directives and define how usernames and passwords are checked, but
2925 there is now a framework to start from.
2927 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2928 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2929 without this. I don't know how relevant this is to other LDAP libraries.
2931 PH/02 Add the verb name to the "unknown ACL verb" error.
2933 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2935 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2937 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2938 directory not expanded when it should be if an expanded home directory
2939 was set for the address (which is overridden by the transport).
2941 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2944 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2945 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2946 because it is too late at that time, and has no effect.
2948 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2949 security issue with \' (bugzilla #107). I could not use the
2950 PQescapeStringConn() function, because it needs a PGconn value as one of
2953 PH/08 When testing addresses using -bt, indicate those final addresses that
2954 are duplicates that would not cause an additional delivery. At least one
2955 person was confused, thinking that -bt output corresponded to deliveries.
2956 (Suppressing duplicates isn't a good idea as you lose the information
2957 about possibly different redirections that led to the duplicates.)
2959 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2960 systems where poll() doesn't work, in particular OS X.
2962 PH/10 Added more information to debugging output for retry time not reached.
2964 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2965 operations in malware.c.
2967 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2970 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2971 syslog_duplication set false, log lines that would normally be written
2972 both the the main log and to the reject log were not written to syslog at
2975 PH/14 In the default configuration, change the use of "message" in ACL warn
2976 statements to "add_header".
2978 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2979 not followed by a command (e.g. "seen endif").
2981 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2982 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2985 PH/17 Added extra conditions to the default value of delay_warning_condition
2989 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2990 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2991 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2994 The Auto-Submitted: and various List- headers are standardised, whereas I
2995 don't think Precedence: ever was.
2997 PH/18 Refactored debugging code in route_finduser() to show more information,
2998 in particular, the error code if getpwnam() issues one.
3000 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3001 This is apparently needed in addition to the PH/07 change above to avoid
3002 any possible encoding problems.
3004 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3005 but not after initializing Perl.
3007 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3008 output them only if debugging. By default they are written stderr,
3009 apparently, which is not desirable.
3011 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3014 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3017 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3019 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3020 authenticated or an ident call has been made. Suppress the default
3021 values for $authenticated_id and $authenticated_sender (but permit -oMai
3022 and -oMas) when testing with -bh.
3024 PH/24 Re-jigged the order of the tests in the default configuration so that the
3025 tests for valid domains and recipients precede the DNS black list and CSA
3026 tests, on the grounds that those ones are more expensive.
3028 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3029 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3030 This bug exists in every version of Exim that I still have, right back to
3033 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3034 However, an attempt to turn on case-sensitivity in a regex key by
3035 including (?-i) didn't work because the subject string was already
3036 lowercased, and the effects were non-intuitive. It turns out that a
3037 one-line patch can be used to allow (?-i) to work as expected.
3043 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3044 other effects) broke the use of negated acl sub-conditions.
3046 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3049 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3050 "Deny" causes Exim to reject the incoming connection with a 554 error.
3051 Unfortunately, if there is a major crisis, such as a disk failure,
3052 tcp-wrappers gives "deny", whereas what one would like would be some
3053 kind of temporary error. A kludge has been added to help with this.
3054 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3055 554 error is used if errno is still zero or contains ENOENT (which occurs
3056 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3059 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3061 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3062 errors. Otherwise a message that provokes a temporary error (when other
3063 messages do not) can cause a whole host to time out.
3065 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3066 addresses were routed directly to files or pipes from a redirect router.
3067 File deliveries just didn't batch; pipe deliveries might have suffered
3070 PH/06 A failure to get a lock for a hints database would erroneously always say
3071 "Failed to get write lock", even when it was really a read lock.
3073 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3074 of 0600. This has been changed to use the value of the lockfile_mode
3075 option (which defaults to 0600).
3077 PH/08 Applied small patch from the Sieve maintainer.
3079 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3080 folder from quota calculations, a direct delivery into this folder messed
3081 up the contents of the maildirsize file. This was because the regex was
3082 used only to exclude .Trash (or whatever) when the size of the mailbox
3083 was calculated. There was no check that a delivery was happening into an
3084 excluded directory. This bug has been fixed by ignoring all quota
3085 processing for deliveries into excluded directories.
3087 PH/10 Added the maildirfolder_create_regex option to appendfile.
3093 PH/01 The code for finding all the local interface addresses on a FreeBSD
3094 system running IPv6 was broken. This may well have applied to all BSD
3095 systems, as well as to others that have similar system calls. The broken
3096 code found IPv4 interfaces correctly, but gave incorrect values for the
3097 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3098 that it would not match correctly against @[] and not recognize the IPv6
3101 PH/02 The ipliteral router was not recognizing addresses of the form user@
3102 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3104 PH/03 Added disable_ipv6.
3106 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3107 decoded form, because it is most often used to construct To: headers
3108 lines in autoreplies, and the decoded form may well be syntactically
3109 invalid. However, $reply_address has leading white space removed, and all
3110 newlines turned into spaces so that the autoreply transport does not
3113 PH/05 If group was specified without a user on a router, and no group or user
3114 was specified on a transport, the group from the router was ignored.
3116 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3117 for visible compile-time settings that can be used to change these
3118 numbers, for those that want even more. Backwards compatibility with old
3119 spool files has been maintained. However, going back to a previous Exim
3120 release will lost any variables that are in spool files.
3122 PH/07 Two small changes when running in the test harness: increase delay when
3123 passing a TCP/IP connection to a new process, in case the original
3124 process has to generate a bounce, and remove special handling of
3125 127.0.0.2 (sic), which is no longer necessary.
3127 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3128 be the same on different OS.
3130 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3133 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3134 whether --show-vars was specified or not
3136 JJ/02 exipick: Added support for new ACL variable spool format introduced
3139 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3140 syntactically invalid From: or Reply-to: line, and a filter used this to
3141 generate an autoreply, and therefore failed to obtain an address for the
3142 autoreply, Exim could try to deliver to a non-existent relative file
3143 name, causing unrelated and misleading errors. What now happens is that
3144 it logs this as a hard delivery error, but does not attempt to create a
3147 PH/11 The exinext utility has a -C option for testing purposes, but although
3148 the given file was scanned by exinext itself; it wasn't being passed on
3149 when Exim was called.
3151 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3152 an end-of-file indication when reading a command response.
3154 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3155 compiled. In many other places in Exim, IPv6 addresses are always
3156 recognized, so I have changed this. It also means that IPv4 domain
3157 literals of the form [IPV4:n.n.n.n] are now always recognized.
3159 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3160 used if the router is not running as root, for example, when verifying at
3161 ACL time, or when using -bh. The debugging output from this situation was
3162 non-existent - all you got was a failure to exec. I have made two
3165 (a) Failures to set uid/gid, the current directory, or a process leader
3166 in a subprocess such as that created by queryprogram now generate
3167 suitable debugging output when -d is set.
3169 (b) The queryprogram router detects when it is not running as root,
3170 outputs suitable debugging information if -d is set, and then runs
3171 the subprocess without attempting to change uid/gid.
3173 PH/15 Minor change to Makefile for building test_host (undocumented testing
3176 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3177 additional section of a DNS packet that returns MX or SRV records.
3178 Instead, it always explicitly searches for A/AAAA records. This avoids
3179 major problems that occur when a DNS server includes only records of one
3180 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3181 fixed another bug: if SRV records were looked up and the corresponding
3182 address records were *not* found in the additional section, the port
3183 values from the SRV records were lost.
3185 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3186 using the correct key (the original address) when searching the retry
3187 rules in order to find which one to use for generating the retry hint.
3189 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3190 adding the default one. Similarly, if it contains a Reply-To: header, the
3191 errors_reply_to option, if set, is not used.
3193 PH/19 When calculating a retry time, Exim used to measure the "time since
3194 failure" by looking at the "first failed" field in the retry record. Now
3195 it does not use this if it is later than than the arrival time of the
3196 message. Instead it uses the arrival time. This makes for better
3197 behaviour in cases where some deliveries succeed, thus re-setting the
3198 "first failed" field. An example is a quota failure for a huge message
3199 when small messages continue to be delivered. Without this change, the
3200 "time since failure" will always be short, possible causing more frequent
3201 delivery attempts for the huge message than are intended.
3202 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3204 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3205 $1, $2, $3) because the numerical variables can be reset during some
3206 expansion items (e.g. "match"), thereby losing the authentication data.
3208 PH/21 Make -bV show the size of off_t variables so that the test suite can
3209 decide whether to run tests for quotas > 2G.
3211 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3212 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3213 filecount value is greater than 2G or if a quota value is greater than 2G
3214 on a system where the size of off_t is not greater than 4, a panic error
3217 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3218 never match. The debug and -bh output now contains an explicit error
3219 message indicating a malformed IPv4 address or mask.
3221 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3222 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3223 PH/23 above applies.
3225 PH/25 Do not write to syslog when running in the test harness. The only
3226 occasion when this arises is a failure to open the main or panic logs
3227 (for which there is an explicit test).
3229 PH/26 Added the /no_tell option to "control=freeze".
3231 PH/27 If a host name lookup failed very early in a connection, for example, if
3232 the IP address matched host_lookup and the reverse lookup yielded a name
3233 that did not have a forward lookup, an error message of the form "no IP
3234 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3235 could be logged. Now it outputs the IP address instead of "NULL".
3237 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3238 allows the sender and the authenticated sender to be set when
3239 submitting a message from within Exim. Since child_open_exim() is
3240 documented for local_scan(), the new function should be too.
3242 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3243 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3244 results in an empty string is now treated as unset.
3246 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3248 PH/31 Added sender_verify_fail logging option.
3250 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3251 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3252 tidied the source and removed it altogether.
3254 PH/33 When a queue run was abandoned because the load average was too high, a
3255 log line was always written; now it is written only if the queue_run log
3256 selector is set. In addition, the log line for abandonment now contains
3257 information about the queue run such as the pid. This is always present
3258 in "start" and "stop" lines but was omitted from the "abandon" line.
3260 PH/34 Omit spaces between a header name and the colon in the error message that
3261 is given when verify = headers_syntax fails (if there are lots of them,
3262 the message gets confusing).
3264 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3265 names, as there are now some PTR records that contain slashes. This check
3266 is only to protect against broken name servers that fall over on strange
3267 characters, so the fact that it applies to all lookups doesn't matter.
3269 PH/36 Now that the new test suite is complete, we can remove some of the
3270 special code in Exim that was needed for the old test suite. For example,
3271 sorting DNS records because real resolvers return them in an arbitrary
3272 order. The new test suite's fake resolver always returns records in the
3275 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3276 bounces) except when queue_only is set, to avoid logging races between
3277 the different processes.
3279 PH/38 Panic-die if .include specifies a non-absolute path.
3281 PH/39 A tweak to the "H" retry rule from its user.
3283 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3284 a label. They prevented compilation on older perls.
3286 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3287 a warning to be raised on newish perls.
3289 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3290 on queue. Changes to match documented behaviour of showing count of
3291 messages matching specified criteria.
3293 PH/40 Changed the default ident timeout from 30s to 5s.
3295 PH/41 Added support for the use of login_cap features, on those BSD systems
3296 that have them, for controlling the resources used by pipe deliveries.
3298 PH/42 The content-scanning code uses fopen() to create files in which to put
3299 message data. Previously it was not paying any attention to the mode of
3300 the files. Exim runs with umask(0) because the rest of the code creates
3301 files with open(), and sets the required mode explicitly. Thus, these
3302 files were ending up world-writeable. This was not a big issue, because,
3303 being within the spool directory, they were not world-accessible. I have
3304 created a function called modefopen, which takes an additional mode
3305 argument. It sets umask(777), creates the file, chmods it to the required
3306 mode, then resets the umask. All the relevant calls to fopen() in the
3307 content scanning code have been changed to use this function.
3309 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3310 to 24 hours. This avoids potential overflow problems when processing G
3311 and H retry rules. I suspect nobody ever tinkers with this value.
3313 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3315 PH/45 When the plaintext authenticator is running as a client, the server's
3316 challenges are checked to ensure they are valid base64 strings. By
3317 default, the authentication attempt is cancelled if an invalid string is
3318 received. Setting client_ignore_invalid_base64 true ignores these errors.
3319 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3320 they are received. Thus, the responses can be made to depend on the
3321 challenges. If an invalid string is ignored, an empty string is placed in
3324 PH/46 Messages that are created by the autoreply transport now contains a
3325 References: header, in accordance with RFCs 2822 and 3834.
3327 PH/47 Added authenticated_sender_force to the smtp transport.
3329 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3331 PH/49 Installed latest patch from the Sieve maintainer.
3333 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3334 was also set, the appendfile transport was unnecessarily scanning a
3335 directory of message files (e.g. for maildir delivery) to find the count
3336 of files (along with the size), even though it did not need this
3337 information. It now does the scan only if it needs to find either the
3338 size of the count of files.
3340 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3342 PH/52 Two bugs concerned with error handling when the smtp transport is
3345 (i) Exim was not creating retry information for temporary errors given
3346 for individual recipients after the DATA command when the smtp transport
3347 was used in LMTP mode. This meant that they could be retried too
3348 frequently, and not timed out correctly.
3350 (ii) Exim was setting the flag that allows error details to be returned
3351 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3352 recipients that were returned after the DATA command.
3354 PH/53 This is related to PH/52, but is more general: for any failing address,
3355 when detailed error information was permitted to be returned to the
3356 sender, but the error was temporary, then after the final timeout, only
3357 "retry timeout exceeded" was returned. Now it returns the full error as
3358 well as "retry timeout exceeded".
3360 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3361 do this, and (what is worse) MTAs that accept it.
3363 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3364 will now be deprecated.
3366 PH/56 New os.c-cygwin from the Cygwin maintainer.
3368 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3369 formats (previously only available in exim formats via -bpr, -bpru,
3370 and -bpra. Now also available in native and exiqgrep formats)
3372 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3373 with very large, slow to parse queues
3375 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3377 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3379 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3380 responses to authentication challenges, though it was showing the
3381 challenges; (ii) I've removed the CR characters from the debug output for
3384 PH/58 Allow for the insertion of a newline as well as a space when a string
3385 is turned into more than one encoded-word during RFC 2047 encoding. The
3386 Sieve code now uses this.
3388 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3389 data_4xx, lost_connection, tls_required.
3391 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3392 message was being sent as an SMTP response.
3394 PH/61 Add -l and -k options to exicyclog.
3396 PH/62 When verifying, if an address was redirected to one new address, so that
3397 verification continued, and the new address failed or deferred after
3398 having set something in $address_data, the value of $address_data was not
3399 passed back to the ACL. This was different to the case when no
3400 redirection occurred. The value is now passed back in both cases.
3402 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3403 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3404 use login_cap.h, so on its own it isn't the distinguishing feature. The
3405 new name refers directly to the setclassresources() function.
3407 PH/65 Added configuration files for NetBSD3.
3409 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3411 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3412 is preferred over IPv4.
3414 PH/68 The bounce_return_message and bounce_return_body options were not being
3415 honoured for bounces generated during the reception of non-SMTP messages.
3416 In particular, this applied to messages rejected by the ACL. This bug has
3417 been fixed. However, if bounce_return_message is true and bounce_return_
3418 body is false, the headers that are returned for a non-SMTP message
3419 include only those that have been read before the error was detected.
3420 (In the case of an ACL rejection, they have all been read.)
3422 PH/69 The HTML version of the specification is now built in a directory called
3423 spec_html instead of spec.html, because the latter looks like a path with
3424 a MIME-type, and this confuses some software.
3426 PH/70 Catch two compiler warnings in sieve.c.
3428 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3429 function verify_get_ident() calls ip_connect() to connect a socket, but
3430 if the "connect()" function timed out, ip_connect() used to close the
3431 socket. However, verify_get_ident() also closes the socket later, and in
3432 between Exim writes to the log, which may get opened at this point. When
3433 the socket was closed in ip_connect(), the log could get the same file
3434 descriptor number as the socket. This naturally causes chaos. The fix is
3435 not to close the socket in ip_connect(); the socket should be closed by
3436 the function that creates it. There was only one place in the code where
3437 this was missing, in the iplookup router, which I don't think anybody now
3438 uses, but I've fixed it anyway.
3440 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3441 well as to direct DNS lookups. Otherwise the handling of names in host
3442 lists is inconsistent and therefore confusing.
3448 PH/01 Two changes to the default runtime configuration:
3450 (1) Move the checks for relay_from_hosts and authenticated clients from
3451 after to before the (commented out) DNS black list checks.
3453 (2) Add control=submission to the relay_from_hosts and authenticated
3454 clients checks, on the grounds that messages accepted by these
3455 statements are most likely to be submissions.
3457 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3459 (1) Generate an error if the third argument for the ${prvs expansion is
3462 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3465 (3) Reset the variables that are obtained from the first argument of
3466 ${prvscheck and used in the second argument before leaving the code,
3467 because their memory is reclaimed, so using them afterwards may do
3470 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3471 one (it's much easier than Tom thought :-).
3473 (5) Because of (4), we can now allow for the use of $prvscheck_result
3474 inside the third argument.
3476 PH/03 For some reason, the default setting of PATH when running a command from
3477 a pipe transport was just "/usr/bin". I have changed it to
3480 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3481 anything to be listed in the output from -bV.
3483 PH/05 When a filter generated an autoreply, the entire To: header line was
3484 quoted in the delivery log line, like this:
3486 => >A.N.Other <ano@some.domain> <original@ddress> ...
3488 This has been changed so that it extracts the operative address. There
3489 may be more than one such address. If so, they are comma-separated, like
3492 => >ano@some.domain,ona@other.domain <original@ddress> ...
3494 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3495 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3496 not being looked up in the rDNS to get a host name, Exim was showing the
3497 IP address twice in Received: lines, even though the IP addresses were
3498 identical. For example:
3500 Received: from [1.2.3.4] (helo=[1.2.3.4])
3502 However, if the real host name was known, it was omitting the HELO data
3503 if it matched the actual IP address. This has been tidied up so that it
3504 doesn't show the same IP address twice.
3506 PH/07 When both +timestamp and +memory debugging was on, the value given by
3507 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3508 called by the debug printing, thereby overwriting the timestamp buffer.
3509 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3511 PH/08 When the original message was included in an autoreply transport, it
3512 always said "this is a copy of the message, including all the headers",
3513 even if body_only or headers_only was set. It now gives an appropriate
3516 PH/09 Applied a patch from the Sieve maintainer which:
3518 o fixes some comments
3519 o adds the (disabled) notify extension core
3520 o adds some debug output for the result of if/elsif tests
3521 o points to the current vacation draft in the documentation
3522 and documents the missing references header update
3526 o fixes a bug in processing the envelope test (when testing
3527 multiple envelope elements, the last element determined the
3530 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3531 Electronic Mail") by including:
3533 Auto-submitted: auto-generated
3535 in the messages that it generates (bounce messages and others, such as
3536 warnings). In the case of bounce messages for non-SMTP messages, there was
3537 also a typo: it was using "Auto_submitted" (underscore instead of
3538 hyphen). Since every message generated by Exim is necessarily in response
3539 to another message, thes have all been changed to:
3541 Auto-Submitted: auto-replied
3543 in accordance with these statements in the RFC:
3545 The auto-replied keyword:
3547 - SHOULD be used on messages sent in direct response to another
3548 message by an automatic process,
3550 - MUST NOT be used on manually-generated messages,
3552 - MAY be used on Delivery Status Notifications (DSNs) and Message
3553 Disposition Notifications (MDNs),
3555 - MUST NOT be used on messages generated by automatic or periodic
3556 processes, except for messages which are automatic responses to
3559 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3560 to the default Received: header definition.
3562 PH/12 Added log selector acl_warn_skipped (default on).
3564 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3565 variables because (a) they are in the wrong storage pool and (b) even if
3566 they were copied, it wouldn't work properly because of the caching.
3568 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3569 checking when decoding. Apparently there are clients that generate
3570 overlong encoded strings. Why am I not surprised?
3572 PH/15 If the first argument of "${if match_address" was not empty, but did not
3573 contain an "@" character, Exim crashed. Now it writes a panic log message
3574 and treats the condition as false.
3576 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3578 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3579 extension "envelope-auth". The code is finished and in agreement with
3580 other implementations, but there is no documentation so far and in fact,
3581 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3582 not changing the active code.
3584 Print executed "if" and "elsif" statements when debugging is used. This
3585 helps a great deal to understand what a filter does.
3587 Document more things not specified clearly in RFC3028. I had all this
3588 sorted out, when out of a sudden new issues came to my mind. Oops."
3590 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3593 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3594 canonical form (as documented). However, after a host name lookup from
3595 the IP address, check_host() was doing a simple string comparison with
3596 addresses acquired from the DNS when checking that the found name did
3597 have the original IP as one of its addresses. Since any found IPv6
3598 addresses are likely to be in abbreviated form, the comparison could
3599 fail. Luckily, there already exists a function for doing the comparison
3600 by converting both addresses to binary, so now that is used instead of
3601 the text comparison.
3603 PH/20 There was another similar case to PH/19, when a complete host name was
3604 given in a host list; looking up its IP address could give an abbreviated
3605 form, whereas the current host's name might or might not be abbreviated.
3606 The same fix has been applied.
3612 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3613 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3616 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3617 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3619 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3621 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3622 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3623 Also, for "Auto-Submitted", treat anything other than "no" as
3624 non-personal, in accordance with RFC 3834. (Previously it treated
3625 anything starting "auto-" as non-personal.)
3627 TF/01 The control=submission/name=... option had a problem with syntax
3628 errors if the name included a slash character. The /name= option
3629 now slurps the rest of the string, so it can include any characters
3630 but it must come last in the list of options (after /sender_retain
3633 PH/05 Some modifications to the interface to the fake nameserver for the new
3641 TK/01 Added the "success_on_redirect" address verification option. See
3642 NewStuff for rationale and an example.
3644 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3646 PH/02 Patch to exigrep to allow it to work on syslog lines.
3648 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3649 fread() to skip over the body file's header line, because in Cygwin the
3650 header line is locked and is inaccessible.
3652 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3653 co-exist for some time) to make it clear that it is the Exim ID that is
3654 referenced, not the Message-ID: header line.
3656 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3657 string_format() function, because snprintf() does not exist on all
3660 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3661 ${stat: expansion item.
3663 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3664 protocol synchronization error", to keep the pedants happy.
3666 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3667 well as for IRIX systems, when gcc is being used. See the host.c source
3670 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3672 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3675 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3676 transport and to the smtp transport in LMTP mode.
3678 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3680 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3681 run a filter in a subprocess. This could lead to confusion in subsequent
3682 lookups in the parent process. There should also be a search_tidyup() at
3683 the end of the subprocess.
3685 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3686 only if the host matched helo_try_verify_hosts, which caused the
3687 verification to occur when the EHLO/HELO command was issued. The ACL just
3688 tested the remembered result. Now, if a previous verification attempt has
3689 not happened, "verify = helo" does it there and then.
3691 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3693 TK/03 Fix log output including CR from clamd.
3695 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3696 exist provoked a memory error which could cause a segfault.
3698 PH/15 Installed PCRE 6.2
3700 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3702 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3703 of the problem. Specifically, suggested +O2 rather than +O1 for the
3706 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3708 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3709 record, the port was not correctly logged when the outgoing_port log
3710 selector was set (it logged the transort's default port).
3712 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3713 fallback_hosts, and "hosts" in the smtp transport.
3715 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3716 host errors such as "Connection refused".
3718 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3719 authentication with radiusclient 0.4.9:
3721 - Error returned from rc_read_config was caught wrongly
3722 - Username/password not passed on to radius server due to wrong length.
3724 The presumption is that some radiusclient API changes for 4.51/PH/17
3725 were not taken care of correctly. The code is still untested by me (my
3726 Linux distribution still has 0.3.2 of radiusclient), but it was
3727 contributed by a Radius user.
3729 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3730 expanding the "port" option of the smtp transport.
3732 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3733 while reading a MIME header. Thanks to Tom Hughes for a patch.
3735 PH/24 Include config.h inside local_scan.h so that configuration settings are
3738 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3739 that in an ACL for RCPT (for example), you can examine exactly what was
3742 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3743 commands, but it was not correctly comparing the address with the actual
3744 client host address. Thus, it would show the EHLO address in Received:
3745 header lines when this was not necessary.
3747 PH/27 Added the % operator to ${eval:}.
3749 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3750 it should be ignoring failures (because with -C, for example, it has lost
3751 privilege). It wasn't ignoring creation failures other than "already
3754 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3757 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3758 because an input line was too long, either on its own, or by virtue of
3759 too many continuations, the temporary file was not being removed, and the
3760 return code was incorrect.
3762 PH/31 Missing "BOOL" in function definition in filtertest.c.
3764 PH/32 Applied Sieve patches from the maintainer.
3766 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3768 PH/33 Added "verify = not_blind".
3770 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3771 Local/Makefile (with some defaults set). These are used in built scripts
3772 such as exicyclog, but they have never been used in the exim_install
3773 script (though there are many overriding facilities there). I have
3774 arranged that the exim_install script now takes note of these two
3777 PH/35 Installed configuration files for Dragonfly.
3779 PH/36 When a locally submitted message by a trusted user did not contain a
3780 From: header, and the sender address was obtained from -f or from an SMTP
3781 MAIL command, and the trusted user did not use -F to supply a sender
3782 name, $originator_name was incorrectly used when constructing a From:
3783 header. Furthermore, $originator_name was used for submission mode
3784 messages from external hosts without From: headers in a similar way,
3785 which is clearly wrong.
3787 PH/37 Added control=suppress_local_fixups.
3789 PH/38 When log_selector = +received_sender was set, and the addition of the
3790 sender made the log line's construction buffer exactly full, or one byte
3791 less than full, an overflow happened when the terminating "\n" was
3794 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3795 when the result of a list match is failure because a DNS lookup failed.
3797 PH/40 RM_COMMAND is now used in the building process.
3799 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3800 the "build-* directories that it finds.
3802 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3803 address in a domain literal was a prefix of an interface address.
3805 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3806 when verifying a sender address, unless rewrite_headers is false.
3808 PH/44 Wrote a long comment about why errors_to addresses are verified as
3809 recipients, not senders.
3811 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3812 the ratelimit ACL was added.
3814 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3816 PH/46 Added extra information about PostgreSQL errors to the error string.
3818 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3819 avoiding the need to install special zones in a real server. This is
3820 backwards compatible; if it can't find the fake resolver, it drops back.
3821 Thus, both old and new test suites can be run.
3823 TF/02 Added util/ratelimit.pl
3825 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3826 clock is set back in time.
3828 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3829 Candler <B.Candler@pobox.com>.
3831 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3832 OK for addresses that are the result of redirections.
3834 PH/48 A number of further additions for the benefit of the new test suite,
3835 including a fake gethostbyname() that interfaces to the fake DNS resolver
3838 TF/06 The fix for widen_domains has also been applied to qualify_single and
3839 search_parents which are the other dnslookup options that can cause
3842 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3845 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3847 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3848 service" if it encountered an MX record with an empty target hostname.
3849 The message is now "an MX or SRV record indicated no SMTP service".
3851 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3852 if the DNS of the sending site is misconfigured. This is quite a
3853 common situation. This change restores the behaviour of treating a
3854 helo verification defer as a failure.
3856 PH/51 If self=fail was set on a router, the bounce message did not include the
3857 actual error message.
3863 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3865 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3866 command itself ended in error, the underlying message about the transport
3867 filter timeout was being overwritten with the pipe command error. Now the
3868 underlying error message should be appended to the second error message.
3870 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3872 PH/02 Reduce the amount of output that "make" produces by default. Full output
3873 can still be requested.
3875 PH/03 The warning log line about a condition test deferring for a "warn" verb
3876 was being output only once per connection, rather than after each
3877 occurrence (because it was using the same function as for successful
3878 "warn" verbs). This seems wrong, so I have changed it.
3880 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3881 it should not have, which might have caused a crash in the right
3882 circumstances, but probably never did.
3884 PH/04 Installed a modified version of Tony Finch's patch to make submission
3885 mode fix the return path as well as the Sender: header line, and to
3886 add a /name= option so that you can make the user's friendly name appear
3889 TF/03 Added the control = fakedefer ACL modifier.
3891 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3892 Mark Lowes for thorough testing.
3894 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3896 TK/03 Merged latest SRS patch from Miles Wilton.
3898 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3899 with the definition in sysexits.h (which is #included earlier).
3900 Fortunately, Exim does not actually use EX_OK. The code used to try to
3901 preserve the sysexits.h value, by assuming that macro definitions were
3902 scanned for macro replacements. I have been disabused of this notion,
3903 so now the code just undefines EX_OK before #including unistd.h.
3905 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3906 in the smtp transport. When a block could not be written in a single
3907 write() function, the timeout was being re-applied to each part-write.
3908 This seems wrong - if the receiver was accepting one byte at a time it
3909 would take for ever. The timeout is now adjusted when this happens. It
3910 doesn't have to be particularly precise.
3912 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3913 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3915 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3916 without a domain if the check to <postmaster@domain> fails.
3918 SC/01 Eximstats: added -xls and the ability to specify output files
3919 (patch written by Frank Heydlauf).
3921 SC/02 Eximstats: use FileHandles for outputting results.
3923 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3925 SC/04 Eximstats: fixed display of large numbers with -nvr option
3927 SC/05 Eximstats: fixed merging of reports with empty tables.
3929 SC/06 Eximstats: added the -include_original_destination flag
3931 SC/07 Eximstats: removed tabs and trailing whitespace.
3933 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3935 TK/06 MBOX spool code: Add real "From " MBOX separator line
3936 so the .eml file is really in mbox format (even though
3937 most programs do not really care). Patch from Alex Miller.
3939 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3940 The latter is generated from $received_to and is only set if the
3941 message has one envelope recipient. SA can use these headers,
3942 obviously out-of-the-box. Patch from Alex Miller.
3944 PH/08 The ${def test on a variable was returning false if the variable's
3945 value was "0", contrary to what the specification has always said!
3946 The result should be true unless the variable is empty.
3948 PH/09 The syntax error of a character other than { following "${if
3949 def:variable_name" (after optional whitespace) was not being diagnosed.
3950 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3951 accidental colon was present, for example, could give incorrect results.
3953 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3954 result is used (not including appendfile, where other changes are about
3957 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3958 This involved changing a lot of size variables from int to off_t. It
3959 should work with maildirs and everything.
3961 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3962 spamd dying while we are connected to it.
3964 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3967 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3968 function for BDB 4.3.
3970 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3972 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3973 the sources for PCRE have been re-arranged and more files are now
3976 PH/15 The code I had for printing potentially long long variables in PH/11
3977 above was not the best (it lost precision). The length of off_t variables
3978 is now inspected at build time, and an appropriate printing format (%ld
3979 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3980 to be "long long int" or "long int". This is needed for the internal
3981 formatting function string_vformat().
3983 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3984 the configuration file to be ":syslog", then the script "guesses" where
3985 the logs files are, rather than using the compiled in default. In our
3986 case the guess is not the same as the compiled default, so the script
3987 suddenly stopped working when I started to use syslog. The patch checks
3988 to see if log_file_path is "". If so, it attempts to read it from exim
3989 with no configuration file to get the compiled in version, before it
3990 falls back to the previous guessing code."
3992 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3993 implementing BATV in an Exim configuration. See NewStuff for the gory
3996 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3997 Makefile that are specific to HP-UX.
3999 PH/18 If the "use_postmaster" option was set for a recipient callout together
4000 with the "random" option, the postmaster address was used as the MAIL
4001 FROM address for the random test, but not for the subsequent recipient
4002 test. It is now used for both.
4004 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4005 patch removes a few documentation additions to RFC 3028, because the
4006 latest draft now contains them. It adds the new en;ascii-case comparator
4007 and a new error check for 8bit text in MIME parts. Comparator and
4008 require names are now matched exactly. I enabled the subaddress
4009 extension, but it is not well tested yet (read: it works for me)."
4011 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4012 rework some of the code of TK/09 above to avoid the hardwired use of
4013 "%lld" and "long long". Replaced the call to snprintf() with a call to
4016 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4017 records point to non-existent hosts", "retry timeout exceeded", and
4018 "retry time not reached for any host after a long failure period".
4020 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4021 experimental DomainKeys support:
4023 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4024 (2) On an error such as an illegally used "control", the wrong name for
4025 the control was given.
4027 These problems did NOT occur unless DomainKeys support was compiled.
4029 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4031 PH/24 Added ${if match_ip condition.
4033 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4034 databases so that it will be absolutely obvious if a crash occurs in the
4035 DB library. This is a regular occurrence (often caused by mis-matched
4038 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4039 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4040 picked up on a user's system that detects such things. There doesn't seem
4041 to be a gcc warning option for this - only an attribute that has to be
4042 put on the function's prototype. It seems that in Fedora Core 4 they have
4043 set this on a number of new functions. No doubt there will be more in due
4046 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4047 specify a transport. However, if an address that was verified by such a
4048 router was the subject of a callout, Exim crashed because it tried to
4049 read the rcpt_include_affixes from the non-existent transport. Now it
4050 just assumes that the setting of that option is false. This bug was
4051 introduced by 4.51/PH/31.
4053 PH/28 Changed -d+all to exclude +memory, because that information is very
4054 rarely of interest, but it makes the output a lot bigger. People tend to
4055 do -d+all out of habit.
4057 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4058 code in os-type was giving problems when libc.so lives in lib64, like on
4061 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4062 aren't the modern standard, and it seems that some systems' include files
4063 don't always have them. Exim was already checking for some of the newer
4064 ones like T_AAAA, and defining it itself. I've added checks for all the
4065 record types that Exim uses.
4067 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4068 not automatically generating a new one, as it is supposed to. This
4069 prevented TLS from working. If the file did exist, but contained invalid
4070 data, a new version was generated, as expected. It was only the case of a
4071 non-existent file that was broken.
4073 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4074 with a change in libdomainkeys > 0.64.
4076 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4077 from DNS. If the selector record carries the flag, it now has
4078 precedence over the domain-wide flag.
4080 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4082 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4083 the use of a port name that isn't defined in /etc/services) occurred, the
4084 message was deferred as in a normal delivery, and thus remained on the
4085 spool, instead of being failed because of the mua_wrapper setting. This
4086 is now fixed, and I tidied up some of the mua_wrapper messages at the
4089 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4090 the messages in a hash of arrays rather than using individual hashes.
4091 This is a bit cleaner and results in dramatic memory savings, albeit
4092 at a slight CPU cost.
4094 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4095 as requested by Marc Sherman.
4097 SC/10 Eximstats: added histograms for user specified patterns as requested
4100 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4102 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4103 fopen() in the content-scanning modules that did not already have it.
4109 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4110 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4112 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4114 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4116 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4117 to test Sieve filters that use "vacation".
4119 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4120 that changes the way the GnuTLS parameters are stored in the cache file.
4121 The new format can be generated externally. For backward compatibility,
4122 if the data in the cache doesn't make sense, Exim assumes it has read an
4123 old-format file, and it generates new data and writes a new file. This
4124 means that you can't go back to an older release without removing the
4127 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4128 work if there are any delivery delays because "one_time" forces the
4129 parent to be marked "delivered", so its unseen clone is never tried
4130 again. For this reason, Exim now forbids the simultaneous setting of
4133 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4134 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4135 entry if you want to know the details. The fix, however, neglected to
4136 consider the case when local delivery batching is involved. The test for
4137 "previously delivered" was not happening when checking to see if an
4138 address could be batched with a previous (undelivered) one; under
4139 certain circumstances this could lead to multiple deliveries to the same
4142 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4143 in its include files, and this causes problems building Exim.
4145 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4146 header_syntax) but Exim was just ignoring anything given after a slash.
4147 In particular, this caused confusion with an attempt to use "verify =
4148 reverse_host_lookup/defer_ok". An error is now given when options are
4149 supplied for verify items that do not have them. (Maybe reverse_host_
4150 lookup should have a defer_ok option, but that's a different point.)
4152 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4153 defined by RFC 821) to 2048, because there were problems with some AUTH
4154 commands, and RFC 1869 says the size should be increased for extended
4155 SMTP commands that take arguments.
4157 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4160 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4161 "unknown" error; now it says that the functionality isn't in the binary.
4163 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4164 an address' error message when a string expansion fails (syntax or
4165 whatever). Otherwise the password may appear in the log. Following change
4166 PH/42 below, there is no longer a chance of it appearing in a bounce
4169 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4171 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4172 of its addresses was ever tried. (Bugzilla bug #2.)
4174 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4175 the result incorrectly in the debug output. (It correctly added a newline
4176 to what was transported.)
4178 TF/01 Added $received_time.
4180 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4181 commented out examples of how to interface to a virus scanner and to
4182 SpamAssassin. Also added commented examples of av_scanner and
4183 spamd_address settings.
4185 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4186 and controls are allowed in which ACLs. There were a couple of minor
4187 errors. Some of the entries in the conditions table (which is a table of
4188 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4189 negation of where the condition IS allowed.
4191 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4193 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4194 header file does not have a version number, so I've had to invent a new
4195 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4196 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4197 radiusclient), but it was contributed by a Radius user.
4199 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4200 files or directories, for OpenSSL.
4202 PH/19 When an Exim process that is running as root has to create an Exim log
4203 file, it does so in a subprocess that runs as exim:exim so as to get the
4204 ownership right at creation (otherwise, other Exim processes might see
4205 the file with the wrong ownership). There was no test for failure of this
4206 fork() call, which would lead to the process getting stuck as it waited
4207 for a non-existent subprocess. Forks do occasionally fail when resources
4208 run out. I reviewed all the other calls to fork(); they all seem to check
4211 PH/20 When checking for unexpected SMTP input at connect time (before writing
4212 the banner), Exim was not dealing correctly with a non-positive return
4213 from the read() function. If the client had disconnected by this time,
4214 the result was a log entry for a synchronization error with an empty
4215 string after "input=" when read() returned zero. If read() returned -1
4216 (an event I could not check), uninitialized data bytes were printed.
4217 There were reports of junk text (parts of files, etc) appearing after
4220 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4222 PH/22 Added support for macro redefinition, and (re)definition in between
4223 driver and ACL definitions.
4225 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4226 forgetting to use the resulting value; it was using the unexpanded value.
4228 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4229 hadn't been configured. The fix is from Juergen Kreileder, who
4230 understands it better than I do:
4232 "Here's what I see happening with three configured cyrus_sasl
4233 authenticators configured (plain, login, cram-md5):
4235 On startup auth_cyrus_sasl_init() gets called for each of these.
4236 This means three calls to sasl_listmech() without a specified mech_list.
4237 => SASL tests which mechs of all available mechs actually work
4238 => three warnings about OTP not working
4239 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4241 With the patch, sasl_listmech() also gets called three times. But now
4242 SASL's mech_list option is set to the server_mech specified in the the
4243 authenticator. Or in other words, the answer from sasl_listmech()
4244 gets limited to just the mech you're testing for (which is different
4246 => the return list contains just 'plain' or 'login', 'cram-md5' or
4247 nothing depending on the value of ob->server_mech.
4249 I've just tested the patch: Authentication still works fine,
4250 unavailable mechs specified in the exim configuration are still
4251 caught, and the auth.log warnings about OTP are gone."
4253 PH/25 When debugging is enabled, the contents of the command line are added
4254 to the debugging output, even when log_selector=+arguments is not
4257 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4258 answer is "GNU", and only if the return is "GNU/something" is the answer
4261 PH/27 $acl_verify_message is now set immediately after the failure of a
4262 verification in an ACL, and so is available in subsequent modifiers. In
4263 particular, the message can be preserved by coding like this:
4265 warn !verify = sender
4266 set acl_m0 = $acl_verify_message
4268 Previously, $acl_verify_message was set only while expanding "message"
4269 and "log_message" when a very denied access.
4271 PH/28 Modified OS/os.c-Linux with
4273 -#ifndef OS_LOAD_AVERAGE
4274 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4276 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4277 nomenclature these days.)
4279 PH/29 Installed patch from the Sieve maintainer that adds the options
4280 sieve_useraddress and sieve_subaddress to the redirect router.
4282 PH/30 In these circumstances:
4283 . Two addresses routed to the same list of hosts;
4284 . First host does not offer TLS;
4285 . First host accepts first address;
4286 . First host gives temporary error to second address;
4287 . Second host offers TLS and a TLS session is established;
4288 . Second host accepts second address.
4289 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4290 and peerdn, if requested) that were in fact used only for the second
4293 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4294 attention to any local part prefix or suffix that was matched by the
4295 router that accepted the address. It now behaves in the same way as it
4296 does for delivery: the affixes are removed from the local part unless
4297 rcpt_include_affixes is set on the transport.
4299 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4300 timeout during the DATA phase of an incoming message.
4302 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4303 applied a patch sanctioned by the Sieve maintainer.
4305 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4306 the uid or gid is negative. A case of a negative gid caused this to be
4307 noticed. The fix allows for either to be negative.
4309 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4310 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4313 PH/36 The change PH/12 above was broken. Fixed it.
4315 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4316 the grounds that routing the same address twice would always produce the
4317 same answer. This might have been true once, but it is certainly no
4318 longer true now. Routing a child address may depend on the previous
4319 routing that produced that child. Some complicated redirection strategies
4320 went wrong when messages had multiple recipients, and made Exim's
4321 behaviour dependent on the order in which the addresses were given.
4323 I have moved the duplicate checking until after the routing is complete.
4324 Exim scans the addresses that are assigned to local and remote
4325 transports, and removes any duplicates. This means that more work will be
4326 done, as duplicates will always all be routed, but duplicates are
4327 presumably rare, so I don't expect this is of any significance.
4329 For deliveries to pipes, files, and autoreplies, the duplicate checking
4330 still happens during the routing process, since they are not going to be
4333 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4334 It corrects a timeout issue with spamd. This is Ian's comment: "The
4335 background is that sometimes spamd either never reads data from a
4336 connection it has accepted, or it never writes response data. The exiscan
4337 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4338 blindly assumes that writes won't block so it may never time out."
4340 PH/39 Allow G after quota size as well as K and M.
4342 PH/40 The value set for $authenticated_id in an authenticator may not contain
4343 binary zeroes or newlines because the value is written to log lines and
4344 to spool files. There was no check on this. Now the value is run through
4345 the string_printing() function so that such characters are converted to
4346 printable escape sequences.
4348 PH/41 $message_linecount is a new variable that contains the total number of
4349 lines in the message. Compare $body_linecount, which is the count for the
4352 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4353 bounce and delay warning messages, except in certain special cases, which
4356 (a) An SMTP error message from a remote host;
4357 (b) A message specified in a :fail: redirection;
4358 (c) A message specified in a "fail" command in a system filter;
4359 (d) A message specified in a FAIL return from the queryprogram router;
4360 (e) A message specified by the cannot_route_message router option.
4362 In these cases only, Exim does include the error details in bounce and
4363 warning messages. There are also a few cases where bland messages such
4364 as "unrouteable address" or "local delivery error" are given.
4366 PH/43 $value is now also set for the "else" part of a ${run expansion.
4368 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4369 being worked on, but at least Exim now implements the latest version to
4372 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4373 process to complete was treated as a delivery failure, a timeout while
4374 writing the message to the pipe was logged, but erroneously treated as a
4375 successful delivery. Such timeouts include transport filter timeouts. For
4376 consistency with the overall process timeout, these timeouts are now
4377 treated as errors, giving rise to delivery failures by default. However,
4378 there is now a new Boolean option for the pipe transport called
4379 timeout_defer, which, if set TRUE, converts the failures into defers for
4380 both kinds of timeout. A transport filter timeout is now identified in
4383 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4384 systems where "make" and "gmake" are different, calling "gmake" at top
4385 level broke things. I've arranged for the value of $(MAKE) to be passed
4386 from the Makefile to this script so that it can call the same version of
4390 A note about Exim versions 4.44 and 4.50
4391 ----------------------------------------
4393 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4394 changes of various kinds. As a consequence, a big documentation update was
4395 needed. This delayed the release for rather longer than seemed good, especially
4396 in the light of a couple of (minor) security issues. Therefore, the changes
4397 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4398 release. So 4.44 and 4.50 are in effect two different branches that both start
4401 I have left the 4.50 change log unchanged; it contains all the changes since
4402 4.43. The change log for 4.44 is below; many of its items are identical to
4403 those for 4.50. This seems to be the most sensible way to preserve the
4404 historical information.
4410 1. Minor wording change to the doc/README.SIEVE file.
4412 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4413 computation of the current number of files was incorrect.
4415 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4416 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4419 4. Give more explanation in the error message when the command for a transport
4420 filter fails to execute.
4422 5. There are several places where Exim runs a non-Exim command in a
4423 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4424 was being done only for the command run by the queryprogram router. It is
4425 now done for all such subprocesses. The other cases are: ${run, transport
4426 filters, and the commands run by the lmtp and pipe transports.
4428 6. Added CONFIGURE_GROUP build-time option.
4430 7. Some older OS have a limit of 256 on the maximum number of file
4431 descriptors. Exim was using setrlimit() to set 1000 as a large value
4432 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4433 systems. I've change it so that if it can't get 1000, it tries for 256.
4435 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4436 was an oversight, and furthermore, ever since the addition of extra
4437 controls (e.g. 4.43/32), the checks on when to allow different forms of
4438 "control" were broken. There should now be diagnostics for all cases when a
4439 control that does not make sense is encountered.
4441 9. Added the /retain_sender option to "control=submission".
4443 10. $recipients is now available in the predata ACL (oversight).
4445 11. Tidy the search cache before the fork to do a delivery from a message
4446 received from the command line. Otherwise the child will trigger a lookup
4447 failure and thereby defer the delivery if it tries to use (for example) a
4448 cached ldap connection that the parent has called unbind on.
4450 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4451 of $address_data from the recipient verification was clobbered by the
4452 sender verification.
4454 13. The value of address_data from a sender verification is now available in
4455 $sender_address_data in subsequent conditions in the ACL statement.
4457 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4459 15. Added a new option "connect=<time>" to callout options, to set a different
4462 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4463 was its contents. (It was OK if the option was not defined at all.)
4465 17. A "Completed" log line is now written for messages that are removed from
4466 the spool by the -Mrm option.
4468 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4469 information about exactly what failed.
4471 19. Added -dd to debug only the daemon process.
4473 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4474 handles timeouts, both on the server side and network timeouts. Renamed the
4475 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4477 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4478 It is now set to "smtps".
4480 22. $host_address is now set to the target address during the checking of
4481 ignore_target_hosts.
4483 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4484 being passed; this would have caused $sender_host_name to have been used if
4485 matching the list had actually called for a host name (not very likely,
4486 since this list is usually IP addresses). A host name is now passed as
4489 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4490 code that specifies a non-restarting handler (typically sigaction() in
4491 modern systems) in an attempt to fix a rare and obscure crash bug.
4493 25. Narrowed the window for a race in the daemon that could cause it to ignore
4494 SIGCHLD signals. This is not a major problem, because they are used only to
4495 wake it up if nothing else does.
4497 26. A malformed maildirsize file could cause Exim to calculate negative values
4498 for the mailbox size or file count. Odd effects could occur as a result.
4499 The maildirsize information is now recalculated if the size or filecount
4502 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4503 support for a long time. Removed HAVE_SYS_VFS_H.
4505 28. Installed the latest version of exipick from John Jetmore.
4507 29. In an address list, if the pattern was not a regular expression, an empty
4508 subject address (from a bounce message) matched only if the pattern was an
4509 empty string. Non-empty patterns were not even tested. This was the wrong
4510 because it is perfectly reasonable to use an empty address as part of a
4511 database query. An empty address is now tested by patterns that are
4512 lookups. However, all the other forms of pattern expect the subject to
4513 contain a local part and a domain, and therefore, for them, an empty
4514 address still always fails if the pattern is not itself empty.
4516 30. Exim went into a mad DNS loop when attempting to do a callout where the
4517 host was specified on an smtp transport, and looking it up yielded more
4518 than one IP address.
4520 31. Re-factored the code for checking spool and log partition space into a
4521 function that finds that data and another that does the check. The former
4522 is then used to implement four new variables: $spool_space, $log_space,
4523 $spool_inodes, and $log_inodes.
4525 32. The RFC2047 encoding function was originally intended for short strings
4526 such as real names; it was not keeping to the 75-character limit for
4527 encoded words that the RFC imposes. It now respects the limit, and
4528 generates multiple encoded words if necessary. To be on the safe side, I
4529 have increased the buffer size for the ${rfc2047: expansion operator from
4532 33. It is now permitted to omit both strings after an "if" condition; if the
4533 condition is true, the result is "true". As before, when the second string
4534 is omitted, a false condition yields an empty string. This makes it less
4535 cumbersome to write custom ACL and router conditions.
4537 34. Failure to deliver a bounce message always caused it to be frozen, even if
4538 there was an errors_to setting on the router. The errors_to setting is now
4541 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4542 canonical form (fully expanded) before being placed in
4543 $sender_host_address.
4545 36. The table in the code that translates DNS record types into text (T_A to
4546 "A" for instance) was missing entries for NS and CNAME. It is just possible
4547 that this could have caused confusion if both these types were looked up
4548 for the same domain, because the text type is used as part of Exim's
4549 per-process caching. But the chance of anyone hitting this buglet seems
4552 37. The dnsdb lookup has been extended in a number of ways.
4554 (1) There is a new type, "zns", which walks up the domain tree until it
4555 finds some nameserver records. It should be used with care.
4557 (2) There is a new type, "mxh", which is like "mx" except that it returns
4558 just the host names, not the priorities.
4560 (3) It is now possible to give a list of domains (or IP addresses) to be
4561 looked up. The behaviour when one of the lookups defers can be
4562 controlled by a keyword.
4564 (4) It is now possible to specify the separator character for use when
4565 multiple records are returned.
4567 38. The dnslists ACL condition has been extended: it is now possible to supply
4568 a list of IP addresses and/or domains to be looked up in a particular DNS
4571 39. Added log_selector=+queue_time_overall.
4573 40. When running the queue in the test harness, wait just a tad after forking a
4574 delivery process, to get repeatability of debugging output.
4576 41. Include certificate and key file names in error message when GnuTLS fails
4577 to set them up, because the GnuTLS error message doesn't include the name
4578 of the failing file when there is a problem reading it.
4580 42. Allow both -bf and -bF in the same test run.
4582 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4584 44. The "Exiscan patch" is now merged into the mainline Exim source.
4586 45. Sometimes the final signoff response after QUIT could fail to get
4587 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4588 < 0 before doing a fflush(). This bug looks as though it goes back to the
4589 introduction of TLS in release 3.20, but "sometimes" must have been rare
4590 because the tests only now provoked it.
4592 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4593 (this can affect the format of dates).
4595 47. exim_tidydb, when checking for the continued existence of a message for
4596 which it has found a message-specific retry record, was not finding
4597 messages that were in split spool directories. Consequently, it was
4598 deleting retry records that should have stayed in existence.
4600 48. Steve fixed some bugs in eximstats.
4602 49. The SPA authentication driver was not abandoning authentication and moving
4603 on to the next authenticator when an expansion was forced to fail,
4604 contradicting the general specification for all authenticators. Instead it
4605 was generating a temporary error. It now behaves as specified.
4607 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4608 (the order specifies the preference for clients). The order is now AES256,
4609 AES128, 3DES, ARCFOUR128.
4611 51. Small patch to Sieve code - explicitly set From: when generating an
4614 52. Exim crashed if a remote delivery caused a very long error message to be
4615 recorded - for instance if somebody sent an entire SpamAssassin report back
4616 as a large number of 550 error lines. This bug was coincidentally fixed by
4617 increasing the size of one of Exim's internal buffers (big_buffer) that
4618 happened as part of the Exiscan merge. However, to be on the safe side, I
4619 have made the code more robust (and fixed the comments that describe what
4622 53. Now that there can be additional text after "Completed" in log lines (if
4623 the queue_time_overall log selector is set), a one-byte patch to exigrep
4624 was needed to allow it to recognize "Completed" as not the last thing in
4627 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4628 patch that reportedly fixes this has been added. I am not expert enough to
4629 create a test for it. This is what the patch creator wrote:
4631 "I found a little strange behaviour of ldap code when working with
4632 Windows 2003 AD Domain, where users was placed in more than one
4633 Organization Units. When I tried to give exim partial DN, the exit code
4634 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4635 But simultaneously result of request was absolutely normal ldap result,
4636 so I produce this patch..."
4638 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4639 REFERENCE, so I have modified the code to exclude the patch when that macro
4642 55. Some experimental protocols are using DNS PTR records for new purposes. The
4643 keys for these records are domain names, not reversed IP addresses. The
4644 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4645 leaves it alone. Component reversal etc. now happens only for IP addresses.
4648 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4650 57. Double the size of the debug message buffer (to 2048) so that more of very
4651 long debug lines gets shown.
4653 58. The exicyclog utility now does better if the number of log files to keep
4654 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4656 59. Two changes related to the smtp_active_hostname option:
4658 (1) $smtp_active_hostname is now available as a variable.
4659 (2) The default for smtp_banner uses $smtp_active_hostname instead
4660 of $primary_hostname.
4662 60. The host_aton() function is supposed to be passed a string that is known
4663 to be a valid IP address. However, in the case of IPv6 addresses, it was
4664 not checking this. This is a hostage to fortune. Exim now panics and dies
4665 if the condition is not met. A case was found where this could be provoked
4666 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4667 components; fortuitously, this particular loophole had already been fixed
4668 by change 4.50/55 above.
4670 If there are any other similar loopholes, the new check in host_aton()
4671 itself should stop them being exploited. The report I received stated that
4672 data on the command line could provoke the exploit when Exim was running as
4673 exim, but did not say which command line option was involved. All I could
4674 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4675 running as the user.
4678 61. There was a buffer overflow vulnerability in the SPA authentication code
4679 (which came originally from the Samba project). I have added a test to the
4680 spa_base64_to_bits() function which I hope fixes it.
4683 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4684 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4686 63. The daemon start-up calls getloadavg() while still root for those OS that
4687 need the first call to be done as root, but it missed one case: when
4688 deliver_queue_load_max is set with deliver_drop_privilege. This is
4689 necessary for the benefit of the queue runner, because there is no re-exec
4690 when deliver_drop_privilege is set.
4692 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4693 This has been fixed.
4695 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4696 was in use, was not putting the data itself into the right store pool;
4697 consequently, it could be overwritten for a subsequent message in the same
4698 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4701 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4703 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4704 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4705 the function were treating the return as a boolean value, which happened to
4706 work because 0=false and not-0=true, but is not correct code.
4708 68. The host_aton() function was not handling scoped IPv6 addresses (those
4709 with, for example, "%eth0" on the end) correctly.
4711 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4712 negated items (that is, ~something) in unsigned ints. Some compilers
4713 apparently mutter when there is no cast.
4715 70. If an address verification called from an ACL failed, and did not produce a
4716 user-specific message (i.e. there was only a "system" message), nothing was
4717 put in $acl_verify_message. In this situation, it now puts the system
4720 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4721 session; change 4.31/43 added the unwanted input to the log line - except
4722 that it did not do this in the start of session case. It now does.
4724 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4725 This is wrong and can cause the other end to generate a synchronization
4726 error if it is another Exim or anything else that does the synchronization
4727 check. A QUIT command is no longer sent after a timeout.
4729 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4730 during host lookups.
4732 74. The defer_ok option of callout verification was not working if it was used
4733 when verifying addresses in header lines, that is, for this case:
4735 verify = header_sender/callout=defer_ok
4737 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4738 those file descriptors could be used for SMTP connections. If anything
4739 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4740 could be sent to the SMTP client, causing chaos. The daemon now opens
4741 stdin, stdout, and stderr to /dev/null when it puts itself into the
4744 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4745 log by default. The user can override this with suitable Perl magic.
4747 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4748 the log message when discard triggers, was not working for the DATA ACL or
4749 for the non-SMTP ACL.
4751 78. Error message wording change in sieve.c.
4753 79. If smtp_accept_max_per_host was set, the number of connections could be
4754 restricted to fewer than expected, because the daemon was trying to set up
4755 a new connection before checking whether the processes handling previous
4756 connections had finished. The check for completed processes is now done
4757 earlier. On busy systems, this bug wouldn't be noticed because something
4758 else would have woken the daemon, and it would have reaped the completed
4761 80. If a message was submitted locally by a user whose login name contained one
4762 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4763 It caused a spool format error. I have fixed the spool reading code. A
4764 related problem was that the "from" clause in the Received: line became
4765 illegal because of the space(s). It is now covered by ${quote_local_part.
4767 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4770 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4772 83. Patch from Sieve maintainer to fix unterminated string problem in
4773 "vacation" handling.
4775 84. Some minor changes to the Linux configuration files to help with other
4776 OS variants using glibc.
4778 85. One more patch for Sieve to update vacation handling to latest spec.
4781 ----------------------------------------------------
4782 See the note above about the 4.44 and 4.50 releases.
4783 ----------------------------------------------------
4789 1. Change 4.43/35 introduced a bug that caused file counts to be
4790 incorrectly computed when quota_filecount was set in an appendfile
4793 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4794 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4797 3. Give more explanation in the error message when the command for a transport
4798 filter fails to execute.
4800 4. There are several places where Exim runs a non-Exim command in a
4801 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4802 was being done only for the command run by the queryprogram router. It is
4803 now done for all such subprocesses. The other cases are: ${run, transport
4804 filters, and the commands run by the lmtp and pipe transports.
4806 5. Some older OS have a limit of 256 on the maximum number of file
4807 descriptors. Exim was using setrlimit() to set 1000 as a large value
4808 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4809 systems. I've change it so that if it can't get 1000, it tries for 256.
4811 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4812 was an oversight, and furthermore, ever since the addition of extra
4813 controls (e.g. 4.43/32), the checks on when to allow different forms of
4814 "control" were broken. There should now be diagnostics for all cases when a
4815 control that does not make sense is encountered.
4817 7. $recipients is now available in the predata ACL (oversight).
4819 8. Tidy the search cache before the fork to do a delivery from a message
4820 received from the command line. Otherwise the child will trigger a lookup
4821 failure and thereby defer the delivery if it tries to use (for example) a
4822 cached ldap connection that the parent has called unbind on.
4824 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4825 of $address_data from the recipient verification was clobbered by the
4826 sender verification.
4828 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4829 was its contents. (It was OK if the option was not defined at all.)
4831 11. A "Completed" log line is now written for messages that are removed from
4832 the spool by the -Mrm option.
4834 12. $host_address is now set to the target address during the checking of
4835 ignore_target_hosts.
4837 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4838 being passed; this would have caused $sender_host_name to have been used if
4839 matching the list had actually called for a host name (not very likely,
4840 since this list is usually IP addresses). A host name is now passed as
4843 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4844 code that specifies a non-restarting handler (typically sigaction() in
4845 modern systems) in an attempt to fix a rare and obscure crash bug.
4847 15. Narrowed the window for a race in the daemon that could cause it to ignore
4848 SIGCHLD signals. This is not a major problem, because they are used only to
4849 wake it up if nothing else does.
4851 16. A malformed maildirsize file could cause Exim to calculate negative values
4852 for the mailbox size or file count. Odd effects could occur as a result.
4853 The maildirsize information is now recalculated if the size or filecount
4856 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4857 support for a long time. Removed HAVE_SYS_VFS_H.
4859 18. Updated exipick to current release from John Jetmore.
4861 19. Allow an empty sender to be matched against a lookup in an address list.
4862 Previously the only cases considered were a regular expression, or an
4865 20. Exim went into a mad DNS lookup loop when doing a callout where the
4866 host was specified on the transport, if the DNS lookup yielded more than
4869 21. The RFC2047 encoding function was originally intended for short strings
4870 such as real names; it was not keeping to the 75-character limit for
4871 encoded words that the RFC imposes. It now respects the limit, and
4872 generates multiple encoded words if necessary. To be on the safe side, I
4873 have increased the buffer size for the ${rfc2047: expansion operator from
4876 22. Failure to deliver a bounce message always caused it to be frozen, even if
4877 there was an errors_to setting on the router. The errors_to setting is now
4880 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4881 canonical form (fully expanded) before being placed in
4882 $sender_host_address.
4884 24. Updated eximstats to version 1.33
4886 25. Include certificate and key file names in error message when GnuTLS fails
4887 to set them up, because the GnuTLS error message doesn't include the name
4888 of the failing file when there is a problem reading it.
4890 26. Expand error message when OpenSSL has problems setting up cert/key files.
4893 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4894 (this can affect the format of dates).
4896 28. exim_tidydb, when checking for the continued existence of a message for
4897 which it has found a message-specific retry record, was not finding
4898 messages that were in split spool directories. Consequently, it was
4899 deleting retry records that should have stayed in existence.
4901 29. eximstats updated to version 1.35
4902 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4903 1.35 - bugfix such that pie charts by volume are generated correctly
4905 30. The SPA authentication driver was not abandoning authentication and moving
4906 on to the next authenticator when an expansion was forced to fail,
4907 contradicting the general specification for all authenticators. Instead it
4908 was generating a temporary error. It now behaves as specified.
4910 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4911 (the order specifies the preference for clients). The order is now AES256,
4912 AES128, 3DES, ARCFOUR128.
4914 31. Small patch to Sieve code - explicitly set From: when generating an
4917 32. Exim crashed if a remote delivery caused a very long error message to be
4918 recorded - for instance if somebody sent an entire SpamAssassin report back
4919 as a large number of 550 error lines. This bug was coincidentally fixed by
4920 increasing the size of one of Exim's internal buffers (big_buffer) that
4921 happened as part of the Exiscan merge. However, to be on the safe side, I
4922 have made the code more robust (and fixed the comments that describe what
4925 33. Some experimental protocols are using DNS PTR records for new purposes. The
4926 keys for these records are domain names, not reversed IP addresses. The
4927 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4928 leaves it alone. Component reversal etc. now happens only for IP addresses.
4931 34. The host_aton() function is supposed to be passed a string that is known
4932 to be a valid IP address. However, in the case of IPv6 addresses, it was
4933 not checking this. This is a hostage to fortune. Exim now panics and dies
4934 if the condition is not met. A case was found where this could be provoked
4935 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4936 components; fortuitously, this particular loophole had already been fixed
4937 by change 4.50/55 or 4.44/33 above.
4939 If there are any other similar loopholes, the new check in host_aton()
4940 itself should stop them being exploited. The report I received stated that
4941 data on the command line could provoke the exploit when Exim was running as
4942 exim, but did not say which command line option was involved. All I could
4943 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4944 running as the user.
4947 35. There was a buffer overflow vulnerability in the SPA authentication code
4948 (which came originally from the Samba project). I have added a test to the
4949 spa_base64_to_bits() function which I hope fixes it.
4952 36. The daemon start-up calls getloadavg() while still root for those OS that
4953 need the first call to be done as root, but it missed one case: when
4954 deliver_queue_load_max is set with deliver_drop_privilege. This is
4955 necessary for the benefit of the queue runner, because there is no re-exec
4956 when deliver_drop_privilege is set.
4958 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4959 was in use, was not putting the data itself into the right store pool;
4960 consequently, it could be overwritten for a subsequent message in the same
4961 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4964 38. Sometimes the final signoff response after QUIT could fail to get
4965 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4966 < 0 before doing a fflush(). This bug looks as though it goes back to the
4967 introduction of TLS in release 3.20, but "sometimes" must have been rare
4968 because the tests only now provoked it.
4974 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4975 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4976 Now it returns an integer. A number of calls were still expecting a T/F
4977 return. Fortuitously, in all cases, the tests worked in OK situations,
4978 which is the norm. However, things would have gone wrong on any write
4979 failures on the smtp file descriptor. This function is used when sending
4980 messages over SMTP and also when doing verify callouts.
4982 2. When Exim is called to do synchronous delivery of a locally submitted
4983 message (the -odf or -odi options), it no longer closes stderr before doing
4986 3. Implemented the mua_wrapper option.
4988 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4990 5. Implemented the functions header_remove(), header_testname(),
4991 header_add_at_position(), and receive_remove_recipient(), and exported them
4994 6. If an ACL "warn" statement specified the addition of headers, Exim already
4995 inserted X-ACL-Warn: at the start if there was no header name. However, it
4996 was not making this test for the second and subsequent header lines if
4997 there were newlines in the string. This meant that an invalid header could
4998 be inserted if Exim was badly configured.
5000 7. Allow an ACL "warn" statement to add header lines at the start or after all
5001 the Received: headers, as well as at the end.
5003 8. Added the rcpt_4xx retry error code.
5005 9. Added postmaster_mailfrom=xxx to callout verification option.
5007 10. Added mailfrom=xxxx to the callout verification option, for verify=
5010 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5011 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5013 12. Inserted some casts to stop certain compilers warning when using pointer
5014 differences as field lengths or precisions in printf-type calls (mostly
5015 affecting debugging statements).
5017 13. Added optional readline() support for -be (dynamically loaded).
5019 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5020 same clock tick as a message's arrival, so that its received time was the
5021 same as the "first fail" time on the retry record, and that message
5022 remained on the queue past the ultimate address timeout, every queue runner
5023 would try a delivery (because it was past the ultimate address timeout) but
5024 after another failure, the ultimate address timeout, which should have then
5025 bounced the address, did not kick in. This was a "< instead of <=" error;
5026 in most cases the first failure would have been in the next clock tick
5027 after the received time, and all would be well.
5029 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5030 being recognized when the domain list was tested by the match_domain
5031 condition in an expansion string.
5033 16. Added the ${str2b64: operator.
5035 17. Exim was always calling setrlimit() to set a large limit for the number of
5036 processes, without checking whether the existing limit was already
5037 adequate. (It did check for the limit on file descriptors.) Furthermore,
5038 errors from getrlimit() and setrlimit() were being ignored. Now they are
5039 logged to the main and panic logs, but Exim does carry on, to try to do its
5040 job under whatever limits there are.
5042 18. Imported PCRE 5.0.
5044 19. Trivial typo in log message " temporarily refused connection" (the leading
5047 20. If the log selector return_path_on_delivery was set and an address was
5048 redirected to /dev/null, the delivery process crashed because it assumed
5049 that a return path would always be set for a "successful" delivery. In this
5050 case, the whole delivery is bypassed as an optimization, and therefore no
5053 21. Internal re-arrangement: the function for sending a challenge and reading
5054 a response while authentication was assuming a zero-terminated challenge
5055 string. It's now changed to take a pointer and a length, to allow for
5056 binary data in such strings.
5058 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5060 23. Exim was not respecting finduser_retries when seeking the login of the
5061 uid under which it was called; it was always trying 10 times. (The default
5062 setting of finduser_retries is zero.) Also, it was sleeping after the final
5063 failure, which is pointless.
5065 24. Implemented tls_on_connect_ports.
5067 25. Implemented acl_smtp_predata.
5069 26. If the domain in control=submission is set empty, Exim assumes that the
5070 authenticated id is a complete email address when it generates From: or
5071 Sender: header lines.
5073 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5074 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5075 chown and chgrp in /bin and hostname in /usr/bin.
5077 28. Exim was keeping the "process log" file open after each use, just as it
5078 does for the main log. This opens the possibility of it remaining open for
5079 long periods when the USR1 signal hits a daemon. Occasional processlog
5080 errors were reported, that could have been caused by this. Anyway, it seems
5081 much more sensible not to leave this file open at all, so that is what now
5084 29. The long-running daemon process does not normally write to the log once it
5085 has entered its main loop, and it closes the log before doing so. This is
5086 so that log files can straightforwardly be renamed and moved. However,
5087 there are a couple of unusual error situations where the daemon does write
5088 log entries, and I had neglected to close the log afterwards.
5090 30. The text of an SMTP error response that was received during a remote
5091 delivery was being truncated at 512 bytes. This is too short for some of
5092 the long messages that one sometimes sees. I've increased the limit to
5095 31. It is now possible to make retry rules that apply only when a message has a
5096 specific sender, in particular, an empty sender.
5098 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5099 it possible to be selective about when SMTP synchronization is enforced.
5101 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5103 32. Implemented hosts_connection_nolog.
5105 33. Added an ACL for QUIT.
5107 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5110 35. Added mailbox_size and mailbox_filecount to appendfile.
5112 36. Added control = no_multiline_responses to ACLs.
5114 37. There was a bug in the logic of the code that waits for the clock to tick
5115 in the case where the clock went backwards by a substantial amount such
5116 that the microsecond fraction of "now" was more than the microsecond
5117 fraction of "then" (but the whole seconds number was less).
5119 38. Added support for the libradius Radius client library this is found on
5120 FreeBSD (previously only the radiusclient library was supported).
5126 1. When certain lookups returned multiple values in the form name=value, the
5127 quoting of the values was not always being done properly. Specifically:
5128 (a) If the value started with a double quote, but contained no whitespace,
5130 (b) If the value contained whitespace other than a space character (i.e.
5131 tabs or newlines or carriage returns) it was not quoted.
5132 This fix has been applied to the mysql and pgsql lookups by writing a
5133 separate quoting function and calling it from the lookup code. The fix
5134 should probably also be applied to nisplus, ibase and oracle lookups, but
5135 since I cannot test any of those, I have not disturbed their existing code.
5137 2. A hit in the callout cache for a specific address caused a log line with no
5138 reason for rejecting RCPT. Now it says "Previous (cached) callout
5139 verification failure".
5141 3. There was an off-by-one bug in the queryprogram router. An over-long
5142 return line was truncated at 256 instead of 255 characters, thereby
5143 overflowing its buffer with the terminating zero. As well as fixing this, I
5144 have increased the buffer size to 1024 (and made a note to document this).
5146 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5147 when Exim is waiting for an SMTP response from a remote server, Exim
5148 restarts its select() call on the socket, thereby resetting its timeout.
5149 This is not a problem when such interrupts are rare. Somebody set up a cron
5150 job to run exiwhat every 2 minutes, which is less than the normal select()
5151 timeout (5 or 10 minutes). This meant that the select() timeout never
5152 kicked in because it was always reset. I have fixed this by comparing the
5153 time when an interrupt arrives with the time at the start of the first call
5154 to select(). If more time than the timeout has elapsed, the interrupt is
5155 treated as a timeout.
5157 5. Some internal re-factoring in preparation for the addition of Sieve
5158 extensions (by MH). In particular, the "personal" test is moved to a
5159 separate function, and given an option for scanning Cc: and Bcc: (which is
5160 not set for Exim filters).
5162 6. When Exim created an email address using the login of the caller as the
5163 local part (e.g. when creating a From: or Sender: header line), it was not
5164 quoting the local part when it contained special characters such as @.
5166 7. Installed new OpenBSD configuration files.
5168 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5169 try to make them clearer.
5171 9. Callout options, other than the timeout value, were being ignored when
5172 verifying sender addresses in header lines. For example, when using
5174 verify = header_sender/callout=no_cache
5176 the cache was (incorrectly) being used.
5178 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5179 only the Cygwin environment.
5181 11. When return_path_on_delivery was set as a log selector, if different remote
5182 addresses in the same message used different return paths and parallel
5183 remote delivery occurred, the wrong values would sometimes be logged.
5184 (Whenever a remote delivery process finished, the return path value from
5185 the most recently started remote delivery process was logged.)
5187 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5188 lines when AUTH and/or TLS are in use. This is the "received protocol"
5189 field. Exim used to use "asmtp" for authenticated SMTP, without any
5190 indication (in the protocol name) for TLS use. Now it follows the RFC and
5191 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5192 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5193 names appear in log lines as well as in Received: header lines.
5195 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5196 extensions, and comparison tests, and to fix some bugs.
5198 14. Changes to the "personal" filter test:
5200 (1) The test was buggy in that it was just doing the equivalent of
5201 "contains" tests on header lines. For example, if a user's address was
5202 anne@some.where, the "personal" test would incorrectly be true for
5204 To: susanne@some.where
5206 This test is now done by extracting each address from the header in turn,
5207 and checking the entire address. Other tests that are part of "personal"
5208 are now done using regular expressions (for example, to check local parts
5209 of addresses in From: header lines).
5211 (2) The list of non-personal local parts in From: addresses has been
5212 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5213 taken from the Sieve specification recommendations.
5215 (3) If the message contains any header line starting with "List-" it is
5216 treated as non-personal.
5218 (4) The test for "circular" in the Subject: header line has been removed
5219 because it now seems ill-conceived.
5221 15. Minor typos in src/EDITME comments corrected.
5223 16. Installed latest exipick from John Jetmore.
5225 17. If headers_add on a router specified a text string that was too long for
5226 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5227 of string_sprintf() is now avoided.
5229 18. $message_body_size was not set (it was always zero) when running the DATA
5230 ACL and the local_scan() function.
5232 19. For the "mail" command in an Exim filter, no default was being set for
5233 the once_repeat time, causing a random time value to be used if "once" was
5234 specified. (If the value happened to be <= 0, no repeat happened.) The
5235 default is now 0s, meaning "never repeat". The "vacation" command was OK
5236 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5237 (I found it when inspecting the code).
5239 20. There is now an overall timeout for performing a callout verification. It
5240 defaults to 4 times the callout timeout, which applies to individual SMTP
5241 commands during the callout. The overall timeout applies when there is more
5242 than one host that can be tried. The timeout is checked before trying the
5243 next host. This prevents very long delays if there are a large number of
5244 hosts and all are timing out (e.g. when the network connections are timing
5245 out). The value of the overall timeout can be changed by specifying an
5246 additional sub-option for "callout", called "maxwait". For example:
5248 verify = sender/callout=5s,maxwait=20s
5250 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5251 to the end before writing, but this should make it even safer).
5253 22. Exim was forgetting that it had advertised PIPELINING for the second and
5254 subsequent messages on an SMTP connection. It was also not resetting its
5255 memory on STARTTLS and an internal HELO.
5257 23. When Exim logs an SMTP synchronization error within a session, it now
5258 records whether PIPELINING has been advertised or not.
5260 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5261 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5262 rather than long int.
5264 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5266 26. Added the never_mail option to autoreply.
5272 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5273 crash if the getsockname() call failed; this can happen if a connection is
5274 closed very soon after it is established. The problem was simply in the
5275 order in which certain operations were done, causing Exim to try to write
5276 to the SMTP stream before it had set up the file descriptor. The bug has
5277 been fixed by making things happen in the correct order.
5283 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5284 before the connection was closed, thus losing the rejection response.
5286 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5287 some early Solaris releases, but causes trouble in current releases where
5288 socklen_t is defined.
5290 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5293 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5296 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5297 that releasing the top of it at the end releases what was used for sub-
5298 expansions (unless the block got too big). However, discard this block if
5299 the first thing is a variable or header, so that we can use its block when
5300 it is dynamic (useful for very large $message_headers, for example).
5302 6. Lookups now cache *every* query, not just the most recent. A new, separate
5303 store pool is used for this. It can be recovered when all lookup caches are
5304 flushed. Lookups now release memory at the end of their result strings.
5305 This has involved some general refactoring of the lookup sources.
5307 7. Some code has been added to the store_xxx() functions to reduce the amount
5308 of flapping under certain conditions.
5310 8. log_incoming_interface used to affect only the <= reception log lines. Now
5311 it causes the local interface and port to be added to several more SMTP log
5312 lines, for example "SMTP connection from", and rejection lines.
5314 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5316 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5318 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5319 could be overwritten at the end of the current message (or the start of a
5320 new message if it was set in a HELO ACL). The value is now preserved for
5321 the duration of the SMTP connection.
5323 12. If a transport had a headers_rewrite setting, and a matching header line
5324 contained an unqualified address, that address was qualified, even if it
5325 did not match any rewriting rules. The underlying bug was that the values
5326 of the flags that permit the existence of unqualified sender and recipient
5327 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5328 non-local messages, and by -bnq for local messages) were not being
5329 preserved with the message after it was received.
5331 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5332 "next input=" as part of the text comprising the host identity instead of
5333 the correct text. The code was using the same buffer for two different
5334 strings. However, depending on which order the printing function evaluated
5335 its arguments, the bug did not always show up. Under Linux, for example, my
5336 test suite worked just fine.
5338 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5339 doesn't work with some older versions of Perl. It has been changed to "my",
5340 which in any case is probably the better facility to use.
5342 15. A really picky compiler found some instances of statements for creating
5343 error messages that either had too many or two few arguments for the format
5346 16. The size of the buffer for calls to the DNS resolver has been increased
5347 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5348 for addresses that have a lot of PTR records. This alleviates a problem; it
5349 does not fully solve it.
5351 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5352 buffer now truncates the list and logs the incident, which is the same
5353 action as happens when Exim is looking up a host name and its aliases.
5354 Previously in this situation something unpredictable would happen;
5355 sometimes it was "internal error: store_reset failed".
5357 18. If a server dropped the connection unexpectedly when an Exim client was
5358 using GnuTLS and trying to read a response, the client delivery process
5359 crashed while trying to generate an error log message.
5361 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5362 string, for example:
5364 warn message = H1: something\nH2: something
5366 the text was added as a single header line from Exim's point of view
5367 though it ended up OK in the delivered message. However, searching for the
5368 second and subsequent header lines using $h_h2: did not work. This has been
5369 fixed. Similarly, if a system filter added multiple headers in this way,
5370 the routers could not see them.
5372 20. Expanded the error message when iplsearch is called with an invalid key to
5373 suggest using net-iplsearch in a host list.
5375 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5376 ACLs are no longer actually imposed (and a message to that effect is
5379 22. If a "gecos" field in a passwd entry contained escaped characters, in
5380 particular, if it contained a \" sequence, Exim got it wrong when building
5381 a From: or a Sender: header from that name. A second bug also caused
5382 incorrect handling when an unquoted " was present following a character
5383 that needed quoting.
5385 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5386 was not being matched caselessly.
5388 24. Arranged for all hyphens in the exim.8 source to be escaped with
5391 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5392 properly. Recipient callout cache records were still being keyed to include
5393 the sender, even when use_sender was set false. This led to far more
5394 callouts that were necessary. The sender is no longer included in the key
5395 when use_sender is false.
5397 26. Added "control = submission" modifier to ACLs.
5399 27. Added the ${base62d: operator to decode base 62 numbers.
5401 28. dnsdb lookups can now access SRV records.
5403 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5404 the configuration file.
5406 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5407 (-v) mode. This makes the output for a verbose queue run more intelligible.
5409 31. Added a use_postmaster feature to recipient callouts.
5411 32. Added the $body_zerocount variable, containing the number of binary zero
5412 bytes in the message body.
5414 33. The time of last modification of the "new" subdirectory is now used as the
5415 "mailbox time last read" when there is a quota error for a maildir
5418 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5420 35. Added +ignore_unknown as a special item in host lists.
5422 36. Code for decoding IPv6 addresses in host lists is now included, even if
5423 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5424 address was recognized as an IP address, but was then not correctly decoded
5425 into binary, causing unexpected and incorrect effects when compared with
5432 1. Very minor rewording of debugging text in manualroute to say "list of
5433 hosts" instead of "hostlist".
5435 2. If verify=header_syntax was set, and a header line with an unqualified
5436 address (no domain) and a large number of spaces between the end of the
5437 name and the colon was received, the reception process suffered a buffer
5438 overflow, and (when I tested it) crashed. This was caused by some obsolete
5439 code that should have been removed. The fix is to remove it!
5441 3. When running in the test harness, delay a bit after writing a bounce
5442 message to get a bit more predictability in the log output.
5444 4. Added a call to search_tidyup() just before forking a reception process. In
5445 theory, someone could use a lookup in the expansion of smtp_accept_max_
5446 per_host which, without the tidyup, could leave open a database connection.
5448 5. Added the variables $recipient_data and $sender_data which get set from a
5449 lookup success in an ACL "recipients" or "senders" condition, or a router
5450 "senders" option, similar to $domain_data and $local_part_data.
5452 6. Moved the writing of debug_print from before to after the "senders" test
5455 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5456 problems for message scanning, either using a data ACL, or using
5457 local_scan() because the Received: header was not generated till after they
5458 were called (in order to set the time as the time of reception completion).
5459 I have revised the way this works. The header is now generated after the
5460 body is received, but before the ACL or local_scan() are called. After they
5461 are run, the timestamp in the header is updated.
5467 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5468 before starting a queue runner without re-exec. This happened only when
5469 deliver_drop_privilege was set or when the Exim user was set to root. The
5470 effect of the bug was that timeouts during subsequent deliveries caused
5471 crashes instead of being properly handled. The handler is now left at its
5472 default (and expected) setting.
5474 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5475 message, again when deliver_drop_privilege is set or Exim is run as root.
5476 The bug described in (1) was not present in this case, but the tidying up
5477 of the other signals was missing. I have made the two cases consistent.
5479 3. The ignore_target_hosts setting on a manualroute router was being ignored
5480 for hosts that were looked up using the /MX notation.
5482 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5485 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5486 operated on the sender address. After changing the $sender_address to <>
5487 for the sender address verify, Exim was re-instated it as the original
5488 (before rewriting) address, but remembering that it had rewritten it, so it
5489 wasn't rewriting it again. This bug also had the effect of breaking the
5490 sender address verification caching when the sender address was rewritten.
5492 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5493 This has been changed so that if the ip literal address matches
5494 ignore_target_hosts, the router declines.
5496 7. Added expansion conditions match_domain, match_address, and match_local_
5497 part (NOT match_host).
5499 8. The placeholder for the Received: header didn't have a length field set.
5501 9. Added code to Exim itself and to exim_lock to test for a specific race
5502 condition that could lead to file corruption when using MBX delivery. The
5503 issue is with the lockfile that is created in /tmp. If this file is removed
5504 after a process has opened it but before that process has acquired a lock,
5505 there is the potential for a second process to recreate the file and also
5506 acquire a lock. This could lead to two Exim processes writing to the file
5507 at the same time. The added code performs the same test as UW imapd; it
5508 checks after acquiring the lock that its file descriptor still refers to
5509 the same named file.
5511 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5512 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5515 11. Added the smtp_active_hostname option. If used, this will typically be made
5516 to depend on the incoming interface address. Because $interface_address is
5517 not set up until the daemon has forked a reception process, error responses
5518 that can happen earlier (such as "too many connections") no longer contain
5521 12. If an expansion in a condition on a "warn" statement fails because a lookup
5522 defers, the "warn" statement is abandoned, and the next ACL statement is
5523 processed. Previously this caused the whole ACL to be aborted.
5525 13. Added the iplsearch lookup type.
5527 14. Added ident_timeout as a log selector.
5529 15. Added tls_certificate_verified as a log selector.
5531 16. Added a global option tls_require_ciphers (compare the smtp transport
5532 option of the same name). This controls incoming TLS connections.
5534 17. I finally figured out how to make tls_require_ciphers do a similar thing
5535 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5536 before starting the TLS session.
5538 18. Tabs are now shown as \t in -bP output.
5540 19. If the log selector return_path_on_delivery was set, Exim crashed when
5541 bouncing a message because it had too many Received: header lines.
5543 20. If two routers both had headers_remove settings, and the first one included
5544 a superfluous trailing colon, the final name in the first list and the
5545 first name in the second list were incorrectly joined into one item (with a
5546 colon in the middle).
5552 1. Added -C and -D options to the exinext utility, mainly to make it easier
5553 to include in the automated testing, but these could be helpful when
5554 multiple configurations are in use.
5556 2. The exinext utility was not formatting the output nicely when there was
5557 an alternate port involved in the retry record key, nor when there was a
5558 message id as well (for retries that were specific to a specific message
5559 and a specific host). It was also confused by IPv6 addresses, because of
5560 the additional colons they contain. I have fixed the IPv4 problem, and
5561 patched it up to do a reasonable job for IPv6.
5563 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5564 delivery, the log line now contains "pipelined" if PIPELINING was used.
5566 4. An SMTP transport process used to panic and die if the bind() call to set
5567 an explicit outgoing interface failed. This has been changed; it is now
5568 treated in the same way as a connect() failure.
5570 5. A reference to $sender_host_name in the part of a conditional expansion
5571 that was being skipped was still causing a DNS lookup. This no longer
5574 6. The def: expansion condition was not recognizing references to header lines
5575 that used bh_ and bheader_.
5577 7. Added the _cache feature to named lists.
5579 8. The code for checking quota_filecount in the appendfile transport was
5580 allowing one more file than it should have been.
5582 9. For compatibility with Sendmail, the command line option
5590 and sets the incoming protocol and host name (for trusted callers). The
5591 host name and its colon can be omitted when only the protocol is to be set.
5592 Note the Exim already has two private options, -pd and -ps, that refer to
5593 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5594 "s", but I don't think that's a major issue.
5596 10. A number of refactoring changes to the code, none of which should affect
5599 (a) The number of logging options was getting close to filling up the
5600 32-bit word that was used as a bit map. I have split them into two classes:
5601 those that are passed in the argument to log_write(), and those that are
5602 only ever tested independently outside of that function. These are now in
5603 separate 32-bit words, so there is plenty of room for expansion again.
5604 There is no change in the user interface or the logging behaviour.
5606 (b) When building, for example, log lines, the code previously used a
5607 macro that called string_cat() twice, in order to add two strings. This is
5608 not really sufficiently general. Furthermore, there was one instance where
5609 it was actually wrong because one of the argument was used twice, and in
5610 one call a function was used. (As it happened, calling the function twice
5611 did not affect the overall behaviour.) The macro has been replaced by a
5612 function that can join an arbitrary number of extra strings onto a growing
5615 (c) The code for expansion conditions now uses a table and a binary chop
5616 instead of a serial search (which was left over from when there were very
5617 few conditions). Also, it now recognizes conditions like "pam" even when
5618 the relevant support is not compiled in: a suitably worded error message is
5619 given if an attempt is made to use such a condition.
5621 11. Added ${time_interval:xxxxx}.
5623 12. A bug was causing one of the ddress fields not to be passed back correctly
5624 from remote delivery subprocesses. The field in question was not being
5625 subsequently used, so this caused to problems in practice.
5627 13. Added new log selectors queue_time and deliver_time.
5629 14. Might have fixed a bug in maildirsizefile handling that threw up
5630 "unexpected character" debug warnings, and recalculated the data
5631 unnecessarily. In any case, I expanded the warning message to give more
5634 15. Added the message "Restricted characters in address" to the statements in
5635 the default ACL that block characters like @ and % in local parts.
5637 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5638 Three changes have been made:
5640 (a) There was a serious bug; a negative response to MAIL caused the whole
5641 recipient domain to be cached as invalid, thereby blocking all messages
5642 to all local parts at the same domain, from all senders. This bug has
5643 been fixed. The domain is no longer cached after a negative response to
5644 MAIL if the sender used is not empty.
5646 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5649 (c) A new callout option, "use_sender" has been added for people who want
5650 the modified behaviour.
5656 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5659 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5660 indeed breaks things for older releases.
5662 3. Added additional logging to the case where there is a problem reading data
5663 from a filter that is running in a subprocess using a pipe, in order to
5664 try to track down a specific problem.
5666 4. Testing facility fudge: when running in the test harness and attempting
5667 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5668 getting "No route to host". Convert this to a timeout.
5670 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5673 6. Some OS don't have socklen_t but use size_t instead. This affects the
5674 fifth argument of getsockopt() amongst other things. This is now
5675 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5676 can be set for individual OS. I have set it for SunOS5, OSF1, and
5677 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5678 some earlier ones do not.
5680 7. Change 4.30/15 was not doing the test caselessly.
5682 8. The standard form for an IPv6 address literal was being rejected by address
5683 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5684 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5685 this, as well as the form without the "IPv6" on the front (but only when
5686 address literals are enabled, of course).
5688 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5690 10. Exim crashed if a message with an empty sender address specified by -f
5691 encountered a router with an errors_to setting. This could be provoked only
5692 by a command such as
5696 where an empty string was supplied; "<>" did not hit this bug.
5698 11. Installed PCRE release 4.5.
5700 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5701 remained set. It is now erased.
5703 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5704 times from message ids (which are base 36 rather than the normal 62).
5706 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5707 were being counted as actual protocol errors, and logged if the log
5708 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5709 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5710 and DATA following a set of rejected RCPTs do not count as protocol errors.
5711 In other words, Exim assumes they were pipelined, though this may not
5712 actually be the case. Of course, in all cases the client gets an
5713 appropriate error code.
5715 15. If a lookup fails in an ACL condition, a message about the failure may
5716 be available; it is used if testing the ACL cannot continue, because most
5717 such messages specify what the cause of the deferral is. However, some
5718 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5719 that caused an old message to be retained and used if a later statement
5720 caused a defer, replacing the real cause of the deferral.
5722 16. If an IP address had so many PTR records that the DNS lookup buffer
5723 was not large enough to hold them, Exim could crash while trying to process
5724 the truncated data. It now detects and logs this case.
5726 17. Further to 4.21/58, another change has been made: if (and only if) the
5727 first line of a message (the first header line) ends with CRLF, a bare LF
5728 in a subsequent header line has a space inserted after it, so as not to
5729 terminate the header.
5731 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5732 unnecessarily, used atoi() instead of strtol(), and didn't check the
5733 termination when getting file sizes from file names by regex.
5735 19. Completely re-implemented the support for maildirsize files, in the light
5736 of a number of problems with the previous contributed implementation
5737 (4.30/29). In particular:
5739 . If the quota is zero, the maildirsize file is maintained, but no quota is
5742 . If the maildir directory does not exist, it is created before any attempt
5743 to write a maildirsize file.
5745 . The quota value in the file is just a cache; if the quota is changed in
5746 the transport, the new value overrides.
5748 . A regular expression is available for excluding directories from the
5751 20. The autoreply transport checks the characters in options that define the
5752 message's headers; it allows continued headers, but it was checking with
5753 isspace() after an embedded newline instead of explicitly looking for a
5756 21. If all the "regular" hosts to which an address was routed had passed their
5757 expiry times, and had not reached their retry times, the address was
5758 bounced, even if fallback hosts were defined. Now Exim should go on to try
5761 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5762 equivalent code in the SMTP transport. Some hosts send humungous responses
5763 to HELO/EHLO, more than 1024 it seems.
5765 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5766 gives compiler warnings in some environments. I've now done it "properly",
5769 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5770 (because of problems with the built-in one) was declared to return uschar *
5771 instead of char *, causing compiler failure.
5773 25. Fixed a file descriptor leak when processing alias/forward files.
5775 26. Fixed a minor format string issue in dbfn.c.
5777 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5779 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5780 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5783 29. When a sender address is verified, it is cached, to save repeating the test
5784 when there is more than one recipient in a message. However, when the
5785 verification involves a callout, it is possible for different callout
5786 options to be set for different recipients. It is too complicated to keep
5787 track of this in the cache, so now Exim always runs a verification when a
5788 callout is required, relying on the callout cache for the optimization.
5789 The overhead is duplication of the address routing, but this should not be
5792 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5793 to be verified with callout=postmaster, and the main callout worked but the
5794 postmaster check failed, the verification correctly failed. However, if a
5795 subsequent RCPT command asked for sender verification *without* the
5796 postmaster check, incorrect caching caused this verification also to fail,
5799 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5800 it was not caching the DNS options (qualify_single, search_parents) that
5801 were used when the lookup failed. A subsequent lookup with different
5802 options therefore always gave the same answer, though there were cases
5803 where it should not have. (Example: a "domains = !$mx_any" option on a
5804 dnslookup router: the "domains" option is always processed without any
5805 widening, but the router might have qualify_single set.) Now Exim uses the
5806 cached value only when the same options are set.
5808 32. Added John Jetmore's "exipick" utility to the distribution.
5810 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5811 than a timeout (e.g. a certificate is required, and is not provided), an
5812 Exim server now closes the connection immediately. Previously it waited for
5813 the client to close - but if the client is SSL, it seems that they each
5814 wait for each other, leading to a delay before one of them times out.
5816 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5817 maintained 0.8.x compatibility because I don't think many are using it, and
5818 it is clearly obsolete.
5820 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5823 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5824 client certificate was expired. A simple patch fixes this, though I don't
5825 understand the full logic of why the verify callback is called multiple
5828 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5829 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5830 which causes problems with some clients (such as the Certicom SSL Plus
5831 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5832 disables the coutermeasure allowing Eudora to connect."
5834 38. Exim was not checking that a write() to a log file succeeded. This could
5835 lead to Bad Things if a log got too big, in particular if it hit a file
5836 size limit. Exim now panics and dies if it cannot write to a log file, just
5837 as it does if it cannot open a log file.
5839 39. Modified OS/Makefile-Linux so that it now contains
5841 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5843 The two -D definitions ensure that Exim is compiled with large file
5844 support, which makes it possible to handle log files that are bigger than
5847 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5848 instance) a domain was checked against a named list that involved a lookup,
5849 causing $domain_data to be set, then another domain was checked against the
5850 same list, then the first domain was re-checked, the value of $domain_data
5851 after the final check could be wrong. In particular, if the second check
5852 failed, it could be set empty. This bug probably also applied to
5855 41. The strip_trailing_dot option was not being applied to the address given
5856 with the -f command-line option.
5858 42. The code for reading a message's header from the spool was incrementing
5859 $received_count, but never initializing it. This meant that the value was
5860 incorrect (doubled) while delivering a message in the same process in which
5861 it was received. In the most common configuration of Exim, this never
5862 happens - a fresh exec is done - but it can happen when
5863 deliver_drop_privilege is set.
5865 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5866 it now includes up to 150 characters of the unexpected data in the log
5869 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5870 and building data strings. The size of both of these buffers was 10 000
5871 bytes - far larger than anybody would *ever* want, thought I. Needless to
5872 say, somebody hit the limit. I have increased the maximum line length to
5873 20 000 and the maximum data length of concatenated lines to 100 000. I have
5874 also fixed two bugs, because there was no checking on these buffers. Tsk,
5875 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5876 buffer is too small.
5878 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5879 lsearch lookups. Now it does.
5881 46. When parsing a route_list item in a manualroute router, a fixed-length
5882 buffer was used for the list of hosts. I made this 1024 bytes long,
5883 thinking that nobody would ever have a list of hosts that long. Wrong.
5884 Somebody had a whole pile of complicated expansion conditions, and the
5885 string was silently truncated, leading to an expansion error. It turns out
5886 that it is easier to change to an unlimited length (owing to other changes
5887 that have happened since this code was originally written) than to build
5888 structure for giving a limitation error. The length of the item that
5889 expands into the list of hosts is now unlimited.
5891 47. The lsearch lookup could not handle data where the length of text line was
5892 more than 4095 characters. Such lines were truncated, leading to shortened
5893 data being returned. It should now handle lines of any length.
5895 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5896 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5899 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5900 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5901 no longer happens in comment lines. A list of replacements is now placed
5902 at the head of all of the source files, except those whose only change is
5903 to replace PERL_COMMAND in the very first #! line.
5905 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5906 messages on the queue, with a bottom-up merge sort, using code contributed
5907 by Michael Haardt. This should make operations like -bp somewhat faster on
5908 large queues. It won't affect queue runners, except when queue_run_in_order
5911 51. Installed eximstats 1.31 in the distribution.
5913 52. Added support for SRV lookups to the dnslookup router.
5915 53. If an ACL referred to $message_body or $message_body_end, the value was not
5916 reset for any messages that followed in the same SMTP session.
5918 54. The store-handling optimization for building very long strings was not
5919 differentiating between the different store pools. I don't think this
5920 actually made any difference in practice, but I've tidied it.
5922 55. While running the routers to verify a sender address, $sender_address
5923 was still set to the sender address. This is wrong, because when routing to
5924 send a bounce to the sender, it would be empty. Therefore, I have changed
5925 it so that, while verifying a sender address, $sender_address is set to <>.
5926 (There is no change to what happens when verifying a recipient address.)
5928 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5929 target A or AAAA records (if not already returned) without resetting the
5930 qualify_single or search_parents options of the DNS resolver. These are
5931 inappropriate in this case because the targets of MX and SRV records must
5932 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5933 target that, when qualified, matched something in the local domain. These
5934 two options are now turned off when doing these lookups.
5936 57. It seems that at least some releases of Reiserfs (which does not have the
5937 concept of a fixed number of inodes) returns zero and not -1 for the
5938 number of available inodes. This interacted badly with check_spool_inodes,
5939 which assumed that -1 was the "no such thing" setting. What I have done is
5940 to check that the total number of inodes is greater than zero before doing
5941 the test of how many are available.
5943 58. When a "warn" ACL statement has a log_message modifier, the message is
5944 remembered, and not repeated. This is to avoid a lot of repetition when a
5945 message has many recipients that cause the same warning to be written.
5946 However, Exim was preserving the list of already written lines for an
5947 entire SMTP session, which doesn't seem right. The memory is now reset if a
5948 new message is started.
5950 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5951 debugging output unless log_rewrite was also set.
5953 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5954 of (int)(handle) when we know that handle contains (void *)(-1).
5956 61. The Exim daemon panic-logs an error return when it closes the incoming
5957 connection. However "connection reset by peer" seems to be common, and
5958 isn't really an error worthy of noting specially, so that particular error
5961 62. When Exim is trying to find all the local interfaces, it used to panic and
5962 die if the ioctl to get the interface flags failed. However, it seems that
5963 on at least one OS (Solaris 9) it is possible to have an interface that is
5964 included in the list of interfaces, but for which you get a failure error
5965 for this call. This happens when the interface is not "plumbed" into a
5966 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5967 failure of the "get flags" call assumes that the interface is down.
5969 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5970 makes life easier for people who are doing arithmetic on fields extracted
5971 from dates, where you often get leading zeros that should not be
5972 interpreted as octal.
5974 64. Added qualify_domain to the redirect router, to override the global
5977 65. If a pathologically long header line contained very many addresses (the
5978 report of this problem mentioned 10 000) and each of them was rewritten,
5979 Exim could use up a very large amount of memory. (It kept on making new
5980 copies of the header line as it rewrote, and never released the old ones.)
5981 At the expense of a bit more processing, the header rewriting function has
5982 been changed so that it no longer eats memory in this way.
5984 66. The generation of the Received: header has been moved from the time that a
5985 message starts to be received, to the time that it finishes. The timestamp
5986 in the Received: header should now be very close to that of the <= log
5987 line. There are two side-effects of this change:
5989 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5990 logged header lines no longer include the local Received: line, because
5991 it has not yet been created. The same applies to a copy of the message
5992 that is returned to a non-SMTP sender when a message is rejected.
5994 (b) When a filter file is tested using -bf, no additional Received: header
5995 is added to the test message. After some thought, I decided that this
5998 This change does not affect the value of $received_for. It is still set
5999 after address rewriting, but before local_scan() is called.
6001 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6003 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6004 gave an unhelpful panic error message, and a defer error. I have managed to
6005 change this behaviour so that it now rejects any supplied certificate,
6006 which seems right, as the list of acceptable certificates is empty.
6008 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6009 gave an unhelpful defer error. I have not managed to make this reject any
6010 supplied certificates, but the error message it gives is "no certificate
6011 supplied", which is not helpful.
6013 70. exigrep's output now also includes lines that are not associated with any
6014 message, but which match the given pattern. Implemented by a patch from
6015 Martin Sluka, which also tidied up the Perl a bit.
6017 71. Recipient callout verification, like sender verification, was using <> in
6018 the MAIL FROM command. This isn't really the right thing, since the actual
6019 sender may affect whether the remote host accepts the recipient or not. I
6020 have changed it to use the actual sender in the callout; this means that
6021 the cache record is now keyed on a recipient/sender pair, not just the
6022 recipient address. There doesn't seem to be a real danger of callout loops,
6023 since a callout by the remote host to check the sender would use <>.
6024 [SEE ABOVE: changed after hitting problems.]
6026 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6027 temporary errors. However, in the case of such a code being given after
6028 the end of a data transmission (i.e. after ".") Exim was failing to write
6029 a retry record for the message. (Yes, there was some broken host that was
6030 actually sending 8xx at this point.)
6032 73. An unknown lookup type in a host list could cause Exim to panic-die when
6033 the list was checked. (An example that provoked this was putting <; in the
6034 middle of a list instead of at the start.) If this happened during a DATA
6035 ACL check, a -D file could be left lying around. This kind of configuration
6036 error no longer causes Exim to die; instead it causes a defer error. The
6037 incident is still logged to the main and panic logs.
6039 74. Buglet left over from Exim 3 conversion. The message "too many messages
6040 in one connection" was written to the rejectlog but not the mainlog, except
6041 when address rewriting (yes!) was being logged.
6043 75. Added write_rejectlog option.
6045 76. When a system filter was run not as root (that is, when system_filter_user
6046 was set), the values of the $n variables were not being returned to the
6047 main process; thus, they were not subsequently available in the $sn
6050 77. Added +return_path_on_delivery log selector.
6052 78. A connection timeout was being treated differently from recipients deferred
6053 when testing hosts_max_try with a message that was older than the host's
6054 retry timeout. (The host should not be counted, thus allowing all hosts to
6055 be tried at least once before bouncing.) This may have been the cause of an
6056 occasionally reported bug whereby a message would remain on the queue
6057 longer than the retry timeout, but would be bounced if a delivery was
6058 forced. I say "may" because I never totally pinned down the problem;
6059 setting up timeout/retry tests is difficult. See also the next item.
6061 79. The ultimate address timeout was not being applied to errors that involved
6062 a combination of host plus message (for example, a timeout on a MAIL
6063 command). When an address resolved to a number of possible hosts, and they
6064 were not all tried for each delivery (e.g. because of hosts_max_try), a
6065 message could remain on the queue longer than the retry timeout.
6067 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6070 81. Fixed an obscure SMTP outgoing bug which required at least the following
6071 conditions: (a) there was another message waiting for the same server;
6072 (b) the server returned 5xx to all RCPT commands in the first message so
6073 that the message was not completed; (c) the server dropped the connection
6074 or gave a negative response to the RSET that Exim sends to abort the
6075 transaction. The observed case was a dropped connection after DATA that had
6076 been sent in pipelining mode. That is, the server had advertised PIPELINING
6077 but was not implementing it correctly. The effect of the bug was incorrect
6078 behaviour, such as trying another host, and this could lead to a crash.
6084 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6085 and daemon.c were passed as pointers to ints; they should have been
6086 pointers to socklen_t variables (which are typically unsigned ints).
6088 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6091 3. Fixed a really odd bug that affected only the testing scheme; patching a
6092 certain fixed string in the binary changed the value of another string that
6093 happened to be identical to the end of the original first string.
6095 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6096 name", it returns that address as the IP address. On some operating
6097 systems (e.g. Solaris), it also passes back the IP address string as the
6098 "host name". However, on others (e.g. Linux), it passes back an empty
6099 string. Exim wasn't checking for this, and was changing the host name to an
6100 empty string, assuming it had been canonicalized.
6102 5. Although rare, it is permitted to have more than one PTR record for a given
6103 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6104 all the names associated with an address, because they do in Solaris.
6105 However, it seems that they do not in Linux for data that comes from the
6106 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6107 I found this out when I moved to a new Linux workstation and tried to run
6108 the Exim test suite.
6110 To get round this problem I have changed the code so that it now does its
6111 own call to the DNS to look up PTR records when searching for a host name.
6112 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6113 addresses that are only in /etc/hosts are still found.
6115 This behaviour is, however, controlled by an option called host_lookup_
6116 order, which defaults to "bydns:byaddr". If people want to use the other
6117 order, or indeed, just use one or the other means of lookup, they can
6118 specify it in this variable.
6120 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6121 some operating systems, this comes back from gethostbyaddr() as an empty
6122 string, and this is what Exim used to test for. However, it seems that in
6123 other systems, "." is yielded. Exim now tests for this case too.
6125 7. The values of check_spool_space and check_log_space are now held internally
6126 as a number of kilobytes instead of an absolute number of bytes. If a
6127 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6128 kilobyte. This means that much larger values can be stored.
6130 8. Exim monitor: an attempt to get the action menu when not actually pointing
6131 at a message produces an empty menu entitled "No message selected". This
6132 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6133 no entries in it ("Shell widget menu has zero width and/or height"). So I
6134 have added a single, blank menu entry in this case.
6136 9. Added ${quote_local_part.
6138 10. MIME decoding is now applied to the contents of Subject: header lines when
6141 11. Now that a reference to $sender_host_address automatically causes a reverse
6142 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6143 host lookup before query-style lookups in lists that might use this
6144 variable. This has therefore been abolished, and the "net-" prefix is no
6145 longer necessary for query-style lookups.
6147 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6148 have been a typo for LFLAGS, so it has been changed.
6150 13. The install script calls Exim with "-C /dev/null" in order to find the
6151 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6152 to be output. However, since Exim outputs its version number before the
6153 error, it didn't break the script. It just looked ugly. I fixed this by
6154 always allowing "-C /dev/null" if the caller is root.
6156 14. Ignore overlarge ACL variable number when reading spool file - insurance
6157 against a later release with more variables having written the file.
6159 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6160 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6161 this, as well as the form without the "IPv6" on the front.
6163 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6164 OS/Makefile-Darwin file.
6166 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6167 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6169 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6170 variable. However, some OS use ints for inodes. I've added cast to long int
6171 to get rid of the compiler warning.
6173 19. I had forgotten to lock out "/../" in configuration file names when
6174 ALT_CONFIG_PREFIX was set.
6176 20. Routers used for verification do not need to specify transports. However,
6177 if such a router generated a host list, and callout was configured, Exim
6178 crashed, because it could not find a port number from the (non-existent)
6179 transport. It now assumes port 25 in this circumstance.
6181 21. Added the -t option to exigrep.
6183 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6184 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6185 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6186 them is set, LOOKUP_LSEARCH is forced.
6188 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6190 24. Added sender and host information to the "rejected by local_scan()" log
6191 line; previously there was no indication of these.
6193 25. Added .include_if_exists.
6195 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6196 Linux. It turns out that not all file systems support this. Apparently some
6197 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6198 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6199 not supported on the file descriptor, is now ignored when Exim is trying to
6200 sync a directory. This applies only to Linux.
6202 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6204 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6205 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6208 29. Added support for maildirsize files from supplied patch (modified a bit).
6210 30. The use of :fail: followed by an empty string could lead Exim to respond to
6211 sender verification failures with (e.g.):
6213 550 Verification failed for <xxx>
6214 550 Sender verify failed
6216 where the first response line was missing the '-' that indicates it is not
6217 the final line of the response.
6219 31. The loop for finding the name of the user that called Exim had a hardwired
6220 limit of 10; it now uses the value of finduser_retries, which is used for
6221 all other user lookups.
6223 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6226 33. Exim was neglecting to zero errno before one call of strtol() when
6227 expanding a string and expecting an integer value. On some systems this
6228 resulted in spurious "integer overflow" errors. Also, it was casting the
6229 result into an int without checking.
6231 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6232 did not work. The code looks as if it has *never* worked, though it appears
6233 to have been documented since at least release 1.62. I have made it work.
6235 35. The "timeout_DNS" error in retry rules, also documented since at least
6236 1.62, also never worked. As it isn't clear exactly what this means, and
6237 clearly it isn't a major issue, I have abolished the feature by treating it
6238 as "timeout", and writing a warning to the main and panic logs.
6240 36. The display of retry rules for -brt wasn't always showing the error code
6243 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6244 timeout_connect_A, timeout_connect_MX.
6246 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6247 to the empty sender.
6249 39. The daemon was not analysing the content of -oX till after it had closed
6250 stderr and disconnected from the controlling terminal. This meant that any
6251 syntax errors were only noted on the panic log, and the return code from
6252 the command was 0. By re-arranging the code a little, I've made the
6253 decoding happen first, so such errors now appear on stderr, and the return
6254 code is 1. However, the actual setting up of the sockets still happens in
6255 the disconnected process, so errors there are still only recorded on the
6258 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6259 connections (as happens on some IP stacks) was logged at start up time as
6260 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6261 it from "IPv6 and IPv4", which means that two separate sockets are being
6264 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6265 says whether AF_INET or AF_INET6 was passed as an argument.
6267 42. Exiwhat output was messed up when time zones were included in log
6270 43. Exiwhat now gives more information about the daemon's listening ports,
6271 and whether -tls-on-connect was used.
6273 44. The "port" option of the smtp transport is now expanded.
6275 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6276 silently ignored. Now an error message is written to the main and panic
6279 46. There's a new ACL modifier called "logwrite" which writes to a log file
6280 as soon as it is encountered.
6282 47. Added $local_user_uid and $local_user_gid at routing time.
6284 48. Exim crashed when trying to verify a sender address that was being
6287 49. Exim was recognizing only a space character after ".include". It now also
6288 recognizes a tab character.
6290 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6291 extracting the relevant information from the specification. The man page no
6292 longer contains scrambled data for the -d option, and I've added a section
6293 at the front about calling Exim under different names.
6295 51. Added "extra_headers" argument to the "mail" command in filter files.
6297 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6300 53. Installed eximstats 1.29.
6302 54. Added transport_filter_timeout as a generic transport option.
6304 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6305 Cc: header lines. This was required by RFC 822, but it not required by RFC
6308 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6309 incoming messages that did not have them. Now it does so only if the
6310 message originates locally, that is, if there is no associated remote host
6311 address. When Resent- header lines are present, this applies to the Resent-
6312 lines rather than the non-Resent- lines.
6314 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6315 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6317 58. Messages for configuration errors now include the name of the main
6318 configuration file - useful now that there may be more than one file in a
6319 list (.included file names were always shown).
6321 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6322 for those rare installations that do not start the daemon as root or run it
6323 setuid root. I've cut out the call to initgroups() if the daemon is not
6326 60. The Exim user and group can now be bound into the binary as text strings
6327 that are looked up at the start of Exim's processing.
6329 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6331 62. Added $mailstore_basename variable.
6333 63. Installed patch to sieve.c from Michael Haardt.
6335 64. When Exim failed to open the panic log after failing to open the main log,
6336 the original message it was trying to log was written to stderr and debug
6337 output, but if they were not available (the usual case in production), it
6338 was lost. Now it is written to syslog before the two lines that record the
6339 failures to open the logs.
6341 65. Users' Exim filters run in subprocesses under the user's uid. It is
6342 possible for a "deliver" command or an alias in a "personal" command to
6343 provoke an address rewrite. If logging of address rewriting is configured,
6344 this fails because the process is not running as root or exim. There may be
6345 a better way of dealing with this, but for the moment (because 4.30 needs
6346 to be released), I have disabled address rewrite logging when running a
6347 filter in a non-root, non-exim process.
6353 1. The buildconfig auxiliary program wasn't quoting the value set for
6354 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6355 not defined. This bug was masked in 4.22 by the effect that was fixed in
6358 2. Some messages that were rejected after a message id was allocated were
6359 shown as "incomplete" by exigrep. It no longer does this for messages that
6360 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6362 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6363 have allow_domain_literals set, the ID did not get logged in the <= line.
6364 Domain literals are now always recognized in Message-ID: header lines.
6366 4. The first argument for a ${extract expansion item is the key name or field
6367 number. Leading and trailing spaces in this item were not being ignored,
6368 causing some misleading effects.
6370 5. When deliver_drop_privilege was set, single queue runner processes started
6371 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6372 same command in the process it spins off) were not dropping privilege.
6374 6. When the daemon running as "exim" started a queue runner, it always
6375 re-executed Exim in the spun-off process. This is a waste of effort when
6376 deliver_drop_privilege is set. The new process now just calls the
6377 queue-runner function directly.
6383 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6386 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6387 ignored. Though the use of -oP was forcing the writing of a pid file, it
6388 was always written to the default place.
6390 3. If the message "no IP address found for host xxxx" is generated during
6391 incoming verification, it is now followed by identification of the incoming
6392 connection (so you can more easily find what provoked it).
6394 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6396 5. Added some features to "harden" Exim a bit more against certain attacks:
6398 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6399 be put in Local/Makefile. This is like the never_users runtime option,
6400 but it cannot be overridden. The default setting is "root".
6402 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6403 prefix string with which any file named in a -C command line option
6406 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6407 is retained for -C and -D only if the caller of Exim is root. Without
6408 it, the exim user may also use -C and -D and retain privilege.
6410 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6411 command line option is disabled.
6413 6. Macro names set by the -D option must start with an upper case letter, just
6414 like macro names defined in the configuration file.
6416 7. Added "dereference=" facility to LDAP.
6418 8. Two instances of the typo "uknown" in the source files are fixed.
6420 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6421 the Configure-Makefile script screwed up while processing it.
6423 10. Incorporated PCRE 4.4.
6425 11. The SMTP synchronization check was not operating right at the start of an
6426 SMTP session. For example, it could not catch a HELO sent before the client
6427 waited for the greeting. There is now a check for outstanding input at the
6428 point when the greeting is written. Because of the duplex, asynchronous
6429 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6430 way, but not yet received, when the check is performed.
6432 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6433 on TCP/IP sockets, because this apparently causes some broken clients to
6436 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6437 unchanged) from the Cygwin maintainer.
6439 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6440 was supported instead of testing for mbx. Effectively a typo.
6442 15. The spa authenticator server code was not checking that the input it
6443 received was valid base64.
6445 16. The debug output line for the "set" modifier in ACLs was not showing the
6446 name of the variable that was being set.
6448 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6450 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6451 lookup on its own. Something else was needed to trigger the lookup. For
6452 example, a match in host_lookup or the need for a host name in a host list.
6453 Now, if $sender_host_name is referenced and the host name has not yet been
6454 looked up, a lookup is performed. If the lookup fails, the variable remains
6455 empty, and $host_lookup_failed is set to "1".
6457 19. Added "eqi" as a case-independent comparison operator.
6459 20. The saslauthd authentication condition could segfault if neither service
6460 nor realm was specified.
6462 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6463 error message that was logged was misleading, and incoming SMTP
6464 connections were dropped. The message is now more accurate, and temporary
6465 errors are given to SMTP connections.
6467 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6468 (except RSET) with a 421 error, until QUIT is received. However, it was
6469 failing to send a response to QUIT.
6471 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6472 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6473 ACL is now run after the helo_try_verify_hosts code.
6475 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6476 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6477 case-independent, so other case variants are also recognized). Apparently
6478 some systems use these upper case variants.
6480 25. If more than two messages were waiting for the same host, and a transport
6481 filter was specified for the transport, Exim sent two messages over the
6482 same TCP/IP connection, and then failed with "socket operation on non-
6483 socket" when it tried to send the third.
6485 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6487 27. The extern definition of crypt16() in expand.c was not being excluded when
6488 the OS had its own crypt16() function.
6490 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6491 as a preferred synonym for return_size_limit, both as an option and as an
6494 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6496 30. Changed the default configuration ACL to relax the local part checking rule
6497 for addresses that are not in any local domains. For these addresses,
6498 slashes and pipe symbols are allowed within local parts, but the sequence
6499 /../ is explicitly forbidden.
6501 31. SPA server authentication was not clearing the challenge buffer before
6504 32. log_message in a "warn" ACL statement was writing to the reject log as
6505 well as to the main log, which contradicts the documentation and doesn't
6506 seem right (because no rejection is happening). So I have stopped it.
6508 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6509 However, I am unable to do any testing of this.
6511 34. Fixed an infelicity in the appendfile transport. When checking directories
6512 for a mailbox, to see if any needed to be created, it was accidentally
6513 using path names with one or more superfluous leading slashes; tracing
6514 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6516 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6517 contents are added to the log line that is written for every discarded
6518 recipient. (Previously a log_message setting was ignored.)
6520 36. The ${quote: operator now quotes the string if it is empty.
6522 37. The install script runs exim in order to find its version number. If for
6523 some reason other than non-existence or emptiness, which it checks, it
6524 could not run './exim', it was installing it with an empty version number,
6525 i.e. as "exim-". This error state is now caught, and the installation is
6528 38. An argument was missing from the function that creates an error message
6529 when Exim fails to connect to the socket for saslauthd authentication.
6530 This could cause Exim to crash, or give a corrupted message.
6532 39. Added isip, isip4, and isip6 to ${if conditions.
6534 40. The ACL variables $acl_xx are now saved with the message, and can be
6535 accessed later in routers, transports, and filters.
6537 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6538 strings in the file are not string-expanded.
6540 42. If a MAIL command specified a SIZE value that was too large to fit into an
6541 int variable, the check against message_size_limit failed. Such values are
6542 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6543 day this will have to be increased, but I don't think I want to be around
6544 when emails are that large.
6551 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6552 iconv() is not standard in FreeBSD.
6554 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6555 IPv6 enabled. The observed symptom was a segmentation fault on return from
6556 the function os_common_find_running_interfaces() in src/os.c.
6558 3. In the check_special_case() function in daemon.c I had used "errno" as an
6559 argument name, which causes warnings on some systems. This was basically a
6560 typo, since it was named "eno" in the comments!
6562 4. The code that waits for the clock to tick (at a resolution of some fraction
6563 of a second) so as to ensure message-id uniqueness was always waiting for
6564 at least one whole tick, when it could have waited for less. [This is
6565 almost certainly not relevant at current processor speeds, where it is
6566 unlikely to ever wait at all. But we try to future-proof.]
6568 5. The function that sleeps for a time interval that includes fractions of a
6569 second contained a race. It did not block SIGALRM between setting the
6570 timer, and suspending (a couple of lines later). If the interval was short
6571 and the sigsuspend() was delayed until after it had expired, the suspension
6572 never ended. On busy systems this could lead to processes getting stuck for
6575 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6576 process, before it forks any delivery processes. The open lookup caching
6577 mechanism meant that the open file or database connection was passed into
6578 the delivery process. The problem was that delivery processes always tidy
6579 up cached lookup data. This could cause a problem for the next delivery
6580 process started by the queue runner, because the external queue runner
6581 process does not know about the closure. So the next delivery process
6582 still has data in the lookup cache. In the case of a file lookup, there was
6583 no problem because closing a file descriptor in a subprocess doesn't affect
6584 the parent. However, if the lookup was caching a connection to a database,
6585 the connection was closed, and the second delivery process was likely to
6586 see errors such as "PGSQL: query failed: server closed the connection
6587 unexpectedly". The problem has been fixed by closing all cached lookups
6588 in a queue runner before running a delivery process.
6590 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6591 seem to have the "const" qualifier which it has on other OS. I've
6594 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6595 *of the same type* (client or server) with the same public name that an
6596 error should be diagnosed.
6598 9. When Exim looked up a host name for an IP address, but failed to find the
6599 original IP address when looking up the host name (a safety check), it
6600 output the message "<ip address> does not match any IP for NULL", which was
6601 confusing, to say the least. The bug was that the host name should have
6602 appeared instead of "NULL".
6604 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6605 user that is built into the binary, and the -C or -D options is used, root
6606 privilege is dropped before the configuration file is read. In addition,
6607 logging is switched to stderr instead of the normal log files. If the
6608 configuration then re-defines the Exim user, the unprivileged environment
6609 is probably not what is expected, so Exim logs a panic warning message (but
6612 However, if deliver_drop_privilege is set, the unprivileged state may well
6613 be exactly what is intended, so the warning has been cut out in that case,
6614 and Exim is allowed to try to write to its normal log files.
6620 1. smtp_return_error_details was not giving details for temporary sender
6621 or receiver verification errors.
6623 2. Diagnose a configuration error if two authenticators have the same public
6626 3. Exim used not to create the message log file for a message until the first
6627 delivery attempt. This could be confusing when incoming messages were held
6628 for policy or load reasons. The message log file is now created at the time
6629 the message is received, and an initial "Received" line is written to it.
6631 4. The automatically generated man page for command line options had a minor
6632 bug that caused no ill effects; however, a more serious problem was that
6633 the procedure for building the man page automatically didn't always
6634 operate. Consequently, release 4.20 contains an out-of-date version. This
6635 shouldn't happen again.
6637 5. When building Exim with embedded Perl support, the script that builds the
6638 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6639 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6641 6. The freeze_tell option was not being used for messages that were frozen on
6642 arrival, either by an ACL or by local_scan().
6644 7. Added the smtp_incomplete_transaction log selector.
6646 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6647 was accepting AUTH without a new EHLO.
6649 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6650 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6653 10. Logging of TCP/IP connections (when configured) now happens in the main
6654 daemon process instead of the child process, so that the TCP/IP connection
6655 count is more accurate (but it can never be perfect).
6657 11. The use of "drop" in a nested ACL was not being handled correctly in the
6658 outer ACL. Now, if condition failure induced by the nested "drop" causes
6659 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6660 or "require"), the connection is dropped.
6662 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6663 that yield "discard" can now be used with an "accept" or a "discard" verb,
6664 but an error is generated for any others (because I can't see a useful way
6665 to define what should happen).
6667 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6668 are now processed in the same way as lines in the Exim configuration file.
6669 In particular, continuation lines are supported.
6671 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6673 15. Added -ti meaning -t -i.
6675 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6676 domains, and warn by logging if others are found.
6678 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6679 returned from GIFCONF when Exim is trying to find the list of interfaces on
6680 a host. The code in os.c has been modified to copy each ifreq to an aligned
6681 structure in all cases.
6683 Also, in some cases, the returned ifreq's were being copied to a 'struct
6684 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6685 means the last couple of bytes of an IPv6 address could be chopped if the
6686 ifreq contained only a normal sockaddr (14 bytes storage).
6688 18. Named domain lists were not supported in the hosts_treat_as_local option.
6689 An entry such as +xxxx was not recognized, and was treated as a literal
6692 19. Ensure that header lines added by a DATA ACL are included in the reject log
6693 if the ACL subsequently rejects the message.
6695 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6696 MD5 (which is deprecated).
6698 21. When testing a filter file using -bf, Exim was writing a message when it
6699 took the sender from a "From " line in the message, but it was not doing so
6700 when it took $return_path from a Return-Path: header line. It now does.
6702 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6703 with a valid header line field name (a series of printing characters
6704 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6706 23. Changed "disc" in the source to "disk" to conform to the documentation and
6707 the book and for uniformity.
6709 24. Ignore Sendmail's -Ooption=value command line item.
6711 25. When execve() failed while trying to run a command in a pipe transport,
6712 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6713 could be confused with a return value of 69 from the command itself. This
6714 has been changed to 127, the value the shell returns if it is asked to run
6715 a non-existent command. The wording for the related log line suggests a
6716 non-existent command as the problem.
6718 26. If received_header_text expands to an empty string, do not add a Received:
6719 header line to the message. (Well, it adds a token one on the spool, but
6720 marks it "old" so that it doesn't get used or transmitted.)
6722 27. Installed eximstats 1.28 (addition of -nt option).
6724 28. There was no check for failure on the call to getsockname() in the daemon
6725 code. This can fail if there is a shortage of resources on the system, with
6726 ENOMEM, for example. A temporary error is now given on failure.
6728 29. Contrary to the C standard, it seems that in some environments, the
6729 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6730 program. Exim now does this explicitly; it affects the formatting of
6731 timestamps using strftime().
6733 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6734 complaints. I've now initialized all the variables, to avoid this.
6736 32. Header lines added by a system filter were not being "seen" during
6737 transport-time rewrites.
6739 33. The info_callback() function passed to OpenSSL is set up with type void
6740 (*)(SSL *, int, int), as described somewhere. However, when calling the
6741 function (actually a macro) that sets it up, the type void(*)() is
6742 expected. I've put in a cast to prevent warnings from picky compilers.
6744 34. If a DNS black list lookup found a CNAME record, but there were no A
6745 records associated with the domain it pointed at, Exim crashed.
6747 35. If a DNS black list lookup returned more than one A record, Exim ignored
6748 all but the first. It now scans all returned addresses if a particular IP
6749 value is being sought. In this situation, the contents of the
6750 $dnslist_value variable are a list of all the addresses, separated by a
6753 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6754 to accept a host name and all its aliases if the forward lookup for any of
6755 them yielded the IP address of the incoming connection. Now it accepts only
6756 those names whose forward lookup yields the correct IP address. Any other
6757 names are discarded. This closes a loophole whereby a rogue DNS
6758 administrator could create reverse DNS records to break through a
6759 wildcarded host restriction in an ACL.
6761 37. If a user filter or a system filter that ran in a subprocess used any of
6762 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6763 the wrong values were passed to the pipe command ($thisaddress had the
6764 value of $0, $0 had the value of $1, etc). This bug was introduced by
6765 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6767 38. Improved the line breaking for long SMTP error messages from ACLs.
6768 Previously, if there was no break point between 40 and 75 characters, Exim
6769 left the rest of the message alone. Two changes have been made: (a) I've
6770 reduced the minimum length to 35 characters; (b) if it can't find a break
6771 point between 35 and 75 characters, it looks ahead and uses the first one
6772 that it finds. This may give the occasional overlong line, but at least the
6773 remaining text gets split now.
6775 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6776 file descriptors might be low, and that setting 1000 would always raise it.
6777 It turns out that in some environments, the limit is already over 1000 and
6778 that lowering it causes trouble. So now Exim takes care not to decrease it.
6780 40. When delivering a message, the value of $return_path is set to $sender_
6781 address at the start of routing (routers may change the value). By an
6782 oversight, this default was not being set up when an address was tested by
6783 -bt or -bv, which affected the outcome if any router or filter referred to
6786 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6787 log only when "message" or "log_message" are set. However, if one of the
6788 conditions was an address verification, or a call to a nested ACL, the
6789 messages generated by the underlying test were being passed through. This
6790 no longer happens. The underlying message is available in $acl_verify_
6791 message for both "message" and "log_message" expansions, so it can be
6792 passed through if needed.
6794 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6795 new expansion $bh_ to give the encoded byte string without charset
6796 translation. Translation happens only if iconv() is available; HAVE_ICONV
6797 indicates this at build time. HEADERS_CHARSET gives the charset to
6798 translate to; headers_charset can change it in the configuration, and
6799 "headers charset" can change it in an individual filter file.
6801 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6802 that creates RFC 2047 encoded "words" labels them as that charset instead
6803 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6804 expansion operator; (ii) when Exim creates a From: line for a local
6805 message; (iii) when a header line is rewritten to include a "phrase" part.
6807 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6808 buggy, causing it to skip the first lines of messages whose message ID
6809 ended in 'D'. This would not have bitten before Exim release 4.14, because
6810 message IDs were unlikely to end in 'D' before then. The effect was to have
6811 incorrect size information for certain domains.
6813 45. #include "config.h" was missing at the start of the crypt16.c module. This
6814 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6817 46. If there was a timeout during a "random" callout check, Exim treated it as
6818 a failure of the random address, and carried on sending RSET and the real
6819 address. If the delay was just some slowness somewhere, the response to the
6820 original RCPT would be taken as a response to RSET and so on, causing
6821 mayhem of various kinds.
6823 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6824 when I implemented it. It didn't allow for the fact that some option values
6825 may legitimately be negative (e.g. size_addition), and it didn't even do
6826 the right test for positive values.
6828 48. Domain names in DNS records are case-independent. Exim always looks them up
6829 in lower case. Some resolvers return domain names in exactly the case they
6830 appear in the zone file, that is, they may contain uppercase letters. Not
6831 all resolvers do this - some return always lower case. Exim was treating a
6832 change of case by a resolver as a change of domain, similar to a widening
6833 of a domain abbreviation. This triggered its re-routing code and so it was
6834 trying to route what was effectively the same domain again. This normally
6835 caused routing to fail (because the router wouldn't handle the domain
6836 twice). Now Exim checks for this case specially, and just changes the
6837 casing of the domain that it ultimately uses when it transmits the message
6840 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6843 50. If a filter generated a file delivery with a non-absolute name (possible if
6844 no home directory exists for the router), the forbid_file option was not
6847 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6848 the existing equality matching.
6850 52. Exim was using ints instead of ino_t variables in some places where it was
6851 dealing with inode numbers.
6853 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6854 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6855 called TMPDIR, and if it finds it is different, it changes its value.
6857 54. The smtp_printf() function is now made available to local_scan() so
6858 additional output lines can be written before returning. There is also an
6859 smtp_fflush() function to enable the detection of a dropped connection.
6860 The variables smtp_input and smtp_batched_input are exported to
6863 55. Changed the default runtime configuration: the message "Unknown user"
6864 has been removed from the ACL, and instead placed on the localuser router,
6865 using the cannot_route_message feature. This means that any verification
6866 failures that generate their own messages won't get overridden. Similarly,
6867 the "Unrouteable address" message that was in the ACL for unverifiable
6868 relay addresses has also been removed.
6870 56. Added hosts_avoid_esmtp to the smtp transport.
6872 57. The exicyclog script was not checking for the esoteric option
6873 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6874 will work only if exicyclog is run under the appropriate euid.
6876 58. Following a discussion on the list, the rules by which Exim recognises line
6877 endings on incoming messages have been changed. The -dropcr and drop_cr
6878 options are now no-ops, retained only for backwards compatibility. The
6879 following line terminators are recognized: LF CRLF CR. However, special
6880 processing applies to CR:
6882 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6883 nor a local message in the state where . is a terminator.
6885 (ii) If a bare CR is encountered in a header line, an extra space is added
6886 after the line terminator so as not to end the header. The reasoning
6887 behind this is that bare CRs in header lines are most likely either
6888 to be mistakes, or people trying to play silly games.
6890 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6891 was being incorrectly given as 18 bytes larger than it should have been.
6892 This is a VOB (very old bug).
6894 60. This may never have affected anything current, but just in case it has:
6895 When the local host is found other than at the start of a list of hosts,
6896 the local host, those with the same MX, and any that follow, are discarded.
6897 When the list in question was part of a longer list of hosts, the following
6898 hosts (not currently being processed) were also being discarded. This no
6899 longer happens. I'm not sure if this situation could ever has previously
6902 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6905 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6906 header. This is something that is recommended in a new Internet Draft, and
6907 is something that is documented as being done by Sendmail. There are two
6908 possible values. For messages generated by the autoreply transport, Exim
6911 Auto-Submitted: auto-replied
6913 whereas for all other generated messages (e.g. bounces) it adds
6915 Auto-Submitted: auto-generated
6917 63. The "personal" condition in filters now includes a test for the
6918 Auto-Submitted: header. If it contains the string "auto-" the message it
6919 not considered personal.
6921 64. Added rcpt_include_affixes as a generic transport option.
6923 65. Added queue_only_override (default true).
6925 66. Added the syslog_duplication option.
6927 67. If what should have been the first header line of a message consisted of
6928 a space followed by a colon, Exim was mis-interpreting it as a header line.
6929 It isn't of course - it is syntactically invalid and should therefore be
6930 treated as the start of the message body. The misbehaviour could have
6931 caused a number of strange effects, including loss of data in subsequent
6932 header lines, and spool format errors.
6934 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6935 client host had authenticated. This control can now be exercised by an ACL
6936 for more flexibility.
6938 69. By default, callouts do not happen when testing with -bh. There is now a
6939 variant, -bhc, which does actually run the callout code, including
6940 consulting and updating the callout cache.
6942 70. Added support for saslauthd authentication, courtesy of Alexander
6945 71. If statvfs() failed on the spool or log directories while checking their
6946 size for availability, Exim confusingly gave the error "space shortage".
6947 Furthermore, in debugging mode it crashed with a floating point exception.
6948 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6949 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6950 serious problem, Exim now writes to the main and panic logs when this
6951 happens, with details of the failure. It then refuses to accept the
6952 incoming message, giving the message "spool directory problem" or "log
6953 directory problem" with a 421 code for SMTP messages.
6955 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6956 0, 1, and 2 exist, because some OS complain for execs without them (see
6957 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6958 descriptors. However, the code omitted to check that the open succeeded,
6959 causing mysterious errors if for some reason the permissions on /dev/null
6960 got screwed. Now Exim writes a message to the main and panic logs, and
6961 bombs out if it can't open /dev/null.
6963 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6964 interact so that it is all more flexible. It is supposed to remain
6965 backwards compatible. Also added extra_local_interfaces.
6967 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6968 to bomb out with an assertion failure - to the client this appears as a
6969 connection drop. This problem occurs in the part of the code that was taken
6970 from the Samba project. Fortunately, the assertion is in a very simple
6971 function, so I have fixed this by reproducing the function inline in the
6972 one place where it is called, and arranging for authentication to fail
6973 instead of killing the process with assert().
6975 75. The SPA client code was not working when the server requested OEM rather
6976 than Unicode encoding.
6978 76. Added code to make require_files with a specific uid setting more usable in
6979 the case where statting the file as root fails - usually a non-root-mounted
6980 NFS file system. When this happens and the failure is EACCES, Exim now
6981 forks a subprocess and does the per-uid checking as the relevant uid.
6983 77. Added process_log_path.
6985 78. If log_file_path was not explicitly set, a setting of check_log_space or
6986 check_log_inodes was ignored.
6988 79. If a space check for the spool or log partitions fails, the incident is now
6989 logged. Of course, in the latter case the data may get lost...
6991 80. Added the %p formatting code to string_format() so that it can be used to
6992 print addresses in debug_print(). Adjusted all the address printing in the
6993 debugging in store.c to use %p rather than %d.
6995 81. There was a concern that a line of code in smtp_in.c could overflow a
6996 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6997 initially expressed, the concern was not well-founded, because trailing
6998 spaces are removed early. However, if the trailing spaces were followed by
6999 a NULL, they did not get removed, so the overflow was possible. Two fixes
7002 (a) I re-wrote the offending code in a cleaner fashion.
7003 (b) If an incoming SMTP command contains a NULL character, it is rejected
7006 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7007 runs initgroups(), so that if the Exim user is in any additional groups,
7008 they will be used during message reception.
7014 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob