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.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
68 JH/01 Rework error string handling in TLS interface so that the caller in
69 more cases is responsible for logging. This permits library-sourced
70 string to be attached to addresses during delivery, and collapses
71 pairs of long lines into single ones.
73 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
74 during configuration. Wildcards are allowed and expanded.
76 JH/02 Rework error string handling in DKIM to pass more info back to callers.
77 This permits better logging.
79 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
80 do not close it down and have the child transport start it up again on
81 the passed-on TCP connection. Instead, proxy the child (and any
82 subsequent ones) for TLS via a unix-domain socket channel. Logging is
83 affected: the continued delivery log lines do not have any DNSSEC, TLS
84 Certificate or OCSP information. TLS cipher information is still logged.
86 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
87 identical IP addresses on different listening ports. Will also affect
90 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
91 add noisy ifdef guards to special-case this sillyness.
92 Patch from Bernd Kuhls.
94 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
95 than 255 are no longer allowed.
97 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
98 Disable the session-cache too, which might reduce our load. Since we
99 currrectly use a new context for every connection, both as server and
100 client, there is no benefit for these.
101 GnuTLS appears to not support tickets server-side by default (we don't
102 call gnutls_session_ticket_enable_server()) but client side is enabled
103 by default on recent versions (3.1.3 +) unless the PFS priority string
106 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
107 <https://reproducible-builds.org/specs/source-date-epoch/>.
109 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
110 the check for any unsuccessful recipients did not notice the limit, and
111 erroneously found still-pending ones.
113 JH/08 Pipeline CHUNKING command and data together, on kernels that support
114 MSG_MORE. Only in-clear (not on TLS connections).
116 JH/09 Avoid using a temporary file during transport using dkim. Unless a
117 transport-filter is involved we can buffer the headers in memory for
118 creating the signature, and read the spool data file once for the
119 signature and again for transmission.
121 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
122 4.77 as the kernel support then wasn't solid, having issues in 64bit
123 mode. Now, it's been long enough. Add support for FreeBSD also.
125 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
126 case where the routing stage had gathered several addresses to send to
127 a host before calling the transport for the first, we previously failed
128 to close down TLS in the old transport process before passing the TCP
129 connection to the new process. The new one sent a STARTTLS command
130 which naturally failed, giving a failed delivery and bloating the retry
131 database. Investigation and fix prototype from Wolfgang Breyha.
133 JH/12 Fix check on SMTP command input synchronisation. Previously there were
134 false-negatives in the check that the sender had not preempted a response
135 or prompt from Exim (running as a server), due to that code's lack of
136 awareness of the SMTP input buffering.
138 PP/04 Add commandline_checks_require_admin option.
139 Exim drops privileges sanely, various checks such as -be aren't a
140 security problem, as long as you trust local users with access to their
141 own account. When invoked by services which pass untrusted data to
142 Exim, this might be an issue. Set this option in main configuration
143 AND make fixes to the calling application, such as using `--` to stop
146 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
147 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
148 responses to those, into a single TLS record each way (this usually means
149 a single packet). As a side issue, smtp_enforce_sync now works on TLS
152 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
153 affects you only if you're dancing at the edge of the param size limits.
154 If you are, and this message makes sense to you, then: raise the
155 configured limit or use OpenSSL 1.1. Nothing we can do for older
158 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
159 cmdline element to get the documented default one. Previously it was
162 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
163 are unsuable for non-message-specific reasons - eg. connection timeout,
164 banner-time rejection.
166 JH/16 Fix logging of delivery remote port, when specified by router, under
169 PP/06 Repair manualroute's ability to take options in any order, even if one
170 is the name of a transport.
173 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
175 JH/17 Change the list-building routines interface to use the expanding-string
176 triplet model, for better allocation and copying behaviour.
178 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
179 Previously it was constructed the first time a possibly-matching string
180 was met in the configuration file input during startup; now it is done
183 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
184 compatible one, to avoid the (poorly documented) possibility of a config
185 file in the working directory redirecting the DB files, possibly correpting
186 some existing file. CVE-2017-10140 assigned for BDB.
188 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
189 cache-hot. Previously, although the result was properly cached, the
190 initial verify call returned a defer.
192 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
193 the main verify for receipient in uncached-mode.
195 JH/22 Retire historical build files to an "unsupported" subdir. These are
196 defined as "ones for which we have no current evidence of testing".
198 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
199 if present. Previously it was ignored.
201 JH/24 Start using specified-initialisers in C structure init coding. This is
202 a C99 feature (it's 2017, so now considered safe).
204 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
205 if was a fixed-sized field and bitmask ops via macros; it is now more
208 PP/07 GitHub PR 56: Apply MariaDB build fix.
209 Patch provided by Jaroslav Škarvada.
211 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
212 during Coverity cleanups [4.87 JH/47]
213 Diagnosis and fix provided by Michael Fischer v. Mollard.
215 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
216 the right size to place the terminating semicolon on its own folded
217 line, the header hash was calculated to an incorrect value thanks to
218 the (relaxed) space the fold became.
220 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
221 and confused the parent.
223 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
224 which could crash as a result. This could lead to undeliverable messages.
226 JH/28 Logging: "next input sent too soon" now shows where input was truncated
229 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
230 matters on fast-turnover and PID-randomising systems, which were getting
231 out-of-order delivery.
233 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
234 a possibly-overlapping copy. The symptom was that "Remote host closed
235 connection in response to HELO" was logged instead of the actual 4xx
238 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
239 Previously only that bufferd was discarded, resulting in SYMTP command
242 JH/32 DKIM: when a message has multiple signatures matching an identity given
243 in dkim_verify_signers, run the dkim acl once for each. Previously only
244 one run was done. Bug 2189.
246 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
247 "panic the current process" to "deliberately defer". The panic log is
248 still written with the problem list name; the mail and reject logs now
249 get a temp-reject line for the message that was being handled, saying
250 something like "domains check lookup or other defer". The SMTP 451
251 message is still "Temporary local problem".
253 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
254 A crafted sequence of BDAT commands could result in in-use memory beeing
255 freed. CVE-2017-16943.
257 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
258 from SMTP input. Previously it was always done; now only done for DATA
259 and not BDAT commands. CVE-2017-16944.
261 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
262 to the message (such as an overlong header line). Previously this was
263 not done and we did not exit BDAT mode. Followon from the previous item
264 though a different problem.
270 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
271 than -2003 did; needs libidn2 in addition to libidn.
273 JH/02 The path option on a pipe transport is now expanded before use.
275 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
276 Patch provided by "Björn", documentation fix added too.
278 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
279 missing a wire-to-host endian conversion.
281 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
282 close after a BDAT command line could be taken as a following command,
283 giving a synch failure. Fix by only checking for synch immediately
284 before acknowledging the chunk.
286 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
287 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
288 macro. Patches provided by Josh Soref.
290 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
291 Previously we did not; the RFC seems ambiguous and VRFY is not listed
292 by IANA as a service extension. However, John Klensin suggests that we
295 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
296 the dkim code may be unix-mode line endings rather than smtp wire-format
297 CRLF, so prepend a CR to any bare LF.
299 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
300 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
302 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
303 the first were themselves being wrongly included in the feed into dkim
304 processing; with most chunk sizes in use this resulted in an incorrect
305 body hash calculated value.
307 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
308 DKIM signature block, for verification. Although advised against by
309 standards it is specifically not ruled illegal.
311 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
313 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
314 missing a body hash (the bh= tag).
316 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
317 It seems that HAProxy sends the Proxy Protocol information in clear and
318 only then does a TLS startup, so do the same.
320 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
321 TCP connections (such as for Spamd) unless the daemon successfully set
322 Fast Open mode on its listening sockets. This fixes breakage seen on
323 too-old kernels or those not configured for Fast Open, at the cost of
324 requiring both directions being enabled for TFO, and TFO never being used
325 by non-daemon-related Exim processes.
327 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
328 endings, at least on the first header line. Try to canonify any that get
329 past that check, despite the cost.
331 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
332 now limited to an arbitrary five deep, while parsing addresses with the
333 strip_excess_angle_brackets option enabled.
335 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
336 instead leave the unprompted TLS handshake in socket buffer for the
337 TLS library to consume.
339 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
341 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
343 JH/16 Drop variables when they go out of scope. Memory management drops a whole
344 region in one operation, for speed, and this leaves assigned pointers
345 dangling. Add checks run only under the testsuite which checks all
346 variables at a store-reset and panics on a dangling pointer; add code
347 explicitly nulling out all the variables discovered. Fixes one known
348 bug: a transport crash, where a dangling pointer for $sending_ip_address
349 originally assigned in a verify callout, is re-used.
351 PP/06 Drop '.' from @INC in various Perl scripts.
353 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
355 PP/08 Reduce a number of compilation warnings under clang; building with
356 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
357 should be warning-free.
359 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
361 HS/01 Fix portability problems introduced by PP/08 for platforms where
362 realloc(NULL) is not equivalent to malloc() [SunOS et al].
364 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
365 chunk. This allows us to accept broken chunked messages. We need a more
366 general solution here.
368 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
369 already-broken messages in the queue.
371 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
373 JH/19 Fix reference counting bug in routing-generated-address tracking.
379 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
380 supports it and a size is available (ie. the sending peer gave us one).
382 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
383 years of being deprecated). The replacements are the ACLs
384 acl_smtp_mime and acl_not_smtp_mime.
386 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
387 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
388 or even an in-clear connection were permitted. Now, if the host lookup
389 was dnssec and dane was requested then the host is only used if the
390 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
391 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
392 if one fails this test.
393 This means that a poorly-configured remote DNS will make it incommunicado;
394 but it protects against a DNS-interception attack on it.
396 JH/04 Bug 1810: make continued-use of an open smtp transport connection
397 non-noisy when a race steals the message being considered.
399 JH/05 If main configuration option tls_certificate is unset, generate a
400 self-signed certificate for inbound TLS connections.
402 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
403 in rewrites and routers.
405 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
406 and logged a warning sing 4.83; now they are a configuration file error.
408 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
409 (lacking @domain). Apply the same qualification processing as RCPT.
411 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
413 JH/10 Support ${sha256:} applied to a string (as well as the previous
416 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
417 a cutthrough deliver is pending, as we always want to make a connection.
418 This also avoids re-routing the message when later placing the cutthrough
419 connection after a verify cache hit.
420 Do not update it with the verify result either.
422 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
423 when routing results in more than one destination address.
425 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
426 signing (which inhibits the cutthrough capability). Previously only
427 the presence of an option was tested; now an expansion evaluating as
428 empty is permissible (obviously it should depend only on data available
429 when the cutthrough connection is made).
431 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
432 the relevant preceding SMTP command did not note the pipelining mode.
434 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
435 Previously they were not counted.
437 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
438 as one having no matching records. Previously we deferred the message
439 that needed the lookup.
441 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
442 distinguished as "(=".
444 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
445 for missing MX records. Previously it only worked for missing A records.
447 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
449 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
450 after the data-go-ahead and data-ack. Patch from Jason Betts.
452 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
453 even for a "none" policy. Patch from Tony Meyer.
455 JH/22 Fix continued use of a connection for further deliveries. If a port was
456 specified by a router, it must also match for the delivery to be
459 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
460 When one of the recipients of a message was unsuitable for the connection
461 (has no matching addresses), we lost track of needing to mark it
462 deferred. As a result mail would be lost.
464 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
466 JH/25 Decoding ACL controls is now done using a binary search; the source code
467 takes up less space and should be simpler to maintain. Merge the ACL
468 condition decode tables also, with similar effect.
470 JH/26 Fix problem with one_time used on a redirect router which returned the
471 parent address unchanged. A retry would see the parent address marked as
472 delivered, so not attempt the (identical) child. As a result mail would
475 JH/27 Fix a possible security hole, wherein a process operating with the Exim
476 UID can gain a root shell. Credit to http://www.halfdog.net/ for
477 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
480 JH/28 Enable {spool,log} filesystem space and inode checks as default.
481 Main config options check_{log,spool}_{inodes,space} are now
482 100 inodes, 10MB unless set otherwise in the configuration.
484 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
485 Previously it only applied to the main-section connection policy
488 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
490 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
491 by me. Added RFC7919 DH primes as an alternative.
493 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
494 are not in the system include path.
496 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
497 GnuTLS, when a session startup failed (eg because the client disconnected)
498 Exim did stdio operations after fclose. This was exposed by a recent
499 change which nulled out the file handle after the fclose.
501 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
502 signed directly by the cert-signing cert, rather than an intermediate
503 OCSP-signing cert. This is the model used by LetsEncrypt.
505 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
507 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
508 an incoming connection.
510 HS/02 Bug 1802: Do not half-close the connection after sending a request
513 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
514 fallback to "prime256v1".
516 JH/34 SECURITY: Use proper copy of DATA command in error message.
517 Could leak key material. Remotely exploitable. CVE-2016-9963.
523 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
524 and 3.4.4 - once the server is enabled to respond to an OCSP request
525 it does even when not requested, resulting in a stapling non-aware
526 client dropping the TLS connection.
528 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
529 support variable-length bit vectors. No functional change.
531 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
532 The I= interface field on outgoing lines is now after the H= remote
533 host field, same as incoming lines. There is a separate
534 outgoing_interface log selector which allows you to disable the
537 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
538 If not running log_selector +smtp_connection the mainlog would be held
539 open indefinitely after a "too many connections" event, including to a
540 deleted file after a log rotate. Leave the per net connection logging
541 leaving it open for efficiency as that will be quickly detected by the
542 check on the next write.
544 HS/01 Bug 1671: Fix post transport crash.
545 Processing the wait-<transport> messages could crash the delivery
546 process if the message IDs didn't exist for some reason. When
547 using 'split_spool_directory=yes' the construction of the spool
548 file name failed already, exposing the same netto behaviour.
550 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
551 mime_regex ACL conditions.
553 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
554 to DSN fail messages (bounces): remote IP, remote greeting, remote response
555 to HELO, local diagnostic string.
557 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
558 log line to debug. Even when configured with a tls authenticator many
559 client connections are expected to not authenticate in this way, so
560 an authenticate fail is not an error.
562 HS/02 Add the Exim version string to the process info. This way exiwhat
563 gives some more detail about the running daemon.
565 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
566 matter for fast-change records such as DNSBLs.
568 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
569 retry record, even if constant. There may be multiple transports with
570 different interface settings and the retry behaviour needs to be kept
573 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
575 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
577 JH/10 Bug 840: fix log_defer_output option of pipe transport
579 JH/11 Bug 830: use same host for all RCPTS of a message, even under
580 hosts_randomize. This matters a lot when combined with mua_wrapper.
582 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
583 ${quote_pgsql:<string>} operator.
585 JH/13 Bug 1708: avoid misaligned access in cached lookup.
587 JH/14 Change header file name for freeradius-client. Relevant if compiling
588 with Radius support; from the Gentoo tree and checked under Fedora.
590 JH/15 Bug 1712: Introduce $prdr_requested flag variable
592 JH/16 Bug 1714: Permit an empty string as expansion result for transport
593 option transport_filter, meaning no filtering.
595 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
597 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
598 defaults to "*" (all hosts). The variable is now available when not built
599 with TLS, default unset, mainly to enable keeping the testsuite sane.
600 If a server certificate is not supplied (via tls_certificate) an error is
601 logged, and clients will find TLS connections fail on startup. Presumably
602 they will retry in-clear.
603 Packagers of Exim are strongly encouraged to create a server certificate
604 at installation time.
606 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
607 with the $config_file variable.
609 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
610 in transport context, after the attempt, and per-recipient. The latter type
611 is per host attempted. The event data is the error message, and the errno
612 information encodes the lookup type (A vs. MX) used for the (first) host,
613 and the trailing two digits of the smtp 4xx response.
615 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
616 to write to mainlog (or rejectlog, paniclog) in the window between file
617 creation and permissions/ownership being changed. Particularly affects
618 installations where exicyclog is run as root, rather than exim user;
619 result is that the running daemon panics and dies.
621 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
623 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
624 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
625 "pri" and "weight". Note that the previous implicit priority given by the
626 list order is no longer honoured.
628 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
631 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
632 by defining SUPPORT_SOCKS.
634 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
635 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
636 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
637 variables are renamed to proxy_{local,external}_{address,port}.
639 JH/27 Move Internationalisation support from Experimental to mainline, enabled
640 for a build by defining SUPPORT_I18N
642 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
643 of the query string, and make ${quote_redis:} do that quoting.
645 JH/29 Move Events support from Experimental to mainline, enabled by default
646 and removable for a build by defining DISABLE_EVENT.
648 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
650 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
651 cached by the daemon.
653 JH/32 Move Redis support from Experimental to mainline, enabled for a build
654 by defining LOOKUP_REDIS. The libhiredis library is required.
656 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
657 keys are given for lookup.
659 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
660 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
661 only supported when built with TLS support. The PolarSSL SHA routines
662 are still used when the TLS library is too old for convenient support.
664 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
665 openssl_options), for security. OpenSSL forces this from version 1.1.0
666 server-side so match that on older versions.
668 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
669 allocation for $value could be released as the expansion processing
670 concluded, but leaving the global pointer active for it.
672 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
673 and to use the domains and local_parts ACL conditions.
675 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
676 incorrectly not doubled on cutthrough transmission, hence seen as a
677 body-termination at the receiving system - resulting in truncated mails.
678 Commonly the sender saw a TCP-level error, and retransmitted the message
679 via the normal store-and-forward channel. This could result in duplicates
680 received - but deduplicating mailstores were liable to retain only the
681 initial truncated version.
683 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
685 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
687 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
688 we're in there, support oversigning also; bug 1309.
690 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
692 HS/04 Add support for keep_environment and add_environment options.
694 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
695 either intentional arithmetic overflow during PRNG, or testing config-
698 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
699 delivery resulted in actual delivery. Cancel cutthrough before DATA
702 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
703 rejects a recipient: pass the reject to the originator.
705 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
706 Many were false-positives and ignorable, but it's worth fixing the
709 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
710 for the new environment-manipulation done at startup. Move the routines
711 from being local to tls.c to being global via the os.c file.
713 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
714 an extract embedded as result-arg for a map, the first arg for extract
715 is unavailable so we cannot tell if this is a numbered or keyed
716 extraction. Accept either.
722 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
725 JH/02 The smtp transport option "multi_domain" is now expanded.
727 JH/03 The smtp transport now requests PRDR by default, if the server offers
730 JH/04 Certificate name checking on server certificates, when exim is a client,
731 is now done by default. The transport option tls_verify_cert_hostnames
732 can be used to disable this per-host. The build option
733 EXPERIMENTAL_CERTNAMES is withdrawn.
735 JH/05 The value of the tls_verify_certificates smtp transport and main options
736 default to the word "system" to access the system default CA bundle.
737 For GnuTLS, only version 3.0.20 or later.
739 JH/06 Verification of the server certificate for a TLS connection is now tried
740 (but not required) by default. The verification status is now logged by
741 default, for both outbound TLS and client-certificate supplying inbound
744 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
747 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
748 Status Notification (bounce) messages are now MIME format per RFC 3464.
749 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
750 under the control of the dsn_advertise_hosts option, and routers may
751 have a dsn_lasthop option.
753 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
754 default, modifiable by a malware= option. The list separator for
755 the options can now be changed in the usual way. Bug 68.
757 JH/10 The smtp_receive_timeout main option is now expanded before use.
759 JH/11 The incoming_interface log option now also enables logging of the
760 local interface on delivery outgoing connections.
762 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
763 if the interface and destination host and port all match.
765 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
768 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
769 Patch from Andrew Lewis.
771 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
772 now supports optional time-restrictions, weighting, and priority
773 modifiers per server. Patch originally by <rommer@active.by>.
775 JH/16 The spamd_address main option now supports a mixed list of local
776 and remote servers. Remote servers can be IPv6 addresses, and
777 specify a port-range.
779 JH/17 Bug 68: The spamd_address main option now supports an optional
780 timeout value per server.
782 JH/18 Bug 1581: Router and transport options headers_add/remove can
783 now have the list separator specified.
785 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
788 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
791 JH/21 Support for the A6 type of dns record is withdrawn.
793 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
794 rather than the verbs used.
796 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
797 from 255 to 1024 chars.
799 JH/24 Verification callouts now attempt to use TLS by default.
801 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
802 are generic router options now. The defaults didn't change.
804 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
805 Original patch from Alexander Shikoff, worked over by JH.
807 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
808 files if ZCAT_COMMAND is not executable.
810 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
812 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
814 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
815 Normally benign, it bites when the pair was led to by a CNAME;
816 modern usage is to not canonicalize the domain to a CNAME target
817 (and we were inconsistent anyway for A-only vs AAAA+A).
819 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
821 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
822 when evaluating $sender_host_dnssec.
824 JH/31 Check the HELO verification lookup for DNSSEC, adding new
825 $sender_helo_dnssec variable.
827 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
829 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
831 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
833 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
834 documented as working, but never had. Support all but $spam_report.
836 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
837 added for tls authenticator.
839 HS/03 Add perl_taintmode main config option
845 TL/01 When running the test suite, the README says that variables such as
846 no_msglog_check are global and can be placed anywhere in a specific
847 test's script, however it was observed that placement needed to be near
848 the beginning for it to behave that way. Changed the runtest perl
849 script to read through the entire script once to detect and set these
850 variables, reset to the beginning of the script, and then run through
851 the script parsing/test process like normal.
853 TL/02 The BSD's have an arc4random API. One of the functions to induce
854 adding randomness was arc4random_stir(), but it has been removed in
855 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
856 function when detected.
858 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
859 cause callback expansion.
861 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
862 syntax errors in an expansion can be treated as a string instead of
863 logging or causing an error, due to the internal use of bool_lax
864 instead of bool when processing it.
866 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
867 server certificates when making smtp deliveries.
869 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
871 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
873 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
875 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
876 Merged patch from Sebastian Wiedenroth.
878 JH/05 Fix results-pipe from transport process. Several recipients, combined
879 with certificate use, exposed issues where response data items split
880 over buffer boundaries were not parsed properly. This eventually
881 resulted in duplicates being sent. This issue only became common enough
882 to notice due to the introduction of connection certificate information,
883 the item size being so much larger. Found and fixed by Wolfgang Breyha.
885 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
886 size buffer was used, resulting in syntax errors when an expansion
889 JH/07 Add support for directories of certificates when compiled with a GnuTLS
890 version 3.3.6 or later.
892 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
893 is EXPERIMENTAL_EVENT, the main-configuration and transport options
894 both become "event_action", the variables become $event_name, $event_data
895 and $event_defer_errno. There is a new variable $verify_mode, usable in
896 routers, transports and related events. The tls:cert event is now also
897 raised for inbound connections, if the main configuration event_action
900 TL/06 In test suite, disable OCSP for old versions of openssl which contained
901 early OCSP support, but no stapling (appears to be less than 1.0.0).
903 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
904 server certificate names available under the smtp transport option
905 "tls_verify_cert_hostname" now do not permit multi-component wildcard
908 JH/10 Time-related extraction expansions from certificates now use the main
909 option "timezone" setting for output formatting, and are consistent
910 between OpenSSL and GnuTLS compilations. Bug 1541.
912 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
913 encoded parameter in the incoming message. Bug 1558.
915 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
916 include certificate info, eximon was claiming there were spoolfile
919 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
921 JH/14 Log delivery-related information more consistently, using the sequence
922 "H=<name> [<ip>]" wherever possible.
924 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
925 are problematic for Debian distribution, omit them from the release
928 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
930 JH/16 Fix string representation of time values on 64bit time_t architectures.
933 JH/17 Fix a null-indirection in certextract expansions when a nondefault
934 output list separator was used.
939 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
940 checkers that were complaining about end of non-void function with no
943 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
944 This was a regression introduced in 4.83 by another bugfix.
946 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
948 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
949 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
955 TF/01 Correctly close the server side of TLS when forking for delivery.
957 When a message was received over SMTP with TLS, Exim failed to clear up
958 the incoming connection properly after forking off the child process to
959 deliver the message. In some situations the subsequent outgoing
960 delivery connection happened to have the same fd number as the incoming
961 connection previously had. Exim would try to use TLS and fail, logging
962 a "Bad file descriptor" error.
964 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
965 utilities have not been installed.
967 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
968 temporary space as the ACL may create new global variables.
970 TL/01 LDAP support uses per connection or global context settings, depending
971 upon the detected version of the libraries at build time.
973 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
974 to extract and use the src ip:port in logging and expansions as if it
975 were a direct connection from the outside internet. PPv2 support was
976 updated based on HAProxy spec change in May 2014.
978 JH/02 Add ${listextract {number}{list}{success}{fail}}.
980 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
981 Properly escape header and check for NULL return.
983 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
986 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
988 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
989 characters in header names, implemented as a verify condition.
990 Contributed by Michael Fischer v. Mollard.
992 TL/05 Rename SPF condition results err_perm and err_temp to standardized
993 results permerror and temperror. Previous values are deprecated but
994 still accepted. In a future release, err_perm and err_temp will be
995 completely removed, which will be a backward incompatibility if the
996 ACL tests for either of these two old results. Patch contributed by
997 user bes-internal on the mailing list.
999 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1001 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1002 selectors, in both main and reject logs.
1004 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1007 JH/07 Add malware type "sock" for talking to simple daemon.
1009 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1011 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1012 routers/transports under cutthrough routing.
1014 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1015 numbers. Touch up "bool" conditional to keep the same definition.
1017 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1019 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1021 JH/12 Expand items in router/transport headers_add or headers_remove lists
1022 individually rather than the list as a whole. Bug 1452.
1024 Required for reasonable handling of multiple headers_ options when
1025 they may be empty; requires that headers_remove items with embedded
1026 colons must have them doubled (or the list-separator changed).
1028 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1029 view the policy declared in the DMARC record. Currently, $dmarc_status
1030 is a combined value of both the record presence and the result of the
1033 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1035 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1036 dnslookup router and the smtp transport (applying to the forward
1039 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1040 of ldap servers used for a specific lookup. Patch provided by Heiko
1043 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1044 New variable $lookup_dnssec_authenticated for observability.
1046 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1047 Patch submitted by Lars Timman.
1049 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1051 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1052 Requires trusted mode and valid format message id, aborts otherwise.
1053 Patch contributed by Heiko Schlichting.
1055 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1056 certextract with support for various fields. Bug 1358.
1058 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1059 is requested by default, modifiable by smtp transport option
1062 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1063 operate on certificate variables to give certificate fingerprints
1064 Also new ${sha256:cert_variable}.
1066 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1068 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1069 Christian Aistleitner.
1071 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1073 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1074 file. Patch from Wolfgang Breyha.
1076 JH/25 Expand the coverage of the delivery $host and $host_address to
1077 client authenticators run in verify callout. Bug 1476.
1079 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1080 align with daemon_smtp_ports. Bug 72.
1082 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1083 support and error reporting did not work properly.
1085 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1086 and is readable. Patch from Andrew Colin Kissa.
1088 TL/14 Enhance documentation of ${run expansion and how it parses the
1089 commandline after expansion, particularly in the case when an
1090 unquoted variable expansion results in an empty value.
1092 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1094 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1095 by renaming away from it. Addresses GH issue 15, reported by
1098 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1100 TL/15 SECURITY: prevent double expansion in math comparison functions
1101 (can expand unsanitized data). Not remotely exploitable.
1108 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1110 PP/02 Make -n do something, by making it not do something.
1111 When combined with -bP, the name of an option is not output.
1113 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1116 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1117 $sender_host_name and config options to manage this, and basic check
1120 PP/05 DSCP support for outbound connections and control modifier for inbound.
1122 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1123 (Only plugin which currently uses this is kerberos4, which nobody should
1124 be using, but we should make it available and other future plugins might
1125 conceivably use it, even though it would break NAT; stuff *should* be
1126 using channel bindings instead).
1128 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1129 name; added for Sendmail compatibility; requires admin caller.
1130 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1131 just ignore it); requires trusted caller.
1132 Also parse but ignore: -Ac -Am -X<logfile>
1135 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1137 TL/02 Add +smtp_confirmation as a default logging option.
1139 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1140 Patch by Magnus Holmgren from 2007-02-20.
1142 TL/04 Bugzilla 1281 - Spec typo.
1143 Bugzilla 1283 - Spec typo.
1144 Bugzilla 1290 - Spec grammar fixes.
1146 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1148 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1150 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1151 mailing list by by Dmitry Isaikin.
1153 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1155 JH/02 Support "G" suffix to numbers in ${if comparisons.
1157 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1159 NM/01 Bugzilla 1197 - Spec typo
1160 Bugzilla 1196 - Spec examples corrections
1162 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1164 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1165 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1168 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1169 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1171 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1172 "acl {{name}{arg}...}", and optional args on acl condition
1175 JH/05 Permit multiple router/transport headers_add/remove lines.
1177 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1179 JH/07 Avoid using a waiting database for a single-message-only transport.
1180 Performance patch from Paul Fisher. Bugzilla 1262.
1182 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1185 JH/09 Add $headers_added variable, with content from use of ACL modifier
1186 add_header (but not yet added to the message). Bugzilla 199.
1188 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1189 Pulled from Bugzilla 817 by Wolfgang Breyha.
1191 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1193 (nb: this is the same fix as in Exim 4.80.1)
1195 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1198 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1201 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1203 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1204 advertise SMTP AUTH mechanism to us, instead of a generic
1205 protocol violation error. Also, make Exim more robust to bad
1206 data from the Dovecot auth socket.
1208 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1210 When a queue runner is handling a message, Exim first routes the
1211 recipient addresses, during which it prunes them based on the retry
1212 hints database. After that it attempts to deliver the message to
1213 any remaining recipients. It then updates the hints database using
1216 So if a recipient address works intermittently, it can get repeatedly
1217 deferred at routing time. The retry hints record remains fresh so the
1218 address never reaches the final cutoff time.
1220 This is a fairly common occurrence when a user is bumping up against
1221 their storage quota. Exim had some logic in its local delivery code
1222 to deal with this. However it did not apply to per-recipient defers
1223 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1225 This change adds a proper retry rule check during routing so that the
1226 final cutoff time is checked against the message's age. We only do
1227 this check if there is an address retry record and there is not a
1228 domain retry record; this implies that previous attempts to handle
1229 the address had the retry_use_local_parts option turned on. We use
1230 this as an approximation for the destination being like a local
1231 delivery, as in LMTP.
1233 I suspect this new check makes the old local delivery cutoff check
1234 redundant, but I have not verified this so I left the code in place.
1236 TF/02 Correct gecos expansion when From: is a prefix of the username.
1238 Test 0254 submits a message to Exim with the header
1242 When I ran the test suite under the user fanf2, Exim expanded
1243 the header to contain my full name, whereas it should have added
1244 a Resent-Sender: header. It erroneously treats any prefix of the
1245 username as equal to the username.
1247 This change corrects that bug.
1249 GF/01 DCC debug and logging tidyup
1250 Error conditions log to paniclog rather than rejectlog.
1251 Debug lines prefixed by "DCC: " to remove any ambiguity.
1253 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1255 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1256 Bug spotted by Jeremy Harris; was flawed since initial commit.
1257 Would have resulted in OCSP responses post-SNI triggering an Exim
1258 NULL dereference and crash.
1260 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1262 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1263 Bug detection, analysis and fix by Samuel Thibault.
1264 Bugzilla 1331, Debian bug #698092.
1266 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1268 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1269 Server implementation by Todd Lyons, client by JH.
1270 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1271 config variable "prdr_enable" controls whether the server
1272 advertises the facility. If the client requests PRDR a new
1273 acl_data_smtp_prdr ACL is called once for each recipient, after
1274 the body content is received and before the acl_smtp_data ACL.
1275 The client is controlled by both of: a hosts_try_prdr option
1276 on the smtp transport, and the server advertisement.
1277 Default client logging of deliveries and rejections involving
1278 PRDR are flagged with the string "PRDR".
1280 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1281 fclose(). Diagnosis by Todd Lyons.
1283 PP/17 Update configure.default to handle IPv6 localhost better.
1284 Patch by Alain Williams (plus minor tweaks).
1287 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1288 This is now consistent with GnuTLS, and is now documented: the
1289 previous undocumented portable approach to treating the option as
1290 unset was to force an expansion failure. That still works, and
1291 an empty string is now equivalent.
1293 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1294 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1295 not performing validation itself.
1297 PP/20 Added force_command boolean option to pipe transport.
1298 Patch from Nick Koston, of cPanel Inc.
1300 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1303 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1305 PP/21 Fix eximon continuous updating with timestamped log-files.
1306 Broken in a format-string cleanup in 4.80, missed when I repaired the
1307 other false fix of the same issue.
1308 Report and fix from Heiko Schlichting.
1311 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1312 Report from Prashanth Katuri.
1314 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1315 It's SecureTransport, so affects any MacOS clients which use the
1316 system-integrated TLS libraries, including email clients.
1318 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1319 using a MIME ACL for non-SMTP local injection.
1320 Report and assistance in diagnosis by Warren Baker.
1322 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1324 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1326 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1327 last id that failed so it may be referenced in subsequent ACL's.
1329 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1332 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1333 ldap library initialization, allowing self-signed CA's to be
1334 used. Also properly sets require_cert option later in code by
1335 using NULL (global ldap config) instead of ldap handle (per
1336 session). Bug diagnosis and testing by alxgomz.
1338 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1339 the src/util/ subdirectory.
1341 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1342 renamed to Transport Post Delivery Action by Jeremy Harris, as
1345 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1346 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1347 redis_servers = needs to be configured which will be used by the redis
1348 lookup. Patch from Warren Baker, of The Packet Hub.
1350 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1352 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1353 hostname or reverse DNS when processing a host list. Used suggestions
1354 from multiple comments on this bug.
1356 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1358 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1359 Missed a few lines, added it to make the runtest require no keyboard
1362 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1363 contains upper case chars. Make router use caseful_local_part.
1365 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1366 support when GnuTLS has been built with p11-kit.
1372 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1374 This, or similar/improved, will also be change PP/11 of 4.82.
1380 PP/01 Handle short writes when writing local log-files.
1381 In practice, only affects FreeBSD (8 onwards).
1382 Bugzilla 1053, with thanks to Dmitry Isaikin.
1384 NM/01 Bugzilla 949 - Documentation tweak
1386 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1389 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1391 PP/02 Implemented gsasl authenticator.
1393 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1395 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1396 `pkg-config foo` for cflags/libs.
1398 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1399 with rest of GSASL and with heimdal_gssapi.
1401 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1402 `pkg-config foo` for cflags/libs for the TLS implementation.
1404 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1405 properties get this fed in as external SSF. A number of robustness
1406 and debugging improvements to the cyrus_sasl authenticator.
1408 PP/08 cyrus_sasl server now expands the server_realm option.
1410 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1411 Patch by Jeremy Harris.
1413 PP/10 Added dbmjz lookup type.
1415 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1417 PP/12 MAIL args handles TAB as well as SP, for better interop with
1418 non-compliant senders.
1419 Analysis and variant patch by Todd Lyons.
1421 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1422 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1423 Patch from Dirk Mueller <dmueller@suse.com>
1425 PP/13 tls_peerdn now print-escaped for spool files.
1426 Observed some $tls_peerdn in wild which contained \n, which resulted
1427 in spool file corruption.
1429 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1430 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1431 or write after TLS renegotiation, which otherwise led to messages
1434 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1435 as a tracking header (ie: a signed header comes before the signature).
1436 Patch from Wolfgang Breyha.
1438 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1439 comma-sep list; embedded commas doubled.
1441 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1443 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1445 Report and patch from Dmitry Banschikov.
1447 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1448 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1449 failures. We appear to now support TLS1.1+ with Exim.
1451 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1452 lets Exim select keys and certificates based upon TLS SNI from client.
1453 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1454 before an outbound SMTP session. New log_selector, +tls_sni.
1456 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1457 NULL dereference. Report and patch from Alun Jones.
1459 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1460 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1461 Not seeing resolver debug output on NetBSD, but suspect this is a
1462 resolver implementation change.
1464 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1465 Left warnings. Added "eximon gdb" invocation mode.
1467 PP/21 Defaulting "accept_8bitmime" to true, not false.
1469 PP/22 Added -bw for inetd wait mode support.
1471 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1472 locate the relevant includes and libraries. Made this the default.
1474 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1475 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1477 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1478 This may cause build issues on older platforms.
1480 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1481 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1482 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1483 Added SNI support via GnuTLS too.
1484 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1486 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1488 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1489 Applied second patch from Janne, implementing suggestion to default
1490 multiple-strings-in-record handling to match SPF spec.
1492 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1494 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1495 read-only, out of scope).
1496 Patch from Wolfgang Breyha, report from Stuart Northfield.
1498 PP/29 Fix three issues highlighted by clang analyser static analysis.
1499 Only crash-plausible issue would require the Cambridge-specific
1500 iplookup router and a misconfiguration.
1501 Report from Marcin Mirosław.
1503 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1505 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1506 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1507 As part of this, removing so much warning spew let me fix some minor
1508 real issues in debug logging.
1510 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1511 assignment on my part. Fixed.
1513 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1514 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1515 Janne Snabb (who went above and beyond: thank you).
1517 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1518 string otherwise requires a connection and a bunch more work and it's
1519 relatively easy to get wrong. Should also expose TLS library linkage
1522 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1523 64-bit ${eval} (JH/03).
1525 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1526 GNU libc to support some of the 64-bit stuff, should not lead to
1527 conflicts. Defined before os.h is pulled in, so if a given platform
1528 needs to override this, it can.
1530 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1531 protection layer was required, which is not implemented.
1532 Bugzilla 1254, patch from Wolfgang Breyha.
1534 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1535 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1536 tls_dhparam take prime identifiers. Also unbreak combination of
1537 OpenSSL+DH_params+TLSSNI.
1539 PP/39 Disable SSLv2 by default in OpenSSL support.
1545 PP/01 Solaris build fix for Oracle's LDAP libraries.
1546 Bugzilla 1109, patch from Stephen Usher.
1548 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1550 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1553 TF/02 Fix a couple more cases where we did not log the error message
1554 when unlink() failed. See also change 4.74-TF/03.
1556 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1557 lock up or crash if it happened to be inside a call to libc when it
1558 got a SIGUSR1 from exiwhat.
1560 The SIGUSR1 handler appends the current process status to the process
1561 log which is later printed by exiwhat. It used to use the general
1562 purpose logging code to do this, but several functions it calls are
1563 not safe for signals.
1565 The new output code in the SIGUSR1 handler is specific to the process
1566 log, and simple enough that it's easy to inspect for signal safety.
1567 Removing some special cases also simplifies the general logging code.
1568 Removing the spurious timestamps from the process log simplifies
1571 TF/04 Improved ratelimit ACL condition.
1573 The /noupdate option has been deprecated in favour of /readonly which
1574 has clearer semantics. The /leaky, /strict, and /readonly update modes
1575 are mutually exclusive. The update mode is no longer included in the
1576 database key; it just determines when the database is updated. (This
1577 means that when you upgrade Exim will forget old rate measurements.)
1579 Exim now checks that the per_* options are used with an update mode that
1580 makes sense for the current ACL. For example, when Exim is processing a
1581 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1582 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1583 must specify per_mail/readonly. If you omit the update mode it defaults to
1584 /leaky where that makes sense (as before) or /readonly where required.
1586 The /noupdate option is now undocumented but still supported for
1587 backwards compatibility. It is equivalent to /readonly except that in
1588 ACLs where /readonly is required you may specify /leaky/noupdate or
1589 /strict/noupdate which are treated the same as /readonly.
1591 A useful new feature is the /count= option. This is a generalization
1592 of the per_byte option, so that you can measure the throughput of other
1593 aggregate values. For example, the per_byte option is now equivalent
1594 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1596 The per_rcpt option has been generalized using the /count= mechanism
1597 (though it's more complicated than the per_byte equivalence). When it is
1598 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1599 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1600 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1601 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1602 Note that using per_rcpt with a non-readonly update mode in more than
1603 one ACL will cause the recipients to be double-counted. (The per_mail
1604 and per_byte options don't have this problem.)
1606 The handling of very low rates has changed slightly. If the computed rate
1607 is less than the event's count (usually one) then this event is the first
1608 after a long gap. In this case the rate is set to the same as this event's
1609 count, so that the first message of a spam run is counted properly.
1611 The major new feature is a mechanism for counting the rate of unique
1612 events. The new per_addr option counts the number of different
1613 recipients that someone has sent messages to in the last time period. It
1614 behaves like per_rcpt if all the recipient addresses are different, but
1615 duplicate recipient addresses do not increase the measured rate. Like
1616 the /count= option this is a general mechanism, so the per_addr option
1617 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1618 example, measure the rate that a client uses different sender addresses
1619 with the options per_mail/unique=$sender_address. There are further
1620 details in the main documentation.
1622 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1624 TF/06 Removed a few PCRE remnants.
1626 TF/07 Automatically extract Exim's version number from tags in the git
1627 repository when doing development or release builds.
1629 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1630 Bugzilla 879. Patch from Paul Fisher.
1632 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1633 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1636 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1638 PP/05 Variable $av_failed, true if the AV scanner deferred.
1639 Bugzilla 1078. Patch from John Horne.
1641 PP/06 Stop make process more reliably on build failure.
1642 Bugzilla 1087. Patch from Heiko Schlittermann.
1644 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1645 Bugzilla 1089. Patch from Heiko Schlittermann.
1647 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1648 Bugzilla 1131. Patch from Holger Weiß.
1650 PP/09 Handle IPv6 addresses with SPF.
1651 Bugzilla 860. Patch from Wolfgang Breyha.
1653 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1655 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1658 PP/11 match_* no longer expand right-hand-side by default.
1659 New compile-time build option, EXPAND_LISTMATCH_RHS.
1660 New expansion conditions, "inlist", "inlisti".
1662 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1664 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1666 PP/14 fix log_write() format string regression from TF/03.
1667 Bugzilla 1152. Patch from Dmitry Isaikin.
1673 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1675 PP/02 Harmonised TLS library version reporting; only show if debugging.
1676 Layout now matches that introduced for other libraries in 4.74 PP/03.
1678 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1680 PP/04 New "dns_use_edns0" global option.
1682 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1685 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1686 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1688 TK/01 Updated PolarSSL code to 0.14.2.
1689 Bugzilla 1097. Patch from Andreas Metzler.
1691 PP/07 Catch divide-by-zero in ${eval:...}.
1692 Fixes bugzilla 1102.
1694 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1697 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1698 format-string attack -- SECURITY: remote arbitrary code execution.
1700 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1701 time unintentionally subject to list matching rules, letting the header
1702 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1703 arbitrary string expansion). This allowed for information disclosure.
1705 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1706 INT_MIN/-1 -- value coerced to INT_MAX.
1712 NM/01 Workaround for PCRE version dependency in version reporting
1715 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1716 This fixes portability to compilers other than gcc, notably
1717 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1719 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1720 makefiles for portability to HP-UX and POSIX correctness.
1722 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1723 Also via indented variable definition in the Makefile.
1724 (Debugging by Oliver Heesakkers).
1726 PP/02 Restore caching of spamd results with expanded spamd_address.
1727 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1729 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1730 Improves build reliability. Fix from: Frank Elsner
1732 NM/02 Fix wide character breakage in the rfc2047 coding
1733 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1735 NM/03 Allow underscore in dnslist lookups
1736 Fixes bug 1026. Patch from Graeme Fowler
1738 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1739 Code patches from Adam Ciarcinski of NetBSD.
1741 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1744 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1745 is logged, to avoid truncation. Patch from John Horne.
1747 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1748 Patch from Jakob Hirsch.
1750 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1751 SQL string expansion failure details.
1752 Patch from Andrey Oktyabrski.
1754 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1755 Patch from Simon Arlott.
1757 PP/09 New lookups functionality failed to compile on old gcc which rejects
1758 extern declarations in function scope.
1759 Patch from Oliver Fleischmann
1761 PP/10 Use sig_atomic_t for flags set from signal handlers.
1762 Check getgroups() return and improve debugging.
1763 Fixed developed for diagnosis in bug 927 (which turned out to be
1766 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1767 Patch from Mark Zealey.
1769 PP/12 Bugzilla 1056: Improved spamd server selection.
1770 Patch from Mark Zealey.
1772 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1773 Based on patch from Heiko Schlittermann.
1775 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1776 Patch from Uwe Doering, sign-off by Michael Haardt.
1778 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1779 variable declaration deep within a block. Bug and patch from
1782 PP/15 lookups-Makefile IRIX compatibility coercion.
1784 PP/16 Make DISABLE_DKIM build knob functional.
1786 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1787 Patch by Simon Arlott
1789 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1790 variable argument macros. Our copy now differs from upstream.
1796 TF/01 Failure to get a lock on a hints database can have serious
1797 consequences so log it to the panic log.
1799 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1800 controlled using the smtp_confirmation log selector.
1802 TF/03 Include the error message when we fail to unlink a spool file.
1804 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1805 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1806 for maintaining out-of-tree patches for some time.
1808 PP/01 Bugzilla 139: Documentation and portability issues.
1809 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1810 Handle per-OS dynamic-module compilation flags.
1812 PP/02 Let /dev/null have normal permissions.
1813 The 4.73 fixes were a little too stringent and complained about the
1814 permissions on /dev/null. Exempt it from some checks.
1815 Reported by Andreas M. Kirchwitz.
1817 PP/03 Report version information for many libraries, including
1818 Exim version information for dynamically loaded libraries. Created
1819 version.h, now support a version extension string for distributors
1820 who patch heavily. Dynamic module ABI change.
1822 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1823 privilege escalation vulnerability whereby the Exim run-time user
1824 can cause root to append content of the attacker's choosing to
1827 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1830 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1831 If dropping privileges for untrusted macros, we disabled normal logging
1832 on the basis that it would fail; for the Exim run-time user, this is not
1833 the case, and it resulted in successful deliveries going unlogged.
1834 Fixed. Reported by Andreas Metzler.
1840 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1841 only prepend for the Resent-* case. Fixes regression introduced in
1842 Exim 4.70 by NM/22 for Bugzilla 607.
1844 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1845 increasing numbers of administrators be bitten by this.
1847 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1849 PP/03 Bugzilla 994: added openssl_options main configuration option.
1851 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1853 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1855 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1857 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1858 without a peer certificate, leading to a segfault because of an
1859 assumption that peers always have certificates. Be a little more
1860 paranoid. Problem reported by Martin Tscholak.
1862 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1863 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1864 NB: ClamAV planning to remove STREAM in "middle of 2010".
1865 CL also introduces -bmalware, various -d+acl logging additions and
1866 more caution in buffer sizes.
1868 PP/09 Implemented reverse_ip expansion operator.
1870 PP/10 Bugzilla 937: provide a "debug" ACL control.
1872 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1874 PP/12 Bugzilla 973: Implement --version.
1876 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1878 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1880 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1882 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1883 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1884 ignore trailing whitespace.
1886 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1888 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1891 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1892 Notification from Dr Andrew Aitchison.
1894 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1895 ExtendedDetectionInfo response format.
1896 Notification from John Horne.
1898 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1901 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1902 XSL and documented dependency on system catalogs, with examples of how
1905 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1908 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1909 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1910 configuration file which is writeable by the Exim user or group.
1912 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1913 of configuration files to cover files specified with the -C option if
1914 they are going to be used with root privileges, not just the default
1917 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1918 option (effectively making it always true).
1920 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1921 files to be used while preserving root privileges.
1923 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1924 that rogue child processes cannot use them.
1926 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1927 run-time user, instead of root.
1929 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1930 Exim run-time user without dropping privileges.
1932 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1933 result string, instead of calling string_vformat() twice with the same
1936 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1937 for other users. Others should always drop root privileges if they use
1938 -C on the command line, even for a whitelisted configure file.
1940 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1942 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1948 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1949 $data_path, and $header_path variables; fixed documentation bugs and
1952 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1953 exipick to access non-standard spools, including the "frozen" queue
1956 NM/01 Bugzilla 965: Support mysql stored procedures.
1957 Patch from Alain Williams
1959 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1961 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1962 Patch from Andreas Metzler
1964 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1965 Patch from Kirill Miazine
1967 NM/05 Bugzilla 671: Added umask to procmail example.
1969 JJ/03 installed exipick 20100323.0, fixing doc bug
1971 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1972 directory. Notification and patch from Dan Rosenberg.
1974 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1976 TK/02 Improve log output when DKIM signing operation fails.
1978 MH/01 Treat the transport option dkim_domain as a colon separated
1979 list, not as a single string, and sign the message with each element,
1980 omitting multiple occurences of the same signer.
1982 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1983 Bugzilla 985, 986. Patch by Simon Arlott
1985 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1986 Patch by Simon Arlott
1988 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1989 MBX locking. Notification from Dan Rosenberg.
1995 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1997 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1999 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2001 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2003 NM/04 Bugzilla 915: Flag broken perl installation during build.
2009 TK/01 Added patch by Johannes Berg that expands the main option
2010 "spamd_address" if it starts with a dollar sign.
2012 TK/02 Write list of recipients to X-Envelope-Sender header when building
2013 the mbox-format spool file for content scanning (suggested by Jakob
2016 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2017 (http://www.dcc-servers.net/) support via dccifd. Activated by
2018 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2020 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2021 by Mark Daniel Reidel <mr@df.eu>.
2023 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2024 When building exim an external PCRE library is now needed -
2025 PCRE is a system library on the majority of modern systems.
2026 See entry on PCRE_LIBS in EDITME file.
2028 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2029 conversation. Added nologin parameter to request.
2030 Patch contributed by Kirill Miazine.
2032 TF/01 Do not log submission mode rewrites if they do not change the address.
2034 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2036 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2037 log files in place. Contributed by Roberto Lima.
2039 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2041 TF/03 Bugzilla 615: When checking the local_parts router precondition
2042 after a local_part_suffix or local_part_prefix option, Exim now
2043 does not use the address's named list lookup cache, since this
2044 contains cached lookups for the whole local part.
2046 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2047 Robert Millan. Documentation is in experimental-spec.txt.
2049 TF/04 Bugzilla 668: Fix parallel build (make -j).
2051 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2053 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2054 Patch provided by Jan Srzednicki.
2056 TF/05 Leading white space used to be stripped from $spam_report which
2057 wrecked the formatting. Now it is preserved.
2059 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2060 that they are available at delivery time.
2062 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2064 TF/08 TLS error reporting now respects the incoming_interface and
2065 incoming_port log selectors.
2067 TF/09 Produce a more useful error message if an SMTP transport's hosts
2068 setting expands to an empty string.
2070 NM/06 Bugzilla 744: EXPN did not work under TLS.
2071 Patch provided by Phil Pennock.
2073 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2074 Patch provided by Richard Godbee.
2076 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2077 acl_smtp_notquit, added index entry.
2079 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2080 Patch provided by Eugene Bujak.
2082 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2083 accept(). Patch provided by Maxim Dounin.
2085 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2086 Patch provided by Phil Pennock.
2088 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2090 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2091 Patch provided by Brad "anomie" Jorsch.
2093 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2094 Patch provided by Dean Brooks.
2096 TK/05 Add native DKIM support (does not depend on external libraries).
2098 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2099 Patch provided by Graeme Fowler.
2101 NM/16 Bugzilla 851: Documentation example syntax fix.
2103 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2105 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2108 NM/19 Bugzilla 745: TLS version reporting.
2109 Patch provided by Phil Pennock.
2111 NM/20 Bugzilla 167: bool: condition support.
2112 Patch provided by Phil Pennock.
2114 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2115 clients. Patch provided by Phil Pennock.
2117 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2118 Patch provided by Brad "anomie" Jorsch.
2120 NM/23 Bugzilla 687: Fix misparses in eximstats.
2121 Patch provided by Heiko Schlittermann.
2123 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2124 Patch provided by Heiko Schlittermann.
2126 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2127 plus update to original patch.
2129 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2131 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2132 Patch provided by David Brownlee.
2134 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2136 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2138 NM/30 Bugzilla 888: TLS documentation bugfixes.
2140 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2142 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2143 Unlike the original bugzilla I have changed all shell scripts in src tree.
2145 NM/33 Bugzilla 898: Transport filter timeout fix.
2146 Patch by Todd Rinaldo.
2148 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2149 Patch by Serge Demonchaux.
2151 NM/35 Bugzilla 39: Base64 decode bug fixes.
2152 Patch by Jakob Hirsch.
2154 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2156 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2158 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2160 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2166 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2167 ALT-N's libdkim that I have put here:
2168 http://duncanthrax.net/exim-experimental/
2170 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2171 were called 'true' and it seems that C99 defines that as a reserved
2172 keyword to be used with 'bool' variable types. That means you could
2173 not include C99-style headers which use bools without triggering
2174 build errors in sieve.c.
2176 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2177 as mailq or other aliases. Changed the --help handling significantly
2178 to do whats expected. exim_usage() emits usage/help information.
2180 SC/01 Added the -bylocaldomain option to eximstats.
2182 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2184 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2186 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2192 PH/01 Another patch from the Sieve maintainer.
2194 PH/02 When an IPv6 address is converted to a string for single-key lookup
2195 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2196 dots are used instead of colons so that keys in lsearch files need not
2197 contain colons. This was done some time before quoting was made available
2198 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2199 (notated using the quote facility) so as to distinguish them from IPv4
2200 keys. This meant that lookups for IP addresses in host lists did not work
2201 for iplsearch lookups.
2203 This has been fixed by arranging for IPv6 addresses to be expressed with
2204 colons if the lookup type is iplsearch. This is not incompatible, because
2205 previously such lookups could never work.
2207 The situation is now rather anomalous, since one *can* have colons in
2208 ordinary lsearch keys. However, making the change in all cases is
2209 incompatible and would probably break a number of configurations.
2211 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2214 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2215 conversion specification without a maximum field width, thereby enabling
2216 a rogue spamd server to cause a buffer overflow. While nobody in their
2217 right mind would setup Exim to query an untrusted spamd server, an
2218 attacker that gains access to a server running spamd could potentially
2219 exploit this vulnerability to run arbitrary code as the Exim user.
2221 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2222 $primary_hostname instead of what libspf2 thinks the hosts name is.
2224 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2225 a directory entry by the name of the lookup key. Previously, if a
2226 symlink pointed to a non-existing file or a file in a directory that
2227 Exim lacked permissions to read, a lookup for a key matching that
2228 symlink would fail. Now it is enough that a matching directory entry
2229 exists, symlink or not. (Bugzilla 503.)
2231 PH/03 The body_linecount and body_zerocount variables are now exported in the
2234 PH/04 Added the $dnslist_matched variable.
2236 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2237 This means they are set thereafter only if the connection becomes
2240 PH/06 Added the client_condition to authenticators so that some can be skipped
2241 by clients under certain conditions.
2243 PH/07 The error message for a badly-placed control=no_multiline_responses left
2244 "_responses" off the end of the name.
2246 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2248 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2249 (without spaces) instead of just copying the configuration text.
2251 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2253 PH/11 Added $max_received_linelength.
2255 PH/12 Added +ignore_defer and +include_defer to host lists.
2257 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2258 way in which PCRE > 7.0 is built.
2260 PH/14 Implemented queue_only_load_latch.
2262 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2263 MAIL command. The effect was to mangle the value on 64-bit systems.
2265 PH/16 Another patch from the Sieve maintainer.
2267 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2269 PH/18 If a system quota error occurred while trying to create the file for
2270 a maildir delivery, the message "Mailbox is full" was not appended to the
2271 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2272 only to a quota excession during the actual writing of the file.
2274 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2275 characters?) which causes problems in log lines. The DN values are now
2276 passed through string_printing() before being added to log lines.
2278 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2279 and InterBase are left for another time.)
2281 PH/21 Added message_body_newlines option.
2283 PH/22 Guard against possible overflow in moan_check_errorcopy().
2285 PH/23 POSIX allows open() to be a macro; guard against that.
2287 PH/24 If the recipient of an error message contained an @ in the local part
2288 (suitably quoted, of course), incorrect values were put in $domain and
2289 $local_part during the evaluation of errors_copy.
2295 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2296 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2299 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2300 issue a MAIL command.
2302 PH/02 In an ACL statement such as
2304 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2306 if a client was not listed at all, or was listed with a value other than
2307 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2308 the condition was not true (as it should be), so access was not denied.
2309 The bug was that the ! inversion was incorrectly passed on to the second
2310 item. This has been fixed.
2312 PH/03 Added additional dnslists conditions == and =& which are different from
2313 = and & when the dns lookup returns more than one IP address.
2315 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2316 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2318 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2319 FSYNC, which compiles an option called disable_fsync that allows for
2320 bypassing fsync(). The documentation is heavily laced with warnings.
2322 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2324 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2325 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2326 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2327 including adding "make clean"; (3) Added -fPIC when compiling the test
2328 dynamically loaded module, to get rid of a warning.
2330 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2331 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2332 = 0s. The bug is otherwise harmless.
2334 PH/07 There was a bug in the dovecot authenticator such that the value of
2335 $auth1 could be overwritten, and so not correctly preserved, after a
2336 successful authentication. This usually meant that the value preserved by
2337 the server_setid option was incorrect.
2339 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2341 PH/09 Installed PCRE release 7.0.
2343 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2344 run for batched SMTP input. It is now run at the start of every message
2345 in the batch. While fixing this I discovered that the process information
2346 (output by running exiwhat) was not always getting set for -bs and -bS
2347 input. This is fixed, and it now also says "batched" for BSMTP.
2349 PH/11 Added control=no_pipelining.
2351 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2352 patch, slightly modified), and move the expansion of helo_data till after
2353 the connection is made in the smtp transport (so it can use these
2356 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2358 PH/14 Added log_selector = +pid.
2360 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2362 PH/16 Add ${if forany and ${if forall.
2364 PH/17 Added dsn_from option to vary the From: line in DSNs.
2366 PH/18 Flush SMTP output before performing a callout, unless control =
2367 no_callout_flush is set.
2369 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2370 was true (the default) a successful delivery failed to delete the retry
2371 item, thus causing premature timeout of the address. The bug is now
2374 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2376 PH/21 Long custom messages for fakedefer and fakereject are now split up
2377 into multiline responses in the same way that messages for "deny" and
2378 other ACL rejections are.
2380 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2381 with slight modification.
2383 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2384 draft, changing the syntax and factoring some duplicate code".
2386 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2387 for deliveries of the second and subsequent messages over the same SMTP
2390 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2391 ${reduce, with only minor "tidies".
2393 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2395 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2396 expansion side effects.
2398 PH/27 When a message times out after an over-quota error from an Exim-imposed
2399 quota, the bounce message says "mailbox is full". This message was not
2400 being given when it was a system quota that was exceeded. It now should
2403 MH/03 Made $recipients available in local_scan(). local_scan() already has
2404 better access to the recipient list through recipients_list[], but
2405 $recipients can be useful in postmaster-provided expansion strings.
2407 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2408 in the case of a MAIL command with additional options following the
2409 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2410 were accidentally chopped off.
2412 PH/29 SMTP synchronization checks are implemented when a command is read -
2413 there is a check that no more input is waiting when there shouldn't be
2414 any. However, for some commands, a delay in an ACL can mean that it is
2415 some time before the response is written. In this time, more input might
2416 arrive, invalidly. So now there are extra checks after an ACL has run for
2417 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2418 pipelining has not been advertised.
2420 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2422 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2423 returned in the SMTP response when smtp_return_error_details was false.
2424 This has been fixed.
2426 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2427 instead of the C I/O that was originally supplied, because problems were
2428 reported on Solaris.
2430 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2431 Exim which did not show up earlier: it was assuming that a call to
2432 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2433 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2434 was a macro that became an assignment, so it seemed to work. This has
2435 changed to a proper function call with a void return, hence the compile
2436 error. Exim's code has been fixed.
2438 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2441 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2443 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2445 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2446 because it turns out to be too big or there is a timeout) there is no
2447 "Completed" line in the log. When some messages of this type were
2448 selected by exigrep, they were listed as "not completed". Others were
2449 picked up by some special patterns. I have improved the selection
2450 criteria to be more general.
2452 PH/38 The host_find_failed option in the manualroute router can now be set
2453 to "ignore", to completely ignore a host whose IP address cannot be
2454 found. If all hosts are ignored, the behaviour is controlled by the new
2455 host_all_ignored option.
2457 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2458 homing or because of multiple MX records with /mx) generated more than
2459 one IP address, and the following item turned out to be the local host,
2460 all the secondary addresses of the first item were incorrectly removed
2461 from the list, along with the local host and any following hosts (which
2462 is what is supposed to happen).
2464 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2465 whoever called Exim into the -H file. In the case of the daemon it was
2466 behaving confusingly. When first started, it used values for whoever
2467 started the daemon, but after a SIGHUP it used the Exim user (because it
2468 calls itself on a restart). I have changed the code so that it now always
2471 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2472 message are rejected with the same error (e.g. no authentication or bad
2473 sender address), and a DATA command is nevertheless sent (as can happen
2474 with PIPELINING or a stupid MUA), the error message that was given to the
2475 RCPT commands is included in the rejection of the DATA command. This is
2476 intended to be helpful for MUAs that show only the final error to their
2479 PH/42 Another patch from the Sieve maintainer.
2481 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2482 Eximstats - Fixed some broken HTML links and added missing column headers
2484 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2485 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2487 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2489 PH/43 Yet another patch from the Sieve maintainer.
2491 PH/44 I found a way to check for a TCP/IP connection going away before sending
2492 the response to the final '.' that terminates a message, but only in the
2493 case where the client has not sent further data following the '.'
2494 (unfortunately, this is allowed). However, in many cases there won't be
2495 any further data because there won't be any more messages to send. A call
2496 to select() can be used: if it shows that the input is "ready", there is
2497 either input waiting, or the socket has been closed. An attempt to read
2498 the next input character can distinguish the two cases. Previously, Exim
2499 would have sent an OK response which the client would never have see.
2500 This could lead to message repetition. This fix should cure that, at
2501 least in a lot of common cases.
2503 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2504 advertised in response to EHLO.
2510 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2511 fixed by 4.65/MH/01 (is this a record?) are fixed:
2513 (i) An empty string was always treated as zero by the numeric comparison
2514 operators. This behaviour has been restored.
2516 (ii) It is documented that the numeric comparison operators always treat
2517 their arguments as decimal numbers. This was broken in that numbers
2518 starting with 0 were being interpreted as octal.
2520 While fixing these problems I realized that there was another issue that
2521 hadn't been noticed. Values of message_size_limit (both the global option
2522 and the transport option) were treated as octal if they started with 0.
2523 The documentation was vague. These values are now always treated as
2524 decimal, and I will make that clear in the documentation.
2530 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2531 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2534 MH/01 Don't check that the operands of numeric comparison operators are
2535 integers when their expansion is in "skipping" mode (fixes bug
2536 introduced by 4.64-PH/07).
2538 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2539 child addresses, Exim now panics and dies. Previously, because the count
2540 is held in a short int, deliveries were likely to be lost. As such a
2541 large number of recipients for a single message is ridiculous
2542 (performance will be very, very poor), I have chosen to impose a limit
2543 rather than extend the field.
2549 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2550 leftover -K file (the existence of which was triggered by #402).
2551 While we were at it, introduced process PID as part of the -K
2552 filename. This should rule out race conditions when creating
2555 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2556 processing considerably. Previous code took too long for large mails,
2557 triggering a timeout which in turn triggers #401.
2559 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2560 in the DK code in transports.c. sendfile() is not really portable,
2561 hence the _LINUX specificness.
2563 TF/01 In the add_headers option to the mail command in an Exim filter,
2564 there was a bug that Exim would claim a syntax error in any
2565 header after the first one which had an odd number of characters
2568 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2569 callout verification, Exim cached a "reject" for the entire domain. This
2570 is correct for most verifications, but it is not correct for a recipient
2571 verification with use_sender or use_postmaster set, because in that case
2572 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2573 case of MAIL FROM:<> rejection from other early rejections (e.g.
2574 rejection of HELO). When verifying a recipient using a non-null MAIL
2575 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2576 Whatever the result of the callout, the value of the domain cache is
2577 left unchanged (for any other kind of callout, getting as far as trying
2578 RCPT means that the domain itself is ok).
2580 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2583 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2584 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2585 session, the daemon ignores ECONNECT errors and logs others; it now
2586 ignores EPIPE as well.
2588 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2589 (quoted-printable decoding).
2591 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2592 later the small subsequent patch to fix an introduced bug.
2594 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2596 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2598 PH/08 An error is now given if message_size_limit is specified negative.
2600 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2601 to be given (somewhat) arbitrary names.
2603 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2606 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2607 miscellaneous code fixes
2609 PH/10 Added the log_reject_target ACL modifier to specify where to log
2612 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2613 hostname. This is wrong, because it relates to the incoming message (and
2614 probably the interface on which it is arriving) and not to the outgoing
2615 callout (which could be using a different interface). This has been
2616 changed to use the value of the helo_data option from the smtp transport
2617 instead - this is what is used when a message is actually being sent. If
2618 there is no remote transport (possible with a router that sets up host
2619 addresses), $smtp_active_hostname is used.
2621 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2622 tweaks were necessary in order to get it to work (see also 21 below):
2623 (a) The code assumed that strncpy() returns a negative number on buffer
2624 overflow, which isn't the case. Replaced with Exim's string_format()
2626 (b) There were several signed/unsigned issues. I just did the minimum
2627 hacking in of casts. There is scope for a larger refactoring.
2628 (c) The code used strcasecmp() which is not a standard C function.
2629 Replaced with Exim's strcmpic() function.
2630 (d) The code set only $1; it now sets $auth1 as well.
2631 (e) A simple test gave the error "authentication client didn't specify
2632 service in request". It would seem that Dovecot has changed its
2633 interface. Fortunately there's a specification; I followed it and
2634 changed what the client sends and it appears to be working now.
2636 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2639 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2640 address A is aliased to B and C, where B exists and C does not. Without
2641 -v the output is "A verified" because verification stops after a
2642 successful redirection if more than one address is generated. However,
2643 with -v the child addresses are also verified. Exim was outputting "A
2644 failed to verify" and then showing the successful verification for C,
2645 with its parentage. It now outputs "B failed to verify", showing B's
2646 parentage before showing the successful verification of C.
2648 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2649 look up a TXT record in a specific list after matching in a combined
2652 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2653 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2654 they consult the DNS. I had assumed they would set it the way they
2655 wanted; and indeed my experiments on Linux seem to show that in some
2656 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2657 To be on the safe side, however, I have now made the interface to
2658 host_find_byname() similar to host_find_bydns(), with an argument
2659 containing the DNS resolver options. The host_find_byname() function now
2660 sets these options at its start, just as host_find_bydns() does. The smtp
2661 transport options dns_qualify_single and dns_search_parents are passed to
2662 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2663 of host_find_byname() use the default settings of RES_DEFNAMES
2664 (qualify_single) but not RES_DNSRCH (search_parents).
2666 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2667 spool_read_header() do less string testing, by means of a preliminary
2668 switch on the second character of optional "-foo" lines. (This is
2669 overdue, caused by the large number of possibilities that now exist.
2670 Originally there were few.) While I was there, I also converted the
2671 str(n)cmp tests so they don't re-test the leading "-" and the first
2672 character, in the hope this might squeeze out yet more improvement.
2674 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2675 flag allowing group syntax was set by the header_syntax check but not
2676 turned off, possible causing trouble later; (2) The flag was not being
2677 set at all for the header_verify test, causing "group"-style headers to
2678 be rejected. I have now set it in this case, and also caused header_
2679 verify to ignore an empty address taken from a group. While doing this, I
2680 came across some other cases where the code for allowing group syntax
2681 while scanning a header line wasn't quite right (mostly, not resetting
2682 the flag correctly in the right place). These bugs could have caused
2683 trouble for malformed header lines. I hope it is now all correct.
2685 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2686 with the "reply" argument non-NULL. The code, however (which originally
2687 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2688 but it didn't always do it. This confused somebody who was copying the
2689 code for some other use. I have removed all the tests.
2691 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2692 feature that was used to support insecure browsers during the U.S. crypto
2693 embargo. It requires special client support, and Exim is probably the
2694 only MTA that supported it -- and would never use it because real RSA is
2695 always available. This code has been removed, because it had the bad
2696 effect of slowing Exim down by computing (never used) parameters for the
2697 RSA_EXPORT functionality.
2699 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2700 authenticator to fail if there's a tab character in the incoming data
2701 (there should never be unless someone is messing about, as it's supposed
2702 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2703 if the connection is using TLS or if the remote IP is the same as the
2704 local IP, and the "valid-client-cert option" if a client certificate has
2707 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2708 authenticators. This can be used for authorization after authentication
2709 succeeds. (In the case of plaintext, it servers for both authentication
2712 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2713 if any retry times were supplied.
2715 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2716 connection, where there is no client IP address to check. In this
2717 situation, the verify now always succeeds.
2719 PH/25 Applied John Jetmore's -Mset patch.
2721 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2723 PH/27 In a string expansion for a processed (not raw) header when multiple
2724 headers of the same name were present, leading whitespace was being
2725 removed from all of them, but trailing whitespace was being removed only
2726 from the last one. Now trailing whitespace is removed from each header
2727 before concatenation. Completely empty headers in a concatenation (as
2728 before) are ignored.
2730 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2731 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2733 PH/29 [Removed. This was a change that I later backed out, and forgot to
2734 correct the ChangeLog entry (that I had efficiently created) before
2735 committing the later change.]
2737 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2738 address errors (4xx response to RCPT) over the same connection as other
2739 messages routed to the same hosts. Such deliveries are always "forced",
2740 so retry times are not inspected. This resulted in far too many retries
2741 for the affected addresses. The effect occurred only when there were more
2742 hosts than the hosts_max_try setting in the smtp transport when it had
2743 the 4xx errors. Those hosts that it had tried were not added to the list
2744 of hosts for which the message was waiting, so if all were tried, there
2745 was no problem. Two fixes have been applied:
2747 (i) If there are any address or message errors in an SMTP delivery, none
2748 of the hosts (tried or untried) are now added to the list of hosts
2749 for which the message is waiting, so the message should not be a
2750 candidate for sending over the same connection that was used for a
2751 successful delivery of some other message. This seems entirely
2752 reasonable: after all the message is NOT "waiting for some host".
2753 This is so "obvious" that I'm not sure why it wasn't done
2754 previously. Hope I haven't missed anything, but it can't do any
2755 harm, as the worst effect is to miss an optimization.
2757 (ii) If, despite (i), such a delivery is accidentally attempted, the
2758 routing retry time is respected, so at least it doesn't keep
2759 hammering the server.
2761 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2762 in ${readsocket because some servers need this prod.
2764 PH/32 Added some extra debug output when updating a wait-xxx database.
2766 PH/33 The hint "could be header name not terminated by colon", which has been
2767 given for certain expansion errors for a long time, was not being given
2768 for the ${if def:h_colon_omitted{... case.
2770 PH/34 The spec says: "With one important exception, whenever a domain list is
2771 being scanned, $domain contains the subject domain." There was at least
2772 one case where this was not true.
2774 PH/35 The error "getsockname() failed: connection reset by peer" was being
2775 written to the panic log as well as the main log, but it isn't really
2776 panic-worthy as it just means the connection died rather early on. I have
2777 removed the panic log writing for the ECONNRESET error when getsockname()
2780 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2781 runs only) independently of the message's sender address. This meant
2782 that, if the 4xx error was in fact related to the sender, a different
2783 message to the same recipient with a different sender could confuse
2784 things. In particular, this can happen when sending to a greylisting
2785 server, but other circumstances could also provoke similar problems.
2786 I have changed the default so that the retry time for these errors is now
2787 based a combination of the sender and recipient addresses. This change
2788 can be overridden by setting address_retry_include_sender=false in the
2791 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2792 remote server are returned as part of bounce messages. This was not
2793 happening for LMTP over a pipe (the lmtp transport), but now it is the
2794 same for both kinds of LMTP.
2796 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2797 in header lines that were in fact CNAMEs. This is no longer the case.
2799 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2800 and queue runs started by the daemon processed all messages. This has
2801 been fixed so that -R and -S can now usefully be given with -q<time>.
2803 PH/40 Import PCRE release 6.7 (fixes some bugs).
2805 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2807 PH/42 Give an error if -q is specified more than once.
2809 PH/43 Renamed the variables $interface_address and $interface_port as
2810 $received_ip_address and $received_port, to make it clear that these
2811 values apply to message reception, and not to the outgoing interface when
2812 a message is delivered. (The old names remain recognized, of course.)
2814 PH/44 There was no timeout on the connect() call when using a Unix domain
2815 socket in the ${readsocket expansion. There now is.
2817 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2818 be meaningful with "accept".
2820 SC/01 Eximstats V1.43
2821 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2823 SC/02 Eximstats V1.44
2824 Use a glob alias rather than an array ref in the generated
2825 parser. This improves both readability and performance.
2827 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2828 Collect SpamAssassin and rejection statistics.
2829 Don't display local sender or destination tables unless
2830 there is data to show.
2831 Added average volumes into the top table text output.
2833 SC/04 Eximstats V1.46
2834 Collect data on the number of addresses (recipients)
2835 as well as the number of messages.
2837 SC/05 Eximstats V1.47
2838 Added 'Message too big' to the list of mail rejection
2839 reasons (thanks to Marco Gaiarin).
2841 SC/06 Eximstats V1.48
2842 Mainlog lines which have GMT offsets and are too short to
2843 have a flag are now skipped.
2845 SC/07 Eximstats V1.49 (Alain Williams)
2846 Added the -emptyok flag.
2848 SC/08 Eximstats V1.50
2849 Fixes for obtaining the IP address from reject messages.
2851 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2852 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2853 whitespace changes from 4.64-PH/27
2855 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2858 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2859 are found, allow negative numbers in numeric criteria)
2861 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2863 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2864 to match changes made in 4.64-PH/43
2866 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2868 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2869 "domain" is actually an IP address, and force a failure. This locks out
2870 those revolvers/nameservers that support "A-for-A" lookups, in
2871 contravention of the specifications.
2873 PH/48 When a host name was looked up from an IP address, and the subsequent
2874 forward lookup of the name timed out, the host name was left in
2875 $sender_host_name, contrary to the specification.
2877 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2878 restricted to single-key lookups, Exim was not diagnosing an error if
2879 * or *@ was used with a query-style lookup.
2881 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2883 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2884 long ago, but noone interested enough thought of it. Let's just say that
2885 the "1.1" means that there are some new functions that weren't there at
2886 some point in the past.
2888 PH/51 Error processing for expansion failure of helo_data from an smtp
2889 transport during callout processing was broken.
2891 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2892 tested/used via the -bh/-bhc/-bs options.
2894 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2895 bug, fixed in subsequent PCRE releases).
2897 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2898 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2900 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2906 SC/01 Use a glob alias rather than an array ref in eximstats generated
2907 parser. This improves both readability and performance.
2909 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2910 Don't display local sender or destination tables in eximstats unless
2911 there is data to show.
2912 Added average volumes into the eximstats top table text output.
2914 SC/03 Collect data on the number of addresses (recipients) as well
2915 as the number of messages in eximstats.
2917 TF/01 Correct an error in the documentation for the redirect router. Exim
2918 does (usually) call initgroups() when daemonizing.
2920 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2921 with consistent privilege compared to when running as a daemon.
2923 TF/03 Note in the spec that $authenticated_id is not set for local
2924 submissions from trusted users.
2926 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2927 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2929 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2930 by adding some example configuration directives to the default
2931 configuration file. A little bit of work is required to uncomment the
2932 directives and define how usernames and passwords are checked, but
2933 there is now a framework to start from.
2935 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2936 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2937 without this. I don't know how relevant this is to other LDAP libraries.
2939 PH/02 Add the verb name to the "unknown ACL verb" error.
2941 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2943 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2945 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2946 directory not expanded when it should be if an expanded home directory
2947 was set for the address (which is overridden by the transport).
2949 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2952 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2953 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2954 because it is too late at that time, and has no effect.
2956 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2957 security issue with \' (bugzilla #107). I could not use the
2958 PQescapeStringConn() function, because it needs a PGconn value as one of
2961 PH/08 When testing addresses using -bt, indicate those final addresses that
2962 are duplicates that would not cause an additional delivery. At least one
2963 person was confused, thinking that -bt output corresponded to deliveries.
2964 (Suppressing duplicates isn't a good idea as you lose the information
2965 about possibly different redirections that led to the duplicates.)
2967 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2968 systems where poll() doesn't work, in particular OS X.
2970 PH/10 Added more information to debugging output for retry time not reached.
2972 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2973 operations in malware.c.
2975 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2978 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2979 syslog_duplication set false, log lines that would normally be written
2980 both the the main log and to the reject log were not written to syslog at
2983 PH/14 In the default configuration, change the use of "message" in ACL warn
2984 statements to "add_header".
2986 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2987 not followed by a command (e.g. "seen endif").
2989 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2990 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2993 PH/17 Added extra conditions to the default value of delay_warning_condition
2997 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2998 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2999 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3002 The Auto-Submitted: and various List- headers are standardised, whereas I
3003 don't think Precedence: ever was.
3005 PH/18 Refactored debugging code in route_finduser() to show more information,
3006 in particular, the error code if getpwnam() issues one.
3008 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3009 This is apparently needed in addition to the PH/07 change above to avoid
3010 any possible encoding problems.
3012 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3013 but not after initializing Perl.
3015 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3016 output them only if debugging. By default they are written stderr,
3017 apparently, which is not desirable.
3019 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3022 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3025 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3027 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3028 authenticated or an ident call has been made. Suppress the default
3029 values for $authenticated_id and $authenticated_sender (but permit -oMai
3030 and -oMas) when testing with -bh.
3032 PH/24 Re-jigged the order of the tests in the default configuration so that the
3033 tests for valid domains and recipients precede the DNS black list and CSA
3034 tests, on the grounds that those ones are more expensive.
3036 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3037 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3038 This bug exists in every version of Exim that I still have, right back to
3041 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3042 However, an attempt to turn on case-sensitivity in a regex key by
3043 including (?-i) didn't work because the subject string was already
3044 lowercased, and the effects were non-intuitive. It turns out that a
3045 one-line patch can be used to allow (?-i) to work as expected.
3051 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3052 other effects) broke the use of negated acl sub-conditions.
3054 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3057 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3058 "Deny" causes Exim to reject the incoming connection with a 554 error.
3059 Unfortunately, if there is a major crisis, such as a disk failure,
3060 tcp-wrappers gives "deny", whereas what one would like would be some
3061 kind of temporary error. A kludge has been added to help with this.
3062 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3063 554 error is used if errno is still zero or contains ENOENT (which occurs
3064 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3067 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3069 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3070 errors. Otherwise a message that provokes a temporary error (when other
3071 messages do not) can cause a whole host to time out.
3073 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3074 addresses were routed directly to files or pipes from a redirect router.
3075 File deliveries just didn't batch; pipe deliveries might have suffered
3078 PH/06 A failure to get a lock for a hints database would erroneously always say
3079 "Failed to get write lock", even when it was really a read lock.
3081 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3082 of 0600. This has been changed to use the value of the lockfile_mode
3083 option (which defaults to 0600).
3085 PH/08 Applied small patch from the Sieve maintainer.
3087 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3088 folder from quota calculations, a direct delivery into this folder messed
3089 up the contents of the maildirsize file. This was because the regex was
3090 used only to exclude .Trash (or whatever) when the size of the mailbox
3091 was calculated. There was no check that a delivery was happening into an
3092 excluded directory. This bug has been fixed by ignoring all quota
3093 processing for deliveries into excluded directories.
3095 PH/10 Added the maildirfolder_create_regex option to appendfile.
3101 PH/01 The code for finding all the local interface addresses on a FreeBSD
3102 system running IPv6 was broken. This may well have applied to all BSD
3103 systems, as well as to others that have similar system calls. The broken
3104 code found IPv4 interfaces correctly, but gave incorrect values for the
3105 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3106 that it would not match correctly against @[] and not recognize the IPv6
3109 PH/02 The ipliteral router was not recognizing addresses of the form user@
3110 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3112 PH/03 Added disable_ipv6.
3114 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3115 decoded form, because it is most often used to construct To: headers
3116 lines in autoreplies, and the decoded form may well be syntactically
3117 invalid. However, $reply_address has leading white space removed, and all
3118 newlines turned into spaces so that the autoreply transport does not
3121 PH/05 If group was specified without a user on a router, and no group or user
3122 was specified on a transport, the group from the router was ignored.
3124 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3125 for visible compile-time settings that can be used to change these
3126 numbers, for those that want even more. Backwards compatibility with old
3127 spool files has been maintained. However, going back to a previous Exim
3128 release will lost any variables that are in spool files.
3130 PH/07 Two small changes when running in the test harness: increase delay when
3131 passing a TCP/IP connection to a new process, in case the original
3132 process has to generate a bounce, and remove special handling of
3133 127.0.0.2 (sic), which is no longer necessary.
3135 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3136 be the same on different OS.
3138 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3141 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3142 whether --show-vars was specified or not
3144 JJ/02 exipick: Added support for new ACL variable spool format introduced
3147 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3148 syntactically invalid From: or Reply-to: line, and a filter used this to
3149 generate an autoreply, and therefore failed to obtain an address for the
3150 autoreply, Exim could try to deliver to a non-existent relative file
3151 name, causing unrelated and misleading errors. What now happens is that
3152 it logs this as a hard delivery error, but does not attempt to create a
3155 PH/11 The exinext utility has a -C option for testing purposes, but although
3156 the given file was scanned by exinext itself; it wasn't being passed on
3157 when Exim was called.
3159 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3160 an end-of-file indication when reading a command response.
3162 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3163 compiled. In many other places in Exim, IPv6 addresses are always
3164 recognized, so I have changed this. It also means that IPv4 domain
3165 literals of the form [IPV4:n.n.n.n] are now always recognized.
3167 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3168 used if the router is not running as root, for example, when verifying at
3169 ACL time, or when using -bh. The debugging output from this situation was
3170 non-existent - all you got was a failure to exec. I have made two
3173 (a) Failures to set uid/gid, the current directory, or a process leader
3174 in a subprocess such as that created by queryprogram now generate
3175 suitable debugging output when -d is set.
3177 (b) The queryprogram router detects when it is not running as root,
3178 outputs suitable debugging information if -d is set, and then runs
3179 the subprocess without attempting to change uid/gid.
3181 PH/15 Minor change to Makefile for building test_host (undocumented testing
3184 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3185 additional section of a DNS packet that returns MX or SRV records.
3186 Instead, it always explicitly searches for A/AAAA records. This avoids
3187 major problems that occur when a DNS server includes only records of one
3188 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3189 fixed another bug: if SRV records were looked up and the corresponding
3190 address records were *not* found in the additional section, the port
3191 values from the SRV records were lost.
3193 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3194 using the correct key (the original address) when searching the retry
3195 rules in order to find which one to use for generating the retry hint.
3197 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3198 adding the default one. Similarly, if it contains a Reply-To: header, the
3199 errors_reply_to option, if set, is not used.
3201 PH/19 When calculating a retry time, Exim used to measure the "time since
3202 failure" by looking at the "first failed" field in the retry record. Now
3203 it does not use this if it is later than than the arrival time of the
3204 message. Instead it uses the arrival time. This makes for better
3205 behaviour in cases where some deliveries succeed, thus re-setting the
3206 "first failed" field. An example is a quota failure for a huge message
3207 when small messages continue to be delivered. Without this change, the
3208 "time since failure" will always be short, possible causing more frequent
3209 delivery attempts for the huge message than are intended.
3210 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3212 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3213 $1, $2, $3) because the numerical variables can be reset during some
3214 expansion items (e.g. "match"), thereby losing the authentication data.
3216 PH/21 Make -bV show the size of off_t variables so that the test suite can
3217 decide whether to run tests for quotas > 2G.
3219 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3220 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3221 filecount value is greater than 2G or if a quota value is greater than 2G
3222 on a system where the size of off_t is not greater than 4, a panic error
3225 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3226 never match. The debug and -bh output now contains an explicit error
3227 message indicating a malformed IPv4 address or mask.
3229 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3230 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3231 PH/23 above applies.
3233 PH/25 Do not write to syslog when running in the test harness. The only
3234 occasion when this arises is a failure to open the main or panic logs
3235 (for which there is an explicit test).
3237 PH/26 Added the /no_tell option to "control=freeze".
3239 PH/27 If a host name lookup failed very early in a connection, for example, if
3240 the IP address matched host_lookup and the reverse lookup yielded a name
3241 that did not have a forward lookup, an error message of the form "no IP
3242 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3243 could be logged. Now it outputs the IP address instead of "NULL".
3245 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3246 allows the sender and the authenticated sender to be set when
3247 submitting a message from within Exim. Since child_open_exim() is
3248 documented for local_scan(), the new function should be too.
3250 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3251 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3252 results in an empty string is now treated as unset.
3254 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3256 PH/31 Added sender_verify_fail logging option.
3258 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3259 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3260 tidied the source and removed it altogether.
3262 PH/33 When a queue run was abandoned because the load average was too high, a
3263 log line was always written; now it is written only if the queue_run log
3264 selector is set. In addition, the log line for abandonment now contains
3265 information about the queue run such as the pid. This is always present
3266 in "start" and "stop" lines but was omitted from the "abandon" line.
3268 PH/34 Omit spaces between a header name and the colon in the error message that
3269 is given when verify = headers_syntax fails (if there are lots of them,
3270 the message gets confusing).
3272 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3273 names, as there are now some PTR records that contain slashes. This check
3274 is only to protect against broken name servers that fall over on strange
3275 characters, so the fact that it applies to all lookups doesn't matter.
3277 PH/36 Now that the new test suite is complete, we can remove some of the
3278 special code in Exim that was needed for the old test suite. For example,
3279 sorting DNS records because real resolvers return them in an arbitrary
3280 order. The new test suite's fake resolver always returns records in the
3283 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3284 bounces) except when queue_only is set, to avoid logging races between
3285 the different processes.
3287 PH/38 Panic-die if .include specifies a non-absolute path.
3289 PH/39 A tweak to the "H" retry rule from its user.
3291 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3292 a label. They prevented compilation on older perls.
3294 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3295 a warning to be raised on newish perls.
3297 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3298 on queue. Changes to match documented behaviour of showing count of
3299 messages matching specified criteria.
3301 PH/40 Changed the default ident timeout from 30s to 5s.
3303 PH/41 Added support for the use of login_cap features, on those BSD systems
3304 that have them, for controlling the resources used by pipe deliveries.
3306 PH/42 The content-scanning code uses fopen() to create files in which to put
3307 message data. Previously it was not paying any attention to the mode of
3308 the files. Exim runs with umask(0) because the rest of the code creates
3309 files with open(), and sets the required mode explicitly. Thus, these
3310 files were ending up world-writeable. This was not a big issue, because,
3311 being within the spool directory, they were not world-accessible. I have
3312 created a function called modefopen, which takes an additional mode
3313 argument. It sets umask(777), creates the file, chmods it to the required
3314 mode, then resets the umask. All the relevant calls to fopen() in the
3315 content scanning code have been changed to use this function.
3317 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3318 to 24 hours. This avoids potential overflow problems when processing G
3319 and H retry rules. I suspect nobody ever tinkers with this value.
3321 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3323 PH/45 When the plaintext authenticator is running as a client, the server's
3324 challenges are checked to ensure they are valid base64 strings. By
3325 default, the authentication attempt is cancelled if an invalid string is
3326 received. Setting client_ignore_invalid_base64 true ignores these errors.
3327 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3328 they are received. Thus, the responses can be made to depend on the
3329 challenges. If an invalid string is ignored, an empty string is placed in
3332 PH/46 Messages that are created by the autoreply transport now contains a
3333 References: header, in accordance with RFCs 2822 and 3834.
3335 PH/47 Added authenticated_sender_force to the smtp transport.
3337 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3339 PH/49 Installed latest patch from the Sieve maintainer.
3341 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3342 was also set, the appendfile transport was unnecessarily scanning a
3343 directory of message files (e.g. for maildir delivery) to find the count
3344 of files (along with the size), even though it did not need this
3345 information. It now does the scan only if it needs to find either the
3346 size of the count of files.
3348 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3350 PH/52 Two bugs concerned with error handling when the smtp transport is
3353 (i) Exim was not creating retry information for temporary errors given
3354 for individual recipients after the DATA command when the smtp transport
3355 was used in LMTP mode. This meant that they could be retried too
3356 frequently, and not timed out correctly.
3358 (ii) Exim was setting the flag that allows error details to be returned
3359 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3360 recipients that were returned after the DATA command.
3362 PH/53 This is related to PH/52, but is more general: for any failing address,
3363 when detailed error information was permitted to be returned to the
3364 sender, but the error was temporary, then after the final timeout, only
3365 "retry timeout exceeded" was returned. Now it returns the full error as
3366 well as "retry timeout exceeded".
3368 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3369 do this, and (what is worse) MTAs that accept it.
3371 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3372 will now be deprecated.
3374 PH/56 New os.c-cygwin from the Cygwin maintainer.
3376 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3377 formats (previously only available in exim formats via -bpr, -bpru,
3378 and -bpra. Now also available in native and exiqgrep formats)
3380 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3381 with very large, slow to parse queues
3383 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3385 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3387 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3388 responses to authentication challenges, though it was showing the
3389 challenges; (ii) I've removed the CR characters from the debug output for
3392 PH/58 Allow for the insertion of a newline as well as a space when a string
3393 is turned into more than one encoded-word during RFC 2047 encoding. The
3394 Sieve code now uses this.
3396 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3397 data_4xx, lost_connection, tls_required.
3399 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3400 message was being sent as an SMTP response.
3402 PH/61 Add -l and -k options to exicyclog.
3404 PH/62 When verifying, if an address was redirected to one new address, so that
3405 verification continued, and the new address failed or deferred after
3406 having set something in $address_data, the value of $address_data was not
3407 passed back to the ACL. This was different to the case when no
3408 redirection occurred. The value is now passed back in both cases.
3410 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3411 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3412 use login_cap.h, so on its own it isn't the distinguishing feature. The
3413 new name refers directly to the setclassresources() function.
3415 PH/65 Added configuration files for NetBSD3.
3417 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3419 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3420 is preferred over IPv4.
3422 PH/68 The bounce_return_message and bounce_return_body options were not being
3423 honoured for bounces generated during the reception of non-SMTP messages.
3424 In particular, this applied to messages rejected by the ACL. This bug has
3425 been fixed. However, if bounce_return_message is true and bounce_return_
3426 body is false, the headers that are returned for a non-SMTP message
3427 include only those that have been read before the error was detected.
3428 (In the case of an ACL rejection, they have all been read.)
3430 PH/69 The HTML version of the specification is now built in a directory called
3431 spec_html instead of spec.html, because the latter looks like a path with
3432 a MIME-type, and this confuses some software.
3434 PH/70 Catch two compiler warnings in sieve.c.
3436 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3437 function verify_get_ident() calls ip_connect() to connect a socket, but
3438 if the "connect()" function timed out, ip_connect() used to close the
3439 socket. However, verify_get_ident() also closes the socket later, and in
3440 between Exim writes to the log, which may get opened at this point. When
3441 the socket was closed in ip_connect(), the log could get the same file
3442 descriptor number as the socket. This naturally causes chaos. The fix is
3443 not to close the socket in ip_connect(); the socket should be closed by
3444 the function that creates it. There was only one place in the code where
3445 this was missing, in the iplookup router, which I don't think anybody now
3446 uses, but I've fixed it anyway.
3448 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3449 well as to direct DNS lookups. Otherwise the handling of names in host
3450 lists is inconsistent and therefore confusing.
3456 PH/01 Two changes to the default runtime configuration:
3458 (1) Move the checks for relay_from_hosts and authenticated clients from
3459 after to before the (commented out) DNS black list checks.
3461 (2) Add control=submission to the relay_from_hosts and authenticated
3462 clients checks, on the grounds that messages accepted by these
3463 statements are most likely to be submissions.
3465 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3467 (1) Generate an error if the third argument for the ${prvs expansion is
3470 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3473 (3) Reset the variables that are obtained from the first argument of
3474 ${prvscheck and used in the second argument before leaving the code,
3475 because their memory is reclaimed, so using them afterwards may do
3478 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3479 one (it's much easier than Tom thought :-).
3481 (5) Because of (4), we can now allow for the use of $prvscheck_result
3482 inside the third argument.
3484 PH/03 For some reason, the default setting of PATH when running a command from
3485 a pipe transport was just "/usr/bin". I have changed it to
3488 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3489 anything to be listed in the output from -bV.
3491 PH/05 When a filter generated an autoreply, the entire To: header line was
3492 quoted in the delivery log line, like this:
3494 => >A.N.Other <ano@some.domain> <original@ddress> ...
3496 This has been changed so that it extracts the operative address. There
3497 may be more than one such address. If so, they are comma-separated, like
3500 => >ano@some.domain,ona@other.domain <original@ddress> ...
3502 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3503 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3504 not being looked up in the rDNS to get a host name, Exim was showing the
3505 IP address twice in Received: lines, even though the IP addresses were
3506 identical. For example:
3508 Received: from [1.2.3.4] (helo=[1.2.3.4])
3510 However, if the real host name was known, it was omitting the HELO data
3511 if it matched the actual IP address. This has been tidied up so that it
3512 doesn't show the same IP address twice.
3514 PH/07 When both +timestamp and +memory debugging was on, the value given by
3515 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3516 called by the debug printing, thereby overwriting the timestamp buffer.
3517 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3519 PH/08 When the original message was included in an autoreply transport, it
3520 always said "this is a copy of the message, including all the headers",
3521 even if body_only or headers_only was set. It now gives an appropriate
3524 PH/09 Applied a patch from the Sieve maintainer which:
3526 o fixes some comments
3527 o adds the (disabled) notify extension core
3528 o adds some debug output for the result of if/elsif tests
3529 o points to the current vacation draft in the documentation
3530 and documents the missing references header update
3534 o fixes a bug in processing the envelope test (when testing
3535 multiple envelope elements, the last element determined the
3538 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3539 Electronic Mail") by including:
3541 Auto-submitted: auto-generated
3543 in the messages that it generates (bounce messages and others, such as
3544 warnings). In the case of bounce messages for non-SMTP messages, there was
3545 also a typo: it was using "Auto_submitted" (underscore instead of
3546 hyphen). Since every message generated by Exim is necessarily in response
3547 to another message, thes have all been changed to:
3549 Auto-Submitted: auto-replied
3551 in accordance with these statements in the RFC:
3553 The auto-replied keyword:
3555 - SHOULD be used on messages sent in direct response to another
3556 message by an automatic process,
3558 - MUST NOT be used on manually-generated messages,
3560 - MAY be used on Delivery Status Notifications (DSNs) and Message
3561 Disposition Notifications (MDNs),
3563 - MUST NOT be used on messages generated by automatic or periodic
3564 processes, except for messages which are automatic responses to
3567 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3568 to the default Received: header definition.
3570 PH/12 Added log selector acl_warn_skipped (default on).
3572 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3573 variables because (a) they are in the wrong storage pool and (b) even if
3574 they were copied, it wouldn't work properly because of the caching.
3576 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3577 checking when decoding. Apparently there are clients that generate
3578 overlong encoded strings. Why am I not surprised?
3580 PH/15 If the first argument of "${if match_address" was not empty, but did not
3581 contain an "@" character, Exim crashed. Now it writes a panic log message
3582 and treats the condition as false.
3584 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3586 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3587 extension "envelope-auth". The code is finished and in agreement with
3588 other implementations, but there is no documentation so far and in fact,
3589 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3590 not changing the active code.
3592 Print executed "if" and "elsif" statements when debugging is used. This
3593 helps a great deal to understand what a filter does.
3595 Document more things not specified clearly in RFC3028. I had all this
3596 sorted out, when out of a sudden new issues came to my mind. Oops."
3598 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3601 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3602 canonical form (as documented). However, after a host name lookup from
3603 the IP address, check_host() was doing a simple string comparison with
3604 addresses acquired from the DNS when checking that the found name did
3605 have the original IP as one of its addresses. Since any found IPv6
3606 addresses are likely to be in abbreviated form, the comparison could
3607 fail. Luckily, there already exists a function for doing the comparison
3608 by converting both addresses to binary, so now that is used instead of
3609 the text comparison.
3611 PH/20 There was another similar case to PH/19, when a complete host name was
3612 given in a host list; looking up its IP address could give an abbreviated
3613 form, whereas the current host's name might or might not be abbreviated.
3614 The same fix has been applied.
3620 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3621 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3624 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3625 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3627 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3629 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3630 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3631 Also, for "Auto-Submitted", treat anything other than "no" as
3632 non-personal, in accordance with RFC 3834. (Previously it treated
3633 anything starting "auto-" as non-personal.)
3635 TF/01 The control=submission/name=... option had a problem with syntax
3636 errors if the name included a slash character. The /name= option
3637 now slurps the rest of the string, so it can include any characters
3638 but it must come last in the list of options (after /sender_retain
3641 PH/05 Some modifications to the interface to the fake nameserver for the new
3649 TK/01 Added the "success_on_redirect" address verification option. See
3650 NewStuff for rationale and an example.
3652 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3654 PH/02 Patch to exigrep to allow it to work on syslog lines.
3656 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3657 fread() to skip over the body file's header line, because in Cygwin the
3658 header line is locked and is inaccessible.
3660 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3661 co-exist for some time) to make it clear that it is the Exim ID that is
3662 referenced, not the Message-ID: header line.
3664 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3665 string_format() function, because snprintf() does not exist on all
3668 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3669 ${stat: expansion item.
3671 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3672 protocol synchronization error", to keep the pedants happy.
3674 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3675 well as for IRIX systems, when gcc is being used. See the host.c source
3678 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3680 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3683 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3684 transport and to the smtp transport in LMTP mode.
3686 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3688 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3689 run a filter in a subprocess. This could lead to confusion in subsequent
3690 lookups in the parent process. There should also be a search_tidyup() at
3691 the end of the subprocess.
3693 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3694 only if the host matched helo_try_verify_hosts, which caused the
3695 verification to occur when the EHLO/HELO command was issued. The ACL just
3696 tested the remembered result. Now, if a previous verification attempt has
3697 not happened, "verify = helo" does it there and then.
3699 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3701 TK/03 Fix log output including CR from clamd.
3703 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3704 exist provoked a memory error which could cause a segfault.
3706 PH/15 Installed PCRE 6.2
3708 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3710 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3711 of the problem. Specifically, suggested +O2 rather than +O1 for the
3714 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3716 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3717 record, the port was not correctly logged when the outgoing_port log
3718 selector was set (it logged the transort's default port).
3720 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3721 fallback_hosts, and "hosts" in the smtp transport.
3723 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3724 host errors such as "Connection refused".
3726 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3727 authentication with radiusclient 0.4.9:
3729 - Error returned from rc_read_config was caught wrongly
3730 - Username/password not passed on to radius server due to wrong length.
3732 The presumption is that some radiusclient API changes for 4.51/PH/17
3733 were not taken care of correctly. The code is still untested by me (my
3734 Linux distribution still has 0.3.2 of radiusclient), but it was
3735 contributed by a Radius user.
3737 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3738 expanding the "port" option of the smtp transport.
3740 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3741 while reading a MIME header. Thanks to Tom Hughes for a patch.
3743 PH/24 Include config.h inside local_scan.h so that configuration settings are
3746 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3747 that in an ACL for RCPT (for example), you can examine exactly what was
3750 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3751 commands, but it was not correctly comparing the address with the actual
3752 client host address. Thus, it would show the EHLO address in Received:
3753 header lines when this was not necessary.
3755 PH/27 Added the % operator to ${eval:}.
3757 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3758 it should be ignoring failures (because with -C, for example, it has lost
3759 privilege). It wasn't ignoring creation failures other than "already
3762 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3765 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3766 because an input line was too long, either on its own, or by virtue of
3767 too many continuations, the temporary file was not being removed, and the
3768 return code was incorrect.
3770 PH/31 Missing "BOOL" in function definition in filtertest.c.
3772 PH/32 Applied Sieve patches from the maintainer.
3774 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3776 PH/33 Added "verify = not_blind".
3778 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3779 Local/Makefile (with some defaults set). These are used in built scripts
3780 such as exicyclog, but they have never been used in the exim_install
3781 script (though there are many overriding facilities there). I have
3782 arranged that the exim_install script now takes note of these two
3785 PH/35 Installed configuration files for Dragonfly.
3787 PH/36 When a locally submitted message by a trusted user did not contain a
3788 From: header, and the sender address was obtained from -f or from an SMTP
3789 MAIL command, and the trusted user did not use -F to supply a sender
3790 name, $originator_name was incorrectly used when constructing a From:
3791 header. Furthermore, $originator_name was used for submission mode
3792 messages from external hosts without From: headers in a similar way,
3793 which is clearly wrong.
3795 PH/37 Added control=suppress_local_fixups.
3797 PH/38 When log_selector = +received_sender was set, and the addition of the
3798 sender made the log line's construction buffer exactly full, or one byte
3799 less than full, an overflow happened when the terminating "\n" was
3802 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3803 when the result of a list match is failure because a DNS lookup failed.
3805 PH/40 RM_COMMAND is now used in the building process.
3807 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3808 the "build-* directories that it finds.
3810 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3811 address in a domain literal was a prefix of an interface address.
3813 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3814 when verifying a sender address, unless rewrite_headers is false.
3816 PH/44 Wrote a long comment about why errors_to addresses are verified as
3817 recipients, not senders.
3819 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3820 the ratelimit ACL was added.
3822 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3824 PH/46 Added extra information about PostgreSQL errors to the error string.
3826 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3827 avoiding the need to install special zones in a real server. This is
3828 backwards compatible; if it can't find the fake resolver, it drops back.
3829 Thus, both old and new test suites can be run.
3831 TF/02 Added util/ratelimit.pl
3833 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3834 clock is set back in time.
3836 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3837 Candler <B.Candler@pobox.com>.
3839 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3840 OK for addresses that are the result of redirections.
3842 PH/48 A number of further additions for the benefit of the new test suite,
3843 including a fake gethostbyname() that interfaces to the fake DNS resolver
3846 TF/06 The fix for widen_domains has also been applied to qualify_single and
3847 search_parents which are the other dnslookup options that can cause
3850 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3853 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3855 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3856 service" if it encountered an MX record with an empty target hostname.
3857 The message is now "an MX or SRV record indicated no SMTP service".
3859 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3860 if the DNS of the sending site is misconfigured. This is quite a
3861 common situation. This change restores the behaviour of treating a
3862 helo verification defer as a failure.
3864 PH/51 If self=fail was set on a router, the bounce message did not include the
3865 actual error message.
3871 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3873 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3874 command itself ended in error, the underlying message about the transport
3875 filter timeout was being overwritten with the pipe command error. Now the
3876 underlying error message should be appended to the second error message.
3878 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3880 PH/02 Reduce the amount of output that "make" produces by default. Full output
3881 can still be requested.
3883 PH/03 The warning log line about a condition test deferring for a "warn" verb
3884 was being output only once per connection, rather than after each
3885 occurrence (because it was using the same function as for successful
3886 "warn" verbs). This seems wrong, so I have changed it.
3888 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3889 it should not have, which might have caused a crash in the right
3890 circumstances, but probably never did.
3892 PH/04 Installed a modified version of Tony Finch's patch to make submission
3893 mode fix the return path as well as the Sender: header line, and to
3894 add a /name= option so that you can make the user's friendly name appear
3897 TF/03 Added the control = fakedefer ACL modifier.
3899 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3900 Mark Lowes for thorough testing.
3902 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3904 TK/03 Merged latest SRS patch from Miles Wilton.
3906 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3907 with the definition in sysexits.h (which is #included earlier).
3908 Fortunately, Exim does not actually use EX_OK. The code used to try to
3909 preserve the sysexits.h value, by assuming that macro definitions were
3910 scanned for macro replacements. I have been disabused of this notion,
3911 so now the code just undefines EX_OK before #including unistd.h.
3913 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3914 in the smtp transport. When a block could not be written in a single
3915 write() function, the timeout was being re-applied to each part-write.
3916 This seems wrong - if the receiver was accepting one byte at a time it
3917 would take for ever. The timeout is now adjusted when this happens. It
3918 doesn't have to be particularly precise.
3920 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3921 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3923 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3924 without a domain if the check to <postmaster@domain> fails.
3926 SC/01 Eximstats: added -xls and the ability to specify output files
3927 (patch written by Frank Heydlauf).
3929 SC/02 Eximstats: use FileHandles for outputting results.
3931 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3933 SC/04 Eximstats: fixed display of large numbers with -nvr option
3935 SC/05 Eximstats: fixed merging of reports with empty tables.
3937 SC/06 Eximstats: added the -include_original_destination flag
3939 SC/07 Eximstats: removed tabs and trailing whitespace.
3941 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3943 TK/06 MBOX spool code: Add real "From " MBOX separator line
3944 so the .eml file is really in mbox format (even though
3945 most programs do not really care). Patch from Alex Miller.
3947 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3948 The latter is generated from $received_to and is only set if the
3949 message has one envelope recipient. SA can use these headers,
3950 obviously out-of-the-box. Patch from Alex Miller.
3952 PH/08 The ${def test on a variable was returning false if the variable's
3953 value was "0", contrary to what the specification has always said!
3954 The result should be true unless the variable is empty.
3956 PH/09 The syntax error of a character other than { following "${if
3957 def:variable_name" (after optional whitespace) was not being diagnosed.
3958 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3959 accidental colon was present, for example, could give incorrect results.
3961 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3962 result is used (not including appendfile, where other changes are about
3965 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3966 This involved changing a lot of size variables from int to off_t. It
3967 should work with maildirs and everything.
3969 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3970 spamd dying while we are connected to it.
3972 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3975 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3976 function for BDB 4.3.
3978 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3980 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3981 the sources for PCRE have been re-arranged and more files are now
3984 PH/15 The code I had for printing potentially long long variables in PH/11
3985 above was not the best (it lost precision). The length of off_t variables
3986 is now inspected at build time, and an appropriate printing format (%ld
3987 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3988 to be "long long int" or "long int". This is needed for the internal
3989 formatting function string_vformat().
3991 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3992 the configuration file to be ":syslog", then the script "guesses" where
3993 the logs files are, rather than using the compiled in default. In our
3994 case the guess is not the same as the compiled default, so the script
3995 suddenly stopped working when I started to use syslog. The patch checks
3996 to see if log_file_path is "". If so, it attempts to read it from exim
3997 with no configuration file to get the compiled in version, before it
3998 falls back to the previous guessing code."
4000 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4001 implementing BATV in an Exim configuration. See NewStuff for the gory
4004 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4005 Makefile that are specific to HP-UX.
4007 PH/18 If the "use_postmaster" option was set for a recipient callout together
4008 with the "random" option, the postmaster address was used as the MAIL
4009 FROM address for the random test, but not for the subsequent recipient
4010 test. It is now used for both.
4012 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4013 patch removes a few documentation additions to RFC 3028, because the
4014 latest draft now contains them. It adds the new en;ascii-case comparator
4015 and a new error check for 8bit text in MIME parts. Comparator and
4016 require names are now matched exactly. I enabled the subaddress
4017 extension, but it is not well tested yet (read: it works for me)."
4019 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4020 rework some of the code of TK/09 above to avoid the hardwired use of
4021 "%lld" and "long long". Replaced the call to snprintf() with a call to
4024 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4025 records point to non-existent hosts", "retry timeout exceeded", and
4026 "retry time not reached for any host after a long failure period".
4028 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4029 experimental DomainKeys support:
4031 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4032 (2) On an error such as an illegally used "control", the wrong name for
4033 the control was given.
4035 These problems did NOT occur unless DomainKeys support was compiled.
4037 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4039 PH/24 Added ${if match_ip condition.
4041 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4042 databases so that it will be absolutely obvious if a crash occurs in the
4043 DB library. This is a regular occurrence (often caused by mis-matched
4046 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4047 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4048 picked up on a user's system that detects such things. There doesn't seem
4049 to be a gcc warning option for this - only an attribute that has to be
4050 put on the function's prototype. It seems that in Fedora Core 4 they have
4051 set this on a number of new functions. No doubt there will be more in due
4054 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4055 specify a transport. However, if an address that was verified by such a
4056 router was the subject of a callout, Exim crashed because it tried to
4057 read the rcpt_include_affixes from the non-existent transport. Now it
4058 just assumes that the setting of that option is false. This bug was
4059 introduced by 4.51/PH/31.
4061 PH/28 Changed -d+all to exclude +memory, because that information is very
4062 rarely of interest, but it makes the output a lot bigger. People tend to
4063 do -d+all out of habit.
4065 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4066 code in os-type was giving problems when libc.so lives in lib64, like on
4069 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4070 aren't the modern standard, and it seems that some systems' include files
4071 don't always have them. Exim was already checking for some of the newer
4072 ones like T_AAAA, and defining it itself. I've added checks for all the
4073 record types that Exim uses.
4075 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4076 not automatically generating a new one, as it is supposed to. This
4077 prevented TLS from working. If the file did exist, but contained invalid
4078 data, a new version was generated, as expected. It was only the case of a
4079 non-existent file that was broken.
4081 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4082 with a change in libdomainkeys > 0.64.
4084 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4085 from DNS. If the selector record carries the flag, it now has
4086 precedence over the domain-wide flag.
4088 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4090 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4091 the use of a port name that isn't defined in /etc/services) occurred, the
4092 message was deferred as in a normal delivery, and thus remained on the
4093 spool, instead of being failed because of the mua_wrapper setting. This
4094 is now fixed, and I tidied up some of the mua_wrapper messages at the
4097 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4098 the messages in a hash of arrays rather than using individual hashes.
4099 This is a bit cleaner and results in dramatic memory savings, albeit
4100 at a slight CPU cost.
4102 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4103 as requested by Marc Sherman.
4105 SC/10 Eximstats: added histograms for user specified patterns as requested
4108 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4110 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4111 fopen() in the content-scanning modules that did not already have it.
4117 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4118 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4120 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4122 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4124 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4125 to test Sieve filters that use "vacation".
4127 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4128 that changes the way the GnuTLS parameters are stored in the cache file.
4129 The new format can be generated externally. For backward compatibility,
4130 if the data in the cache doesn't make sense, Exim assumes it has read an
4131 old-format file, and it generates new data and writes a new file. This
4132 means that you can't go back to an older release without removing the
4135 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4136 work if there are any delivery delays because "one_time" forces the
4137 parent to be marked "delivered", so its unseen clone is never tried
4138 again. For this reason, Exim now forbids the simultaneous setting of
4141 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4142 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4143 entry if you want to know the details. The fix, however, neglected to
4144 consider the case when local delivery batching is involved. The test for
4145 "previously delivered" was not happening when checking to see if an
4146 address could be batched with a previous (undelivered) one; under
4147 certain circumstances this could lead to multiple deliveries to the same
4150 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4151 in its include files, and this causes problems building Exim.
4153 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4154 header_syntax) but Exim was just ignoring anything given after a slash.
4155 In particular, this caused confusion with an attempt to use "verify =
4156 reverse_host_lookup/defer_ok". An error is now given when options are
4157 supplied for verify items that do not have them. (Maybe reverse_host_
4158 lookup should have a defer_ok option, but that's a different point.)
4160 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4161 defined by RFC 821) to 2048, because there were problems with some AUTH
4162 commands, and RFC 1869 says the size should be increased for extended
4163 SMTP commands that take arguments.
4165 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4168 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4169 "unknown" error; now it says that the functionality isn't in the binary.
4171 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4172 an address' error message when a string expansion fails (syntax or
4173 whatever). Otherwise the password may appear in the log. Following change
4174 PH/42 below, there is no longer a chance of it appearing in a bounce
4177 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4179 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4180 of its addresses was ever tried. (Bugzilla bug #2.)
4182 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4183 the result incorrectly in the debug output. (It correctly added a newline
4184 to what was transported.)
4186 TF/01 Added $received_time.
4188 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4189 commented out examples of how to interface to a virus scanner and to
4190 SpamAssassin. Also added commented examples of av_scanner and
4191 spamd_address settings.
4193 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4194 and controls are allowed in which ACLs. There were a couple of minor
4195 errors. Some of the entries in the conditions table (which is a table of
4196 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4197 negation of where the condition IS allowed.
4199 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4201 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4202 header file does not have a version number, so I've had to invent a new
4203 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4204 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4205 radiusclient), but it was contributed by a Radius user.
4207 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4208 files or directories, for OpenSSL.
4210 PH/19 When an Exim process that is running as root has to create an Exim log
4211 file, it does so in a subprocess that runs as exim:exim so as to get the
4212 ownership right at creation (otherwise, other Exim processes might see
4213 the file with the wrong ownership). There was no test for failure of this
4214 fork() call, which would lead to the process getting stuck as it waited
4215 for a non-existent subprocess. Forks do occasionally fail when resources
4216 run out. I reviewed all the other calls to fork(); they all seem to check
4219 PH/20 When checking for unexpected SMTP input at connect time (before writing
4220 the banner), Exim was not dealing correctly with a non-positive return
4221 from the read() function. If the client had disconnected by this time,
4222 the result was a log entry for a synchronization error with an empty
4223 string after "input=" when read() returned zero. If read() returned -1
4224 (an event I could not check), uninitialized data bytes were printed.
4225 There were reports of junk text (parts of files, etc) appearing after
4228 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4230 PH/22 Added support for macro redefinition, and (re)definition in between
4231 driver and ACL definitions.
4233 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4234 forgetting to use the resulting value; it was using the unexpanded value.
4236 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4237 hadn't been configured. The fix is from Juergen Kreileder, who
4238 understands it better than I do:
4240 "Here's what I see happening with three configured cyrus_sasl
4241 authenticators configured (plain, login, cram-md5):
4243 On startup auth_cyrus_sasl_init() gets called for each of these.
4244 This means three calls to sasl_listmech() without a specified mech_list.
4245 => SASL tests which mechs of all available mechs actually work
4246 => three warnings about OTP not working
4247 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4249 With the patch, sasl_listmech() also gets called three times. But now
4250 SASL's mech_list option is set to the server_mech specified in the the
4251 authenticator. Or in other words, the answer from sasl_listmech()
4252 gets limited to just the mech you're testing for (which is different
4254 => the return list contains just 'plain' or 'login', 'cram-md5' or
4255 nothing depending on the value of ob->server_mech.
4257 I've just tested the patch: Authentication still works fine,
4258 unavailable mechs specified in the exim configuration are still
4259 caught, and the auth.log warnings about OTP are gone."
4261 PH/25 When debugging is enabled, the contents of the command line are added
4262 to the debugging output, even when log_selector=+arguments is not
4265 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4266 answer is "GNU", and only if the return is "GNU/something" is the answer
4269 PH/27 $acl_verify_message is now set immediately after the failure of a
4270 verification in an ACL, and so is available in subsequent modifiers. In
4271 particular, the message can be preserved by coding like this:
4273 warn !verify = sender
4274 set acl_m0 = $acl_verify_message
4276 Previously, $acl_verify_message was set only while expanding "message"
4277 and "log_message" when a very denied access.
4279 PH/28 Modified OS/os.c-Linux with
4281 -#ifndef OS_LOAD_AVERAGE
4282 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4284 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4285 nomenclature these days.)
4287 PH/29 Installed patch from the Sieve maintainer that adds the options
4288 sieve_useraddress and sieve_subaddress to the redirect router.
4290 PH/30 In these circumstances:
4291 . Two addresses routed to the same list of hosts;
4292 . First host does not offer TLS;
4293 . First host accepts first address;
4294 . First host gives temporary error to second address;
4295 . Second host offers TLS and a TLS session is established;
4296 . Second host accepts second address.
4297 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4298 and peerdn, if requested) that were in fact used only for the second
4301 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4302 attention to any local part prefix or suffix that was matched by the
4303 router that accepted the address. It now behaves in the same way as it
4304 does for delivery: the affixes are removed from the local part unless
4305 rcpt_include_affixes is set on the transport.
4307 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4308 timeout during the DATA phase of an incoming message.
4310 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4311 applied a patch sanctioned by the Sieve maintainer.
4313 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4314 the uid or gid is negative. A case of a negative gid caused this to be
4315 noticed. The fix allows for either to be negative.
4317 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4318 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4321 PH/36 The change PH/12 above was broken. Fixed it.
4323 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4324 the grounds that routing the same address twice would always produce the
4325 same answer. This might have been true once, but it is certainly no
4326 longer true now. Routing a child address may depend on the previous
4327 routing that produced that child. Some complicated redirection strategies
4328 went wrong when messages had multiple recipients, and made Exim's
4329 behaviour dependent on the order in which the addresses were given.
4331 I have moved the duplicate checking until after the routing is complete.
4332 Exim scans the addresses that are assigned to local and remote
4333 transports, and removes any duplicates. This means that more work will be
4334 done, as duplicates will always all be routed, but duplicates are
4335 presumably rare, so I don't expect this is of any significance.
4337 For deliveries to pipes, files, and autoreplies, the duplicate checking
4338 still happens during the routing process, since they are not going to be
4341 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4342 It corrects a timeout issue with spamd. This is Ian's comment: "The
4343 background is that sometimes spamd either never reads data from a
4344 connection it has accepted, or it never writes response data. The exiscan
4345 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4346 blindly assumes that writes won't block so it may never time out."
4348 PH/39 Allow G after quota size as well as K and M.
4350 PH/40 The value set for $authenticated_id in an authenticator may not contain
4351 binary zeroes or newlines because the value is written to log lines and
4352 to spool files. There was no check on this. Now the value is run through
4353 the string_printing() function so that such characters are converted to
4354 printable escape sequences.
4356 PH/41 $message_linecount is a new variable that contains the total number of
4357 lines in the message. Compare $body_linecount, which is the count for the
4360 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4361 bounce and delay warning messages, except in certain special cases, which
4364 (a) An SMTP error message from a remote host;
4365 (b) A message specified in a :fail: redirection;
4366 (c) A message specified in a "fail" command in a system filter;
4367 (d) A message specified in a FAIL return from the queryprogram router;
4368 (e) A message specified by the cannot_route_message router option.
4370 In these cases only, Exim does include the error details in bounce and
4371 warning messages. There are also a few cases where bland messages such
4372 as "unrouteable address" or "local delivery error" are given.
4374 PH/43 $value is now also set for the "else" part of a ${run expansion.
4376 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4377 being worked on, but at least Exim now implements the latest version to
4380 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4381 process to complete was treated as a delivery failure, a timeout while
4382 writing the message to the pipe was logged, but erroneously treated as a
4383 successful delivery. Such timeouts include transport filter timeouts. For
4384 consistency with the overall process timeout, these timeouts are now
4385 treated as errors, giving rise to delivery failures by default. However,
4386 there is now a new Boolean option for the pipe transport called
4387 timeout_defer, which, if set TRUE, converts the failures into defers for
4388 both kinds of timeout. A transport filter timeout is now identified in
4391 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4392 systems where "make" and "gmake" are different, calling "gmake" at top
4393 level broke things. I've arranged for the value of $(MAKE) to be passed
4394 from the Makefile to this script so that it can call the same version of
4398 A note about Exim versions 4.44 and 4.50
4399 ----------------------------------------
4401 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4402 changes of various kinds. As a consequence, a big documentation update was
4403 needed. This delayed the release for rather longer than seemed good, especially
4404 in the light of a couple of (minor) security issues. Therefore, the changes
4405 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4406 release. So 4.44 and 4.50 are in effect two different branches that both start
4409 I have left the 4.50 change log unchanged; it contains all the changes since
4410 4.43. The change log for 4.44 is below; many of its items are identical to
4411 those for 4.50. This seems to be the most sensible way to preserve the
4412 historical information.
4418 1. Minor wording change to the doc/README.SIEVE file.
4420 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4421 computation of the current number of files was incorrect.
4423 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4424 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4427 4. Give more explanation in the error message when the command for a transport
4428 filter fails to execute.
4430 5. There are several places where Exim runs a non-Exim command in a
4431 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4432 was being done only for the command run by the queryprogram router. It is
4433 now done for all such subprocesses. The other cases are: ${run, transport
4434 filters, and the commands run by the lmtp and pipe transports.
4436 6. Added CONFIGURE_GROUP build-time option.
4438 7. Some older OS have a limit of 256 on the maximum number of file
4439 descriptors. Exim was using setrlimit() to set 1000 as a large value
4440 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4441 systems. I've change it so that if it can't get 1000, it tries for 256.
4443 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4444 was an oversight, and furthermore, ever since the addition of extra
4445 controls (e.g. 4.43/32), the checks on when to allow different forms of
4446 "control" were broken. There should now be diagnostics for all cases when a
4447 control that does not make sense is encountered.
4449 9. Added the /retain_sender option to "control=submission".
4451 10. $recipients is now available in the predata ACL (oversight).
4453 11. Tidy the search cache before the fork to do a delivery from a message
4454 received from the command line. Otherwise the child will trigger a lookup
4455 failure and thereby defer the delivery if it tries to use (for example) a
4456 cached ldap connection that the parent has called unbind on.
4458 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4459 of $address_data from the recipient verification was clobbered by the
4460 sender verification.
4462 13. The value of address_data from a sender verification is now available in
4463 $sender_address_data in subsequent conditions in the ACL statement.
4465 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4467 15. Added a new option "connect=<time>" to callout options, to set a different
4470 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4471 was its contents. (It was OK if the option was not defined at all.)
4473 17. A "Completed" log line is now written for messages that are removed from
4474 the spool by the -Mrm option.
4476 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4477 information about exactly what failed.
4479 19. Added -dd to debug only the daemon process.
4481 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4482 handles timeouts, both on the server side and network timeouts. Renamed the
4483 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4485 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4486 It is now set to "smtps".
4488 22. $host_address is now set to the target address during the checking of
4489 ignore_target_hosts.
4491 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4492 being passed; this would have caused $sender_host_name to have been used if
4493 matching the list had actually called for a host name (not very likely,
4494 since this list is usually IP addresses). A host name is now passed as
4497 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4498 code that specifies a non-restarting handler (typically sigaction() in
4499 modern systems) in an attempt to fix a rare and obscure crash bug.
4501 25. Narrowed the window for a race in the daemon that could cause it to ignore
4502 SIGCHLD signals. This is not a major problem, because they are used only to
4503 wake it up if nothing else does.
4505 26. A malformed maildirsize file could cause Exim to calculate negative values
4506 for the mailbox size or file count. Odd effects could occur as a result.
4507 The maildirsize information is now recalculated if the size or filecount
4510 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4511 support for a long time. Removed HAVE_SYS_VFS_H.
4513 28. Installed the latest version of exipick from John Jetmore.
4515 29. In an address list, if the pattern was not a regular expression, an empty
4516 subject address (from a bounce message) matched only if the pattern was an
4517 empty string. Non-empty patterns were not even tested. This was the wrong
4518 because it is perfectly reasonable to use an empty address as part of a
4519 database query. An empty address is now tested by patterns that are
4520 lookups. However, all the other forms of pattern expect the subject to
4521 contain a local part and a domain, and therefore, for them, an empty
4522 address still always fails if the pattern is not itself empty.
4524 30. Exim went into a mad DNS loop when attempting to do a callout where the
4525 host was specified on an smtp transport, and looking it up yielded more
4526 than one IP address.
4528 31. Re-factored the code for checking spool and log partition space into a
4529 function that finds that data and another that does the check. The former
4530 is then used to implement four new variables: $spool_space, $log_space,
4531 $spool_inodes, and $log_inodes.
4533 32. The RFC2047 encoding function was originally intended for short strings
4534 such as real names; it was not keeping to the 75-character limit for
4535 encoded words that the RFC imposes. It now respects the limit, and
4536 generates multiple encoded words if necessary. To be on the safe side, I
4537 have increased the buffer size for the ${rfc2047: expansion operator from
4540 33. It is now permitted to omit both strings after an "if" condition; if the
4541 condition is true, the result is "true". As before, when the second string
4542 is omitted, a false condition yields an empty string. This makes it less
4543 cumbersome to write custom ACL and router conditions.
4545 34. Failure to deliver a bounce message always caused it to be frozen, even if
4546 there was an errors_to setting on the router. The errors_to setting is now
4549 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4550 canonical form (fully expanded) before being placed in
4551 $sender_host_address.
4553 36. The table in the code that translates DNS record types into text (T_A to
4554 "A" for instance) was missing entries for NS and CNAME. It is just possible
4555 that this could have caused confusion if both these types were looked up
4556 for the same domain, because the text type is used as part of Exim's
4557 per-process caching. But the chance of anyone hitting this buglet seems
4560 37. The dnsdb lookup has been extended in a number of ways.
4562 (1) There is a new type, "zns", which walks up the domain tree until it
4563 finds some nameserver records. It should be used with care.
4565 (2) There is a new type, "mxh", which is like "mx" except that it returns
4566 just the host names, not the priorities.
4568 (3) It is now possible to give a list of domains (or IP addresses) to be
4569 looked up. The behaviour when one of the lookups defers can be
4570 controlled by a keyword.
4572 (4) It is now possible to specify the separator character for use when
4573 multiple records are returned.
4575 38. The dnslists ACL condition has been extended: it is now possible to supply
4576 a list of IP addresses and/or domains to be looked up in a particular DNS
4579 39. Added log_selector=+queue_time_overall.
4581 40. When running the queue in the test harness, wait just a tad after forking a
4582 delivery process, to get repeatability of debugging output.
4584 41. Include certificate and key file names in error message when GnuTLS fails
4585 to set them up, because the GnuTLS error message doesn't include the name
4586 of the failing file when there is a problem reading it.
4588 42. Allow both -bf and -bF in the same test run.
4590 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4592 44. The "Exiscan patch" is now merged into the mainline Exim source.
4594 45. Sometimes the final signoff response after QUIT could fail to get
4595 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4596 < 0 before doing a fflush(). This bug looks as though it goes back to the
4597 introduction of TLS in release 3.20, but "sometimes" must have been rare
4598 because the tests only now provoked it.
4600 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4601 (this can affect the format of dates).
4603 47. exim_tidydb, when checking for the continued existence of a message for
4604 which it has found a message-specific retry record, was not finding
4605 messages that were in split spool directories. Consequently, it was
4606 deleting retry records that should have stayed in existence.
4608 48. Steve fixed some bugs in eximstats.
4610 49. The SPA authentication driver was not abandoning authentication and moving
4611 on to the next authenticator when an expansion was forced to fail,
4612 contradicting the general specification for all authenticators. Instead it
4613 was generating a temporary error. It now behaves as specified.
4615 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4616 (the order specifies the preference for clients). The order is now AES256,
4617 AES128, 3DES, ARCFOUR128.
4619 51. Small patch to Sieve code - explicitly set From: when generating an
4622 52. Exim crashed if a remote delivery caused a very long error message to be
4623 recorded - for instance if somebody sent an entire SpamAssassin report back
4624 as a large number of 550 error lines. This bug was coincidentally fixed by
4625 increasing the size of one of Exim's internal buffers (big_buffer) that
4626 happened as part of the Exiscan merge. However, to be on the safe side, I
4627 have made the code more robust (and fixed the comments that describe what
4630 53. Now that there can be additional text after "Completed" in log lines (if
4631 the queue_time_overall log selector is set), a one-byte patch to exigrep
4632 was needed to allow it to recognize "Completed" as not the last thing in
4635 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4636 patch that reportedly fixes this has been added. I am not expert enough to
4637 create a test for it. This is what the patch creator wrote:
4639 "I found a little strange behaviour of ldap code when working with
4640 Windows 2003 AD Domain, where users was placed in more than one
4641 Organization Units. When I tried to give exim partial DN, the exit code
4642 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4643 But simultaneously result of request was absolutely normal ldap result,
4644 so I produce this patch..."
4646 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4647 REFERENCE, so I have modified the code to exclude the patch when that macro
4650 55. Some experimental protocols are using DNS PTR records for new purposes. The
4651 keys for these records are domain names, not reversed IP addresses. The
4652 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4653 leaves it alone. Component reversal etc. now happens only for IP addresses.
4656 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4658 57. Double the size of the debug message buffer (to 2048) so that more of very
4659 long debug lines gets shown.
4661 58. The exicyclog utility now does better if the number of log files to keep
4662 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4664 59. Two changes related to the smtp_active_hostname option:
4666 (1) $smtp_active_hostname is now available as a variable.
4667 (2) The default for smtp_banner uses $smtp_active_hostname instead
4668 of $primary_hostname.
4670 60. The host_aton() function is supposed to be passed a string that is known
4671 to be a valid IP address. However, in the case of IPv6 addresses, it was
4672 not checking this. This is a hostage to fortune. Exim now panics and dies
4673 if the condition is not met. A case was found where this could be provoked
4674 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4675 components; fortuitously, this particular loophole had already been fixed
4676 by change 4.50/55 above.
4678 If there are any other similar loopholes, the new check in host_aton()
4679 itself should stop them being exploited. The report I received stated that
4680 data on the command line could provoke the exploit when Exim was running as
4681 exim, but did not say which command line option was involved. All I could
4682 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4683 running as the user.
4686 61. There was a buffer overflow vulnerability in the SPA authentication code
4687 (which came originally from the Samba project). I have added a test to the
4688 spa_base64_to_bits() function which I hope fixes it.
4691 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4692 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4694 63. The daemon start-up calls getloadavg() while still root for those OS that
4695 need the first call to be done as root, but it missed one case: when
4696 deliver_queue_load_max is set with deliver_drop_privilege. This is
4697 necessary for the benefit of the queue runner, because there is no re-exec
4698 when deliver_drop_privilege is set.
4700 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4701 This has been fixed.
4703 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4704 was in use, was not putting the data itself into the right store pool;
4705 consequently, it could be overwritten for a subsequent message in the same
4706 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4709 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4711 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4712 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4713 the function were treating the return as a boolean value, which happened to
4714 work because 0=false and not-0=true, but is not correct code.
4716 68. The host_aton() function was not handling scoped IPv6 addresses (those
4717 with, for example, "%eth0" on the end) correctly.
4719 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4720 negated items (that is, ~something) in unsigned ints. Some compilers
4721 apparently mutter when there is no cast.
4723 70. If an address verification called from an ACL failed, and did not produce a
4724 user-specific message (i.e. there was only a "system" message), nothing was
4725 put in $acl_verify_message. In this situation, it now puts the system
4728 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4729 session; change 4.31/43 added the unwanted input to the log line - except
4730 that it did not do this in the start of session case. It now does.
4732 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4733 This is wrong and can cause the other end to generate a synchronization
4734 error if it is another Exim or anything else that does the synchronization
4735 check. A QUIT command is no longer sent after a timeout.
4737 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4738 during host lookups.
4740 74. The defer_ok option of callout verification was not working if it was used
4741 when verifying addresses in header lines, that is, for this case:
4743 verify = header_sender/callout=defer_ok
4745 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4746 those file descriptors could be used for SMTP connections. If anything
4747 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4748 could be sent to the SMTP client, causing chaos. The daemon now opens
4749 stdin, stdout, and stderr to /dev/null when it puts itself into the
4752 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4753 log by default. The user can override this with suitable Perl magic.
4755 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4756 the log message when discard triggers, was not working for the DATA ACL or
4757 for the non-SMTP ACL.
4759 78. Error message wording change in sieve.c.
4761 79. If smtp_accept_max_per_host was set, the number of connections could be
4762 restricted to fewer than expected, because the daemon was trying to set up
4763 a new connection before checking whether the processes handling previous
4764 connections had finished. The check for completed processes is now done
4765 earlier. On busy systems, this bug wouldn't be noticed because something
4766 else would have woken the daemon, and it would have reaped the completed
4769 80. If a message was submitted locally by a user whose login name contained one
4770 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4771 It caused a spool format error. I have fixed the spool reading code. A
4772 related problem was that the "from" clause in the Received: line became
4773 illegal because of the space(s). It is now covered by ${quote_local_part.
4775 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4778 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4780 83. Patch from Sieve maintainer to fix unterminated string problem in
4781 "vacation" handling.
4783 84. Some minor changes to the Linux configuration files to help with other
4784 OS variants using glibc.
4786 85. One more patch for Sieve to update vacation handling to latest spec.
4789 ----------------------------------------------------
4790 See the note above about the 4.44 and 4.50 releases.
4791 ----------------------------------------------------
4797 1. Change 4.43/35 introduced a bug that caused file counts to be
4798 incorrectly computed when quota_filecount was set in an appendfile
4801 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4802 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4805 3. Give more explanation in the error message when the command for a transport
4806 filter fails to execute.
4808 4. There are several places where Exim runs a non-Exim command in a
4809 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4810 was being done only for the command run by the queryprogram router. It is
4811 now done for all such subprocesses. The other cases are: ${run, transport
4812 filters, and the commands run by the lmtp and pipe transports.
4814 5. Some older OS have a limit of 256 on the maximum number of file
4815 descriptors. Exim was using setrlimit() to set 1000 as a large value
4816 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4817 systems. I've change it so that if it can't get 1000, it tries for 256.
4819 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4820 was an oversight, and furthermore, ever since the addition of extra
4821 controls (e.g. 4.43/32), the checks on when to allow different forms of
4822 "control" were broken. There should now be diagnostics for all cases when a
4823 control that does not make sense is encountered.
4825 7. $recipients is now available in the predata ACL (oversight).
4827 8. Tidy the search cache before the fork to do a delivery from a message
4828 received from the command line. Otherwise the child will trigger a lookup
4829 failure and thereby defer the delivery if it tries to use (for example) a
4830 cached ldap connection that the parent has called unbind on.
4832 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4833 of $address_data from the recipient verification was clobbered by the
4834 sender verification.
4836 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4837 was its contents. (It was OK if the option was not defined at all.)
4839 11. A "Completed" log line is now written for messages that are removed from
4840 the spool by the -Mrm option.
4842 12. $host_address is now set to the target address during the checking of
4843 ignore_target_hosts.
4845 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4846 being passed; this would have caused $sender_host_name to have been used if
4847 matching the list had actually called for a host name (not very likely,
4848 since this list is usually IP addresses). A host name is now passed as
4851 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4852 code that specifies a non-restarting handler (typically sigaction() in
4853 modern systems) in an attempt to fix a rare and obscure crash bug.
4855 15. Narrowed the window for a race in the daemon that could cause it to ignore
4856 SIGCHLD signals. This is not a major problem, because they are used only to
4857 wake it up if nothing else does.
4859 16. A malformed maildirsize file could cause Exim to calculate negative values
4860 for the mailbox size or file count. Odd effects could occur as a result.
4861 The maildirsize information is now recalculated if the size or filecount
4864 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4865 support for a long time. Removed HAVE_SYS_VFS_H.
4867 18. Updated exipick to current release from John Jetmore.
4869 19. Allow an empty sender to be matched against a lookup in an address list.
4870 Previously the only cases considered were a regular expression, or an
4873 20. Exim went into a mad DNS lookup loop when doing a callout where the
4874 host was specified on the transport, if the DNS lookup yielded more than
4877 21. The RFC2047 encoding function was originally intended for short strings
4878 such as real names; it was not keeping to the 75-character limit for
4879 encoded words that the RFC imposes. It now respects the limit, and
4880 generates multiple encoded words if necessary. To be on the safe side, I
4881 have increased the buffer size for the ${rfc2047: expansion operator from
4884 22. Failure to deliver a bounce message always caused it to be frozen, even if
4885 there was an errors_to setting on the router. The errors_to setting is now
4888 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4889 canonical form (fully expanded) before being placed in
4890 $sender_host_address.
4892 24. Updated eximstats to version 1.33
4894 25. Include certificate and key file names in error message when GnuTLS fails
4895 to set them up, because the GnuTLS error message doesn't include the name
4896 of the failing file when there is a problem reading it.
4898 26. Expand error message when OpenSSL has problems setting up cert/key files.
4901 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4902 (this can affect the format of dates).
4904 28. exim_tidydb, when checking for the continued existence of a message for
4905 which it has found a message-specific retry record, was not finding
4906 messages that were in split spool directories. Consequently, it was
4907 deleting retry records that should have stayed in existence.
4909 29. eximstats updated to version 1.35
4910 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4911 1.35 - bugfix such that pie charts by volume are generated correctly
4913 30. The SPA authentication driver was not abandoning authentication and moving
4914 on to the next authenticator when an expansion was forced to fail,
4915 contradicting the general specification for all authenticators. Instead it
4916 was generating a temporary error. It now behaves as specified.
4918 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4919 (the order specifies the preference for clients). The order is now AES256,
4920 AES128, 3DES, ARCFOUR128.
4922 31. Small patch to Sieve code - explicitly set From: when generating an
4925 32. Exim crashed if a remote delivery caused a very long error message to be
4926 recorded - for instance if somebody sent an entire SpamAssassin report back
4927 as a large number of 550 error lines. This bug was coincidentally fixed by
4928 increasing the size of one of Exim's internal buffers (big_buffer) that
4929 happened as part of the Exiscan merge. However, to be on the safe side, I
4930 have made the code more robust (and fixed the comments that describe what
4933 33. Some experimental protocols are using DNS PTR records for new purposes. The
4934 keys for these records are domain names, not reversed IP addresses. The
4935 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4936 leaves it alone. Component reversal etc. now happens only for IP addresses.
4939 34. The host_aton() function is supposed to be passed a string that is known
4940 to be a valid IP address. However, in the case of IPv6 addresses, it was
4941 not checking this. This is a hostage to fortune. Exim now panics and dies
4942 if the condition is not met. A case was found where this could be provoked
4943 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4944 components; fortuitously, this particular loophole had already been fixed
4945 by change 4.50/55 or 4.44/33 above.
4947 If there are any other similar loopholes, the new check in host_aton()
4948 itself should stop them being exploited. The report I received stated that
4949 data on the command line could provoke the exploit when Exim was running as
4950 exim, but did not say which command line option was involved. All I could
4951 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4952 running as the user.
4955 35. There was a buffer overflow vulnerability in the SPA authentication code
4956 (which came originally from the Samba project). I have added a test to the
4957 spa_base64_to_bits() function which I hope fixes it.
4960 36. The daemon start-up calls getloadavg() while still root for those OS that
4961 need the first call to be done as root, but it missed one case: when
4962 deliver_queue_load_max is set with deliver_drop_privilege. This is
4963 necessary for the benefit of the queue runner, because there is no re-exec
4964 when deliver_drop_privilege is set.
4966 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4967 was in use, was not putting the data itself into the right store pool;
4968 consequently, it could be overwritten for a subsequent message in the same
4969 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4972 38. Sometimes the final signoff response after QUIT could fail to get
4973 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4974 < 0 before doing a fflush(). This bug looks as though it goes back to the
4975 introduction of TLS in release 3.20, but "sometimes" must have been rare
4976 because the tests only now provoked it.
4982 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4983 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4984 Now it returns an integer. A number of calls were still expecting a T/F
4985 return. Fortuitously, in all cases, the tests worked in OK situations,
4986 which is the norm. However, things would have gone wrong on any write
4987 failures on the smtp file descriptor. This function is used when sending
4988 messages over SMTP and also when doing verify callouts.
4990 2. When Exim is called to do synchronous delivery of a locally submitted
4991 message (the -odf or -odi options), it no longer closes stderr before doing
4994 3. Implemented the mua_wrapper option.
4996 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4998 5. Implemented the functions header_remove(), header_testname(),
4999 header_add_at_position(), and receive_remove_recipient(), and exported them
5002 6. If an ACL "warn" statement specified the addition of headers, Exim already
5003 inserted X-ACL-Warn: at the start if there was no header name. However, it
5004 was not making this test for the second and subsequent header lines if
5005 there were newlines in the string. This meant that an invalid header could
5006 be inserted if Exim was badly configured.
5008 7. Allow an ACL "warn" statement to add header lines at the start or after all
5009 the Received: headers, as well as at the end.
5011 8. Added the rcpt_4xx retry error code.
5013 9. Added postmaster_mailfrom=xxx to callout verification option.
5015 10. Added mailfrom=xxxx to the callout verification option, for verify=
5018 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5019 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5021 12. Inserted some casts to stop certain compilers warning when using pointer
5022 differences as field lengths or precisions in printf-type calls (mostly
5023 affecting debugging statements).
5025 13. Added optional readline() support for -be (dynamically loaded).
5027 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5028 same clock tick as a message's arrival, so that its received time was the
5029 same as the "first fail" time on the retry record, and that message
5030 remained on the queue past the ultimate address timeout, every queue runner
5031 would try a delivery (because it was past the ultimate address timeout) but
5032 after another failure, the ultimate address timeout, which should have then
5033 bounced the address, did not kick in. This was a "< instead of <=" error;
5034 in most cases the first failure would have been in the next clock tick
5035 after the received time, and all would be well.
5037 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5038 being recognized when the domain list was tested by the match_domain
5039 condition in an expansion string.
5041 16. Added the ${str2b64: operator.
5043 17. Exim was always calling setrlimit() to set a large limit for the number of
5044 processes, without checking whether the existing limit was already
5045 adequate. (It did check for the limit on file descriptors.) Furthermore,
5046 errors from getrlimit() and setrlimit() were being ignored. Now they are
5047 logged to the main and panic logs, but Exim does carry on, to try to do its
5048 job under whatever limits there are.
5050 18. Imported PCRE 5.0.
5052 19. Trivial typo in log message " temporarily refused connection" (the leading
5055 20. If the log selector return_path_on_delivery was set and an address was
5056 redirected to /dev/null, the delivery process crashed because it assumed
5057 that a return path would always be set for a "successful" delivery. In this
5058 case, the whole delivery is bypassed as an optimization, and therefore no
5061 21. Internal re-arrangement: the function for sending a challenge and reading
5062 a response while authentication was assuming a zero-terminated challenge
5063 string. It's now changed to take a pointer and a length, to allow for
5064 binary data in such strings.
5066 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5068 23. Exim was not respecting finduser_retries when seeking the login of the
5069 uid under which it was called; it was always trying 10 times. (The default
5070 setting of finduser_retries is zero.) Also, it was sleeping after the final
5071 failure, which is pointless.
5073 24. Implemented tls_on_connect_ports.
5075 25. Implemented acl_smtp_predata.
5077 26. If the domain in control=submission is set empty, Exim assumes that the
5078 authenticated id is a complete email address when it generates From: or
5079 Sender: header lines.
5081 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5082 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5083 chown and chgrp in /bin and hostname in /usr/bin.
5085 28. Exim was keeping the "process log" file open after each use, just as it
5086 does for the main log. This opens the possibility of it remaining open for
5087 long periods when the USR1 signal hits a daemon. Occasional processlog
5088 errors were reported, that could have been caused by this. Anyway, it seems
5089 much more sensible not to leave this file open at all, so that is what now
5092 29. The long-running daemon process does not normally write to the log once it
5093 has entered its main loop, and it closes the log before doing so. This is
5094 so that log files can straightforwardly be renamed and moved. However,
5095 there are a couple of unusual error situations where the daemon does write
5096 log entries, and I had neglected to close the log afterwards.
5098 30. The text of an SMTP error response that was received during a remote
5099 delivery was being truncated at 512 bytes. This is too short for some of
5100 the long messages that one sometimes sees. I've increased the limit to
5103 31. It is now possible to make retry rules that apply only when a message has a
5104 specific sender, in particular, an empty sender.
5106 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5107 it possible to be selective about when SMTP synchronization is enforced.
5109 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5111 32. Implemented hosts_connection_nolog.
5113 33. Added an ACL for QUIT.
5115 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5118 35. Added mailbox_size and mailbox_filecount to appendfile.
5120 36. Added control = no_multiline_responses to ACLs.
5122 37. There was a bug in the logic of the code that waits for the clock to tick
5123 in the case where the clock went backwards by a substantial amount such
5124 that the microsecond fraction of "now" was more than the microsecond
5125 fraction of "then" (but the whole seconds number was less).
5127 38. Added support for the libradius Radius client library this is found on
5128 FreeBSD (previously only the radiusclient library was supported).
5134 1. When certain lookups returned multiple values in the form name=value, the
5135 quoting of the values was not always being done properly. Specifically:
5136 (a) If the value started with a double quote, but contained no whitespace,
5138 (b) If the value contained whitespace other than a space character (i.e.
5139 tabs or newlines or carriage returns) it was not quoted.
5140 This fix has been applied to the mysql and pgsql lookups by writing a
5141 separate quoting function and calling it from the lookup code. The fix
5142 should probably also be applied to nisplus, ibase and oracle lookups, but
5143 since I cannot test any of those, I have not disturbed their existing code.
5145 2. A hit in the callout cache for a specific address caused a log line with no
5146 reason for rejecting RCPT. Now it says "Previous (cached) callout
5147 verification failure".
5149 3. There was an off-by-one bug in the queryprogram router. An over-long
5150 return line was truncated at 256 instead of 255 characters, thereby
5151 overflowing its buffer with the terminating zero. As well as fixing this, I
5152 have increased the buffer size to 1024 (and made a note to document this).
5154 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5155 when Exim is waiting for an SMTP response from a remote server, Exim
5156 restarts its select() call on the socket, thereby resetting its timeout.
5157 This is not a problem when such interrupts are rare. Somebody set up a cron
5158 job to run exiwhat every 2 minutes, which is less than the normal select()
5159 timeout (5 or 10 minutes). This meant that the select() timeout never
5160 kicked in because it was always reset. I have fixed this by comparing the
5161 time when an interrupt arrives with the time at the start of the first call
5162 to select(). If more time than the timeout has elapsed, the interrupt is
5163 treated as a timeout.
5165 5. Some internal re-factoring in preparation for the addition of Sieve
5166 extensions (by MH). In particular, the "personal" test is moved to a
5167 separate function, and given an option for scanning Cc: and Bcc: (which is
5168 not set for Exim filters).
5170 6. When Exim created an email address using the login of the caller as the
5171 local part (e.g. when creating a From: or Sender: header line), it was not
5172 quoting the local part when it contained special characters such as @.
5174 7. Installed new OpenBSD configuration files.
5176 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5177 try to make them clearer.
5179 9. Callout options, other than the timeout value, were being ignored when
5180 verifying sender addresses in header lines. For example, when using
5182 verify = header_sender/callout=no_cache
5184 the cache was (incorrectly) being used.
5186 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5187 only the Cygwin environment.
5189 11. When return_path_on_delivery was set as a log selector, if different remote
5190 addresses in the same message used different return paths and parallel
5191 remote delivery occurred, the wrong values would sometimes be logged.
5192 (Whenever a remote delivery process finished, the return path value from
5193 the most recently started remote delivery process was logged.)
5195 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5196 lines when AUTH and/or TLS are in use. This is the "received protocol"
5197 field. Exim used to use "asmtp" for authenticated SMTP, without any
5198 indication (in the protocol name) for TLS use. Now it follows the RFC and
5199 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5200 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5201 names appear in log lines as well as in Received: header lines.
5203 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5204 extensions, and comparison tests, and to fix some bugs.
5206 14. Changes to the "personal" filter test:
5208 (1) The test was buggy in that it was just doing the equivalent of
5209 "contains" tests on header lines. For example, if a user's address was
5210 anne@some.where, the "personal" test would incorrectly be true for
5212 To: susanne@some.where
5214 This test is now done by extracting each address from the header in turn,
5215 and checking the entire address. Other tests that are part of "personal"
5216 are now done using regular expressions (for example, to check local parts
5217 of addresses in From: header lines).
5219 (2) The list of non-personal local parts in From: addresses has been
5220 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5221 taken from the Sieve specification recommendations.
5223 (3) If the message contains any header line starting with "List-" it is
5224 treated as non-personal.
5226 (4) The test for "circular" in the Subject: header line has been removed
5227 because it now seems ill-conceived.
5229 15. Minor typos in src/EDITME comments corrected.
5231 16. Installed latest exipick from John Jetmore.
5233 17. If headers_add on a router specified a text string that was too long for
5234 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5235 of string_sprintf() is now avoided.
5237 18. $message_body_size was not set (it was always zero) when running the DATA
5238 ACL and the local_scan() function.
5240 19. For the "mail" command in an Exim filter, no default was being set for
5241 the once_repeat time, causing a random time value to be used if "once" was
5242 specified. (If the value happened to be <= 0, no repeat happened.) The
5243 default is now 0s, meaning "never repeat". The "vacation" command was OK
5244 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5245 (I found it when inspecting the code).
5247 20. There is now an overall timeout for performing a callout verification. It
5248 defaults to 4 times the callout timeout, which applies to individual SMTP
5249 commands during the callout. The overall timeout applies when there is more
5250 than one host that can be tried. The timeout is checked before trying the
5251 next host. This prevents very long delays if there are a large number of
5252 hosts and all are timing out (e.g. when the network connections are timing
5253 out). The value of the overall timeout can be changed by specifying an
5254 additional sub-option for "callout", called "maxwait". For example:
5256 verify = sender/callout=5s,maxwait=20s
5258 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5259 to the end before writing, but this should make it even safer).
5261 22. Exim was forgetting that it had advertised PIPELINING for the second and
5262 subsequent messages on an SMTP connection. It was also not resetting its
5263 memory on STARTTLS and an internal HELO.
5265 23. When Exim logs an SMTP synchronization error within a session, it now
5266 records whether PIPELINING has been advertised or not.
5268 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5269 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5270 rather than long int.
5272 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5274 26. Added the never_mail option to autoreply.
5280 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5281 crash if the getsockname() call failed; this can happen if a connection is
5282 closed very soon after it is established. The problem was simply in the
5283 order in which certain operations were done, causing Exim to try to write
5284 to the SMTP stream before it had set up the file descriptor. The bug has
5285 been fixed by making things happen in the correct order.
5291 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5292 before the connection was closed, thus losing the rejection response.
5294 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5295 some early Solaris releases, but causes trouble in current releases where
5296 socklen_t is defined.
5298 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5301 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5304 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5305 that releasing the top of it at the end releases what was used for sub-
5306 expansions (unless the block got too big). However, discard this block if
5307 the first thing is a variable or header, so that we can use its block when
5308 it is dynamic (useful for very large $message_headers, for example).
5310 6. Lookups now cache *every* query, not just the most recent. A new, separate
5311 store pool is used for this. It can be recovered when all lookup caches are
5312 flushed. Lookups now release memory at the end of their result strings.
5313 This has involved some general refactoring of the lookup sources.
5315 7. Some code has been added to the store_xxx() functions to reduce the amount
5316 of flapping under certain conditions.
5318 8. log_incoming_interface used to affect only the <= reception log lines. Now
5319 it causes the local interface and port to be added to several more SMTP log
5320 lines, for example "SMTP connection from", and rejection lines.
5322 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5324 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5326 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5327 could be overwritten at the end of the current message (or the start of a
5328 new message if it was set in a HELO ACL). The value is now preserved for
5329 the duration of the SMTP connection.
5331 12. If a transport had a headers_rewrite setting, and a matching header line
5332 contained an unqualified address, that address was qualified, even if it
5333 did not match any rewriting rules. The underlying bug was that the values
5334 of the flags that permit the existence of unqualified sender and recipient
5335 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5336 non-local messages, and by -bnq for local messages) were not being
5337 preserved with the message after it was received.
5339 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5340 "next input=" as part of the text comprising the host identity instead of
5341 the correct text. The code was using the same buffer for two different
5342 strings. However, depending on which order the printing function evaluated
5343 its arguments, the bug did not always show up. Under Linux, for example, my
5344 test suite worked just fine.
5346 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5347 doesn't work with some older versions of Perl. It has been changed to "my",
5348 which in any case is probably the better facility to use.
5350 15. A really picky compiler found some instances of statements for creating
5351 error messages that either had too many or two few arguments for the format
5354 16. The size of the buffer for calls to the DNS resolver has been increased
5355 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5356 for addresses that have a lot of PTR records. This alleviates a problem; it
5357 does not fully solve it.
5359 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5360 buffer now truncates the list and logs the incident, which is the same
5361 action as happens when Exim is looking up a host name and its aliases.
5362 Previously in this situation something unpredictable would happen;
5363 sometimes it was "internal error: store_reset failed".
5365 18. If a server dropped the connection unexpectedly when an Exim client was
5366 using GnuTLS and trying to read a response, the client delivery process
5367 crashed while trying to generate an error log message.
5369 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5370 string, for example:
5372 warn message = H1: something\nH2: something
5374 the text was added as a single header line from Exim's point of view
5375 though it ended up OK in the delivered message. However, searching for the
5376 second and subsequent header lines using $h_h2: did not work. This has been
5377 fixed. Similarly, if a system filter added multiple headers in this way,
5378 the routers could not see them.
5380 20. Expanded the error message when iplsearch is called with an invalid key to
5381 suggest using net-iplsearch in a host list.
5383 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5384 ACLs are no longer actually imposed (and a message to that effect is
5387 22. If a "gecos" field in a passwd entry contained escaped characters, in
5388 particular, if it contained a \" sequence, Exim got it wrong when building
5389 a From: or a Sender: header from that name. A second bug also caused
5390 incorrect handling when an unquoted " was present following a character
5391 that needed quoting.
5393 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5394 was not being matched caselessly.
5396 24. Arranged for all hyphens in the exim.8 source to be escaped with
5399 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5400 properly. Recipient callout cache records were still being keyed to include
5401 the sender, even when use_sender was set false. This led to far more
5402 callouts that were necessary. The sender is no longer included in the key
5403 when use_sender is false.
5405 26. Added "control = submission" modifier to ACLs.
5407 27. Added the ${base62d: operator to decode base 62 numbers.
5409 28. dnsdb lookups can now access SRV records.
5411 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5412 the configuration file.
5414 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5415 (-v) mode. This makes the output for a verbose queue run more intelligible.
5417 31. Added a use_postmaster feature to recipient callouts.
5419 32. Added the $body_zerocount variable, containing the number of binary zero
5420 bytes in the message body.
5422 33. The time of last modification of the "new" subdirectory is now used as the
5423 "mailbox time last read" when there is a quota error for a maildir
5426 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5428 35. Added +ignore_unknown as a special item in host lists.
5430 36. Code for decoding IPv6 addresses in host lists is now included, even if
5431 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5432 address was recognized as an IP address, but was then not correctly decoded
5433 into binary, causing unexpected and incorrect effects when compared with
5440 1. Very minor rewording of debugging text in manualroute to say "list of
5441 hosts" instead of "hostlist".
5443 2. If verify=header_syntax was set, and a header line with an unqualified
5444 address (no domain) and a large number of spaces between the end of the
5445 name and the colon was received, the reception process suffered a buffer
5446 overflow, and (when I tested it) crashed. This was caused by some obsolete
5447 code that should have been removed. The fix is to remove it!
5449 3. When running in the test harness, delay a bit after writing a bounce
5450 message to get a bit more predictability in the log output.
5452 4. Added a call to search_tidyup() just before forking a reception process. In
5453 theory, someone could use a lookup in the expansion of smtp_accept_max_
5454 per_host which, without the tidyup, could leave open a database connection.
5456 5. Added the variables $recipient_data and $sender_data which get set from a
5457 lookup success in an ACL "recipients" or "senders" condition, or a router
5458 "senders" option, similar to $domain_data and $local_part_data.
5460 6. Moved the writing of debug_print from before to after the "senders" test
5463 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5464 problems for message scanning, either using a data ACL, or using
5465 local_scan() because the Received: header was not generated till after they
5466 were called (in order to set the time as the time of reception completion).
5467 I have revised the way this works. The header is now generated after the
5468 body is received, but before the ACL or local_scan() are called. After they
5469 are run, the timestamp in the header is updated.
5475 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5476 before starting a queue runner without re-exec. This happened only when
5477 deliver_drop_privilege was set or when the Exim user was set to root. The
5478 effect of the bug was that timeouts during subsequent deliveries caused
5479 crashes instead of being properly handled. The handler is now left at its
5480 default (and expected) setting.
5482 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5483 message, again when deliver_drop_privilege is set or Exim is run as root.
5484 The bug described in (1) was not present in this case, but the tidying up
5485 of the other signals was missing. I have made the two cases consistent.
5487 3. The ignore_target_hosts setting on a manualroute router was being ignored
5488 for hosts that were looked up using the /MX notation.
5490 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5493 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5494 operated on the sender address. After changing the $sender_address to <>
5495 for the sender address verify, Exim was re-instated it as the original
5496 (before rewriting) address, but remembering that it had rewritten it, so it
5497 wasn't rewriting it again. This bug also had the effect of breaking the
5498 sender address verification caching when the sender address was rewritten.
5500 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5501 This has been changed so that if the ip literal address matches
5502 ignore_target_hosts, the router declines.
5504 7. Added expansion conditions match_domain, match_address, and match_local_
5505 part (NOT match_host).
5507 8. The placeholder for the Received: header didn't have a length field set.
5509 9. Added code to Exim itself and to exim_lock to test for a specific race
5510 condition that could lead to file corruption when using MBX delivery. The
5511 issue is with the lockfile that is created in /tmp. If this file is removed
5512 after a process has opened it but before that process has acquired a lock,
5513 there is the potential for a second process to recreate the file and also
5514 acquire a lock. This could lead to two Exim processes writing to the file
5515 at the same time. The added code performs the same test as UW imapd; it
5516 checks after acquiring the lock that its file descriptor still refers to
5517 the same named file.
5519 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5520 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5523 11. Added the smtp_active_hostname option. If used, this will typically be made
5524 to depend on the incoming interface address. Because $interface_address is
5525 not set up until the daemon has forked a reception process, error responses
5526 that can happen earlier (such as "too many connections") no longer contain
5529 12. If an expansion in a condition on a "warn" statement fails because a lookup
5530 defers, the "warn" statement is abandoned, and the next ACL statement is
5531 processed. Previously this caused the whole ACL to be aborted.
5533 13. Added the iplsearch lookup type.
5535 14. Added ident_timeout as a log selector.
5537 15. Added tls_certificate_verified as a log selector.
5539 16. Added a global option tls_require_ciphers (compare the smtp transport
5540 option of the same name). This controls incoming TLS connections.
5542 17. I finally figured out how to make tls_require_ciphers do a similar thing
5543 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5544 before starting the TLS session.
5546 18. Tabs are now shown as \t in -bP output.
5548 19. If the log selector return_path_on_delivery was set, Exim crashed when
5549 bouncing a message because it had too many Received: header lines.
5551 20. If two routers both had headers_remove settings, and the first one included
5552 a superfluous trailing colon, the final name in the first list and the
5553 first name in the second list were incorrectly joined into one item (with a
5554 colon in the middle).
5560 1. Added -C and -D options to the exinext utility, mainly to make it easier
5561 to include in the automated testing, but these could be helpful when
5562 multiple configurations are in use.
5564 2. The exinext utility was not formatting the output nicely when there was
5565 an alternate port involved in the retry record key, nor when there was a
5566 message id as well (for retries that were specific to a specific message
5567 and a specific host). It was also confused by IPv6 addresses, because of
5568 the additional colons they contain. I have fixed the IPv4 problem, and
5569 patched it up to do a reasonable job for IPv6.
5571 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5572 delivery, the log line now contains "pipelined" if PIPELINING was used.
5574 4. An SMTP transport process used to panic and die if the bind() call to set
5575 an explicit outgoing interface failed. This has been changed; it is now
5576 treated in the same way as a connect() failure.
5578 5. A reference to $sender_host_name in the part of a conditional expansion
5579 that was being skipped was still causing a DNS lookup. This no longer
5582 6. The def: expansion condition was not recognizing references to header lines
5583 that used bh_ and bheader_.
5585 7. Added the _cache feature to named lists.
5587 8. The code for checking quota_filecount in the appendfile transport was
5588 allowing one more file than it should have been.
5590 9. For compatibility with Sendmail, the command line option
5598 and sets the incoming protocol and host name (for trusted callers). The
5599 host name and its colon can be omitted when only the protocol is to be set.
5600 Note the Exim already has two private options, -pd and -ps, that refer to
5601 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5602 "s", but I don't think that's a major issue.
5604 10. A number of refactoring changes to the code, none of which should affect
5607 (a) The number of logging options was getting close to filling up the
5608 32-bit word that was used as a bit map. I have split them into two classes:
5609 those that are passed in the argument to log_write(), and those that are
5610 only ever tested independently outside of that function. These are now in
5611 separate 32-bit words, so there is plenty of room for expansion again.
5612 There is no change in the user interface or the logging behaviour.
5614 (b) When building, for example, log lines, the code previously used a
5615 macro that called string_cat() twice, in order to add two strings. This is
5616 not really sufficiently general. Furthermore, there was one instance where
5617 it was actually wrong because one of the argument was used twice, and in
5618 one call a function was used. (As it happened, calling the function twice
5619 did not affect the overall behaviour.) The macro has been replaced by a
5620 function that can join an arbitrary number of extra strings onto a growing
5623 (c) The code for expansion conditions now uses a table and a binary chop
5624 instead of a serial search (which was left over from when there were very
5625 few conditions). Also, it now recognizes conditions like "pam" even when
5626 the relevant support is not compiled in: a suitably worded error message is
5627 given if an attempt is made to use such a condition.
5629 11. Added ${time_interval:xxxxx}.
5631 12. A bug was causing one of the ddress fields not to be passed back correctly
5632 from remote delivery subprocesses. The field in question was not being
5633 subsequently used, so this caused to problems in practice.
5635 13. Added new log selectors queue_time and deliver_time.
5637 14. Might have fixed a bug in maildirsizefile handling that threw up
5638 "unexpected character" debug warnings, and recalculated the data
5639 unnecessarily. In any case, I expanded the warning message to give more
5642 15. Added the message "Restricted characters in address" to the statements in
5643 the default ACL that block characters like @ and % in local parts.
5645 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5646 Three changes have been made:
5648 (a) There was a serious bug; a negative response to MAIL caused the whole
5649 recipient domain to be cached as invalid, thereby blocking all messages
5650 to all local parts at the same domain, from all senders. This bug has
5651 been fixed. The domain is no longer cached after a negative response to
5652 MAIL if the sender used is not empty.
5654 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5657 (c) A new callout option, "use_sender" has been added for people who want
5658 the modified behaviour.
5664 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5667 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5668 indeed breaks things for older releases.
5670 3. Added additional logging to the case where there is a problem reading data
5671 from a filter that is running in a subprocess using a pipe, in order to
5672 try to track down a specific problem.
5674 4. Testing facility fudge: when running in the test harness and attempting
5675 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5676 getting "No route to host". Convert this to a timeout.
5678 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5681 6. Some OS don't have socklen_t but use size_t instead. This affects the
5682 fifth argument of getsockopt() amongst other things. This is now
5683 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5684 can be set for individual OS. I have set it for SunOS5, OSF1, and
5685 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5686 some earlier ones do not.
5688 7. Change 4.30/15 was not doing the test caselessly.
5690 8. The standard form for an IPv6 address literal was being rejected by address
5691 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5692 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5693 this, as well as the form without the "IPv6" on the front (but only when
5694 address literals are enabled, of course).
5696 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5698 10. Exim crashed if a message with an empty sender address specified by -f
5699 encountered a router with an errors_to setting. This could be provoked only
5700 by a command such as
5704 where an empty string was supplied; "<>" did not hit this bug.
5706 11. Installed PCRE release 4.5.
5708 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5709 remained set. It is now erased.
5711 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5712 times from message ids (which are base 36 rather than the normal 62).
5714 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5715 were being counted as actual protocol errors, and logged if the log
5716 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5717 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5718 and DATA following a set of rejected RCPTs do not count as protocol errors.
5719 In other words, Exim assumes they were pipelined, though this may not
5720 actually be the case. Of course, in all cases the client gets an
5721 appropriate error code.
5723 15. If a lookup fails in an ACL condition, a message about the failure may
5724 be available; it is used if testing the ACL cannot continue, because most
5725 such messages specify what the cause of the deferral is. However, some
5726 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5727 that caused an old message to be retained and used if a later statement
5728 caused a defer, replacing the real cause of the deferral.
5730 16. If an IP address had so many PTR records that the DNS lookup buffer
5731 was not large enough to hold them, Exim could crash while trying to process
5732 the truncated data. It now detects and logs this case.
5734 17. Further to 4.21/58, another change has been made: if (and only if) the
5735 first line of a message (the first header line) ends with CRLF, a bare LF
5736 in a subsequent header line has a space inserted after it, so as not to
5737 terminate the header.
5739 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5740 unnecessarily, used atoi() instead of strtol(), and didn't check the
5741 termination when getting file sizes from file names by regex.
5743 19. Completely re-implemented the support for maildirsize files, in the light
5744 of a number of problems with the previous contributed implementation
5745 (4.30/29). In particular:
5747 . If the quota is zero, the maildirsize file is maintained, but no quota is
5750 . If the maildir directory does not exist, it is created before any attempt
5751 to write a maildirsize file.
5753 . The quota value in the file is just a cache; if the quota is changed in
5754 the transport, the new value overrides.
5756 . A regular expression is available for excluding directories from the
5759 20. The autoreply transport checks the characters in options that define the
5760 message's headers; it allows continued headers, but it was checking with
5761 isspace() after an embedded newline instead of explicitly looking for a
5764 21. If all the "regular" hosts to which an address was routed had passed their
5765 expiry times, and had not reached their retry times, the address was
5766 bounced, even if fallback hosts were defined. Now Exim should go on to try
5769 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5770 equivalent code in the SMTP transport. Some hosts send humungous responses
5771 to HELO/EHLO, more than 1024 it seems.
5773 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5774 gives compiler warnings in some environments. I've now done it "properly",
5777 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5778 (because of problems with the built-in one) was declared to return uschar *
5779 instead of char *, causing compiler failure.
5781 25. Fixed a file descriptor leak when processing alias/forward files.
5783 26. Fixed a minor format string issue in dbfn.c.
5785 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5787 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5788 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5791 29. When a sender address is verified, it is cached, to save repeating the test
5792 when there is more than one recipient in a message. However, when the
5793 verification involves a callout, it is possible for different callout
5794 options to be set for different recipients. It is too complicated to keep
5795 track of this in the cache, so now Exim always runs a verification when a
5796 callout is required, relying on the callout cache for the optimization.
5797 The overhead is duplication of the address routing, but this should not be
5800 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5801 to be verified with callout=postmaster, and the main callout worked but the
5802 postmaster check failed, the verification correctly failed. However, if a
5803 subsequent RCPT command asked for sender verification *without* the
5804 postmaster check, incorrect caching caused this verification also to fail,
5807 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5808 it was not caching the DNS options (qualify_single, search_parents) that
5809 were used when the lookup failed. A subsequent lookup with different
5810 options therefore always gave the same answer, though there were cases
5811 where it should not have. (Example: a "domains = !$mx_any" option on a
5812 dnslookup router: the "domains" option is always processed without any
5813 widening, but the router might have qualify_single set.) Now Exim uses the
5814 cached value only when the same options are set.
5816 32. Added John Jetmore's "exipick" utility to the distribution.
5818 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5819 than a timeout (e.g. a certificate is required, and is not provided), an
5820 Exim server now closes the connection immediately. Previously it waited for
5821 the client to close - but if the client is SSL, it seems that they each
5822 wait for each other, leading to a delay before one of them times out.
5824 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5825 maintained 0.8.x compatibility because I don't think many are using it, and
5826 it is clearly obsolete.
5828 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5831 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5832 client certificate was expired. A simple patch fixes this, though I don't
5833 understand the full logic of why the verify callback is called multiple
5836 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5837 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5838 which causes problems with some clients (such as the Certicom SSL Plus
5839 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5840 disables the coutermeasure allowing Eudora to connect."
5842 38. Exim was not checking that a write() to a log file succeeded. This could
5843 lead to Bad Things if a log got too big, in particular if it hit a file
5844 size limit. Exim now panics and dies if it cannot write to a log file, just
5845 as it does if it cannot open a log file.
5847 39. Modified OS/Makefile-Linux so that it now contains
5849 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5851 The two -D definitions ensure that Exim is compiled with large file
5852 support, which makes it possible to handle log files that are bigger than
5855 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5856 instance) a domain was checked against a named list that involved a lookup,
5857 causing $domain_data to be set, then another domain was checked against the
5858 same list, then the first domain was re-checked, the value of $domain_data
5859 after the final check could be wrong. In particular, if the second check
5860 failed, it could be set empty. This bug probably also applied to
5863 41. The strip_trailing_dot option was not being applied to the address given
5864 with the -f command-line option.
5866 42. The code for reading a message's header from the spool was incrementing
5867 $received_count, but never initializing it. This meant that the value was
5868 incorrect (doubled) while delivering a message in the same process in which
5869 it was received. In the most common configuration of Exim, this never
5870 happens - a fresh exec is done - but it can happen when
5871 deliver_drop_privilege is set.
5873 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5874 it now includes up to 150 characters of the unexpected data in the log
5877 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5878 and building data strings. The size of both of these buffers was 10 000
5879 bytes - far larger than anybody would *ever* want, thought I. Needless to
5880 say, somebody hit the limit. I have increased the maximum line length to
5881 20 000 and the maximum data length of concatenated lines to 100 000. I have
5882 also fixed two bugs, because there was no checking on these buffers. Tsk,
5883 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5884 buffer is too small.
5886 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5887 lsearch lookups. Now it does.
5889 46. When parsing a route_list item in a manualroute router, a fixed-length
5890 buffer was used for the list of hosts. I made this 1024 bytes long,
5891 thinking that nobody would ever have a list of hosts that long. Wrong.
5892 Somebody had a whole pile of complicated expansion conditions, and the
5893 string was silently truncated, leading to an expansion error. It turns out
5894 that it is easier to change to an unlimited length (owing to other changes
5895 that have happened since this code was originally written) than to build
5896 structure for giving a limitation error. The length of the item that
5897 expands into the list of hosts is now unlimited.
5899 47. The lsearch lookup could not handle data where the length of text line was
5900 more than 4095 characters. Such lines were truncated, leading to shortened
5901 data being returned. It should now handle lines of any length.
5903 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5904 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5907 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5908 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5909 no longer happens in comment lines. A list of replacements is now placed
5910 at the head of all of the source files, except those whose only change is
5911 to replace PERL_COMMAND in the very first #! line.
5913 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5914 messages on the queue, with a bottom-up merge sort, using code contributed
5915 by Michael Haardt. This should make operations like -bp somewhat faster on
5916 large queues. It won't affect queue runners, except when queue_run_in_order
5919 51. Installed eximstats 1.31 in the distribution.
5921 52. Added support for SRV lookups to the dnslookup router.
5923 53. If an ACL referred to $message_body or $message_body_end, the value was not
5924 reset for any messages that followed in the same SMTP session.
5926 54. The store-handling optimization for building very long strings was not
5927 differentiating between the different store pools. I don't think this
5928 actually made any difference in practice, but I've tidied it.
5930 55. While running the routers to verify a sender address, $sender_address
5931 was still set to the sender address. This is wrong, because when routing to
5932 send a bounce to the sender, it would be empty. Therefore, I have changed
5933 it so that, while verifying a sender address, $sender_address is set to <>.
5934 (There is no change to what happens when verifying a recipient address.)
5936 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5937 target A or AAAA records (if not already returned) without resetting the
5938 qualify_single or search_parents options of the DNS resolver. These are
5939 inappropriate in this case because the targets of MX and SRV records must
5940 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5941 target that, when qualified, matched something in the local domain. These
5942 two options are now turned off when doing these lookups.
5944 57. It seems that at least some releases of Reiserfs (which does not have the
5945 concept of a fixed number of inodes) returns zero and not -1 for the
5946 number of available inodes. This interacted badly with check_spool_inodes,
5947 which assumed that -1 was the "no such thing" setting. What I have done is
5948 to check that the total number of inodes is greater than zero before doing
5949 the test of how many are available.
5951 58. When a "warn" ACL statement has a log_message modifier, the message is
5952 remembered, and not repeated. This is to avoid a lot of repetition when a
5953 message has many recipients that cause the same warning to be written.
5954 However, Exim was preserving the list of already written lines for an
5955 entire SMTP session, which doesn't seem right. The memory is now reset if a
5956 new message is started.
5958 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5959 debugging output unless log_rewrite was also set.
5961 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5962 of (int)(handle) when we know that handle contains (void *)(-1).
5964 61. The Exim daemon panic-logs an error return when it closes the incoming
5965 connection. However "connection reset by peer" seems to be common, and
5966 isn't really an error worthy of noting specially, so that particular error
5969 62. When Exim is trying to find all the local interfaces, it used to panic and
5970 die if the ioctl to get the interface flags failed. However, it seems that
5971 on at least one OS (Solaris 9) it is possible to have an interface that is
5972 included in the list of interfaces, but for which you get a failure error
5973 for this call. This happens when the interface is not "plumbed" into a
5974 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5975 failure of the "get flags" call assumes that the interface is down.
5977 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5978 makes life easier for people who are doing arithmetic on fields extracted
5979 from dates, where you often get leading zeros that should not be
5980 interpreted as octal.
5982 64. Added qualify_domain to the redirect router, to override the global
5985 65. If a pathologically long header line contained very many addresses (the
5986 report of this problem mentioned 10 000) and each of them was rewritten,
5987 Exim could use up a very large amount of memory. (It kept on making new
5988 copies of the header line as it rewrote, and never released the old ones.)
5989 At the expense of a bit more processing, the header rewriting function has
5990 been changed so that it no longer eats memory in this way.
5992 66. The generation of the Received: header has been moved from the time that a
5993 message starts to be received, to the time that it finishes. The timestamp
5994 in the Received: header should now be very close to that of the <= log
5995 line. There are two side-effects of this change:
5997 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5998 logged header lines no longer include the local Received: line, because
5999 it has not yet been created. The same applies to a copy of the message
6000 that is returned to a non-SMTP sender when a message is rejected.
6002 (b) When a filter file is tested using -bf, no additional Received: header
6003 is added to the test message. After some thought, I decided that this
6006 This change does not affect the value of $received_for. It is still set
6007 after address rewriting, but before local_scan() is called.
6009 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6011 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6012 gave an unhelpful panic error message, and a defer error. I have managed to
6013 change this behaviour so that it now rejects any supplied certificate,
6014 which seems right, as the list of acceptable certificates is empty.
6016 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6017 gave an unhelpful defer error. I have not managed to make this reject any
6018 supplied certificates, but the error message it gives is "no certificate
6019 supplied", which is not helpful.
6021 70. exigrep's output now also includes lines that are not associated with any
6022 message, but which match the given pattern. Implemented by a patch from
6023 Martin Sluka, which also tidied up the Perl a bit.
6025 71. Recipient callout verification, like sender verification, was using <> in
6026 the MAIL FROM command. This isn't really the right thing, since the actual
6027 sender may affect whether the remote host accepts the recipient or not. I
6028 have changed it to use the actual sender in the callout; this means that
6029 the cache record is now keyed on a recipient/sender pair, not just the
6030 recipient address. There doesn't seem to be a real danger of callout loops,
6031 since a callout by the remote host to check the sender would use <>.
6032 [SEE ABOVE: changed after hitting problems.]
6034 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6035 temporary errors. However, in the case of such a code being given after
6036 the end of a data transmission (i.e. after ".") Exim was failing to write
6037 a retry record for the message. (Yes, there was some broken host that was
6038 actually sending 8xx at this point.)
6040 73. An unknown lookup type in a host list could cause Exim to panic-die when
6041 the list was checked. (An example that provoked this was putting <; in the
6042 middle of a list instead of at the start.) If this happened during a DATA
6043 ACL check, a -D file could be left lying around. This kind of configuration
6044 error no longer causes Exim to die; instead it causes a defer error. The
6045 incident is still logged to the main and panic logs.
6047 74. Buglet left over from Exim 3 conversion. The message "too many messages
6048 in one connection" was written to the rejectlog but not the mainlog, except
6049 when address rewriting (yes!) was being logged.
6051 75. Added write_rejectlog option.
6053 76. When a system filter was run not as root (that is, when system_filter_user
6054 was set), the values of the $n variables were not being returned to the
6055 main process; thus, they were not subsequently available in the $sn
6058 77. Added +return_path_on_delivery log selector.
6060 78. A connection timeout was being treated differently from recipients deferred
6061 when testing hosts_max_try with a message that was older than the host's
6062 retry timeout. (The host should not be counted, thus allowing all hosts to
6063 be tried at least once before bouncing.) This may have been the cause of an
6064 occasionally reported bug whereby a message would remain on the queue
6065 longer than the retry timeout, but would be bounced if a delivery was
6066 forced. I say "may" because I never totally pinned down the problem;
6067 setting up timeout/retry tests is difficult. See also the next item.
6069 79. The ultimate address timeout was not being applied to errors that involved
6070 a combination of host plus message (for example, a timeout on a MAIL
6071 command). When an address resolved to a number of possible hosts, and they
6072 were not all tried for each delivery (e.g. because of hosts_max_try), a
6073 message could remain on the queue longer than the retry timeout.
6075 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6078 81. Fixed an obscure SMTP outgoing bug which required at least the following
6079 conditions: (a) there was another message waiting for the same server;
6080 (b) the server returned 5xx to all RCPT commands in the first message so
6081 that the message was not completed; (c) the server dropped the connection
6082 or gave a negative response to the RSET that Exim sends to abort the
6083 transaction. The observed case was a dropped connection after DATA that had
6084 been sent in pipelining mode. That is, the server had advertised PIPELINING
6085 but was not implementing it correctly. The effect of the bug was incorrect
6086 behaviour, such as trying another host, and this could lead to a crash.
6092 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6093 and daemon.c were passed as pointers to ints; they should have been
6094 pointers to socklen_t variables (which are typically unsigned ints).
6096 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6099 3. Fixed a really odd bug that affected only the testing scheme; patching a
6100 certain fixed string in the binary changed the value of another string that
6101 happened to be identical to the end of the original first string.
6103 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6104 name", it returns that address as the IP address. On some operating
6105 systems (e.g. Solaris), it also passes back the IP address string as the
6106 "host name". However, on others (e.g. Linux), it passes back an empty
6107 string. Exim wasn't checking for this, and was changing the host name to an
6108 empty string, assuming it had been canonicalized.
6110 5. Although rare, it is permitted to have more than one PTR record for a given
6111 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6112 all the names associated with an address, because they do in Solaris.
6113 However, it seems that they do not in Linux for data that comes from the
6114 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6115 I found this out when I moved to a new Linux workstation and tried to run
6116 the Exim test suite.
6118 To get round this problem I have changed the code so that it now does its
6119 own call to the DNS to look up PTR records when searching for a host name.
6120 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6121 addresses that are only in /etc/hosts are still found.
6123 This behaviour is, however, controlled by an option called host_lookup_
6124 order, which defaults to "bydns:byaddr". If people want to use the other
6125 order, or indeed, just use one or the other means of lookup, they can
6126 specify it in this variable.
6128 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6129 some operating systems, this comes back from gethostbyaddr() as an empty
6130 string, and this is what Exim used to test for. However, it seems that in
6131 other systems, "." is yielded. Exim now tests for this case too.
6133 7. The values of check_spool_space and check_log_space are now held internally
6134 as a number of kilobytes instead of an absolute number of bytes. If a
6135 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6136 kilobyte. This means that much larger values can be stored.
6138 8. Exim monitor: an attempt to get the action menu when not actually pointing
6139 at a message produces an empty menu entitled "No message selected". This
6140 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6141 no entries in it ("Shell widget menu has zero width and/or height"). So I
6142 have added a single, blank menu entry in this case.
6144 9. Added ${quote_local_part.
6146 10. MIME decoding is now applied to the contents of Subject: header lines when
6149 11. Now that a reference to $sender_host_address automatically causes a reverse
6150 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6151 host lookup before query-style lookups in lists that might use this
6152 variable. This has therefore been abolished, and the "net-" prefix is no
6153 longer necessary for query-style lookups.
6155 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6156 have been a typo for LFLAGS, so it has been changed.
6158 13. The install script calls Exim with "-C /dev/null" in order to find the
6159 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6160 to be output. However, since Exim outputs its version number before the
6161 error, it didn't break the script. It just looked ugly. I fixed this by
6162 always allowing "-C /dev/null" if the caller is root.
6164 14. Ignore overlarge ACL variable number when reading spool file - insurance
6165 against a later release with more variables having written the file.
6167 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6168 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6169 this, as well as the form without the "IPv6" on the front.
6171 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6172 OS/Makefile-Darwin file.
6174 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6175 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6177 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6178 variable. However, some OS use ints for inodes. I've added cast to long int
6179 to get rid of the compiler warning.
6181 19. I had forgotten to lock out "/../" in configuration file names when
6182 ALT_CONFIG_PREFIX was set.
6184 20. Routers used for verification do not need to specify transports. However,
6185 if such a router generated a host list, and callout was configured, Exim
6186 crashed, because it could not find a port number from the (non-existent)
6187 transport. It now assumes port 25 in this circumstance.
6189 21. Added the -t option to exigrep.
6191 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6192 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6193 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6194 them is set, LOOKUP_LSEARCH is forced.
6196 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6198 24. Added sender and host information to the "rejected by local_scan()" log
6199 line; previously there was no indication of these.
6201 25. Added .include_if_exists.
6203 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6204 Linux. It turns out that not all file systems support this. Apparently some
6205 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6206 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6207 not supported on the file descriptor, is now ignored when Exim is trying to
6208 sync a directory. This applies only to Linux.
6210 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6212 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6213 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6216 29. Added support for maildirsize files from supplied patch (modified a bit).
6218 30. The use of :fail: followed by an empty string could lead Exim to respond to
6219 sender verification failures with (e.g.):
6221 550 Verification failed for <xxx>
6222 550 Sender verify failed
6224 where the first response line was missing the '-' that indicates it is not
6225 the final line of the response.
6227 31. The loop for finding the name of the user that called Exim had a hardwired
6228 limit of 10; it now uses the value of finduser_retries, which is used for
6229 all other user lookups.
6231 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6234 33. Exim was neglecting to zero errno before one call of strtol() when
6235 expanding a string and expecting an integer value. On some systems this
6236 resulted in spurious "integer overflow" errors. Also, it was casting the
6237 result into an int without checking.
6239 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6240 did not work. The code looks as if it has *never* worked, though it appears
6241 to have been documented since at least release 1.62. I have made it work.
6243 35. The "timeout_DNS" error in retry rules, also documented since at least
6244 1.62, also never worked. As it isn't clear exactly what this means, and
6245 clearly it isn't a major issue, I have abolished the feature by treating it
6246 as "timeout", and writing a warning to the main and panic logs.
6248 36. The display of retry rules for -brt wasn't always showing the error code
6251 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6252 timeout_connect_A, timeout_connect_MX.
6254 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6255 to the empty sender.
6257 39. The daemon was not analysing the content of -oX till after it had closed
6258 stderr and disconnected from the controlling terminal. This meant that any
6259 syntax errors were only noted on the panic log, and the return code from
6260 the command was 0. By re-arranging the code a little, I've made the
6261 decoding happen first, so such errors now appear on stderr, and the return
6262 code is 1. However, the actual setting up of the sockets still happens in
6263 the disconnected process, so errors there are still only recorded on the
6266 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6267 connections (as happens on some IP stacks) was logged at start up time as
6268 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6269 it from "IPv6 and IPv4", which means that two separate sockets are being
6272 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6273 says whether AF_INET or AF_INET6 was passed as an argument.
6275 42. Exiwhat output was messed up when time zones were included in log
6278 43. Exiwhat now gives more information about the daemon's listening ports,
6279 and whether -tls-on-connect was used.
6281 44. The "port" option of the smtp transport is now expanded.
6283 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6284 silently ignored. Now an error message is written to the main and panic
6287 46. There's a new ACL modifier called "logwrite" which writes to a log file
6288 as soon as it is encountered.
6290 47. Added $local_user_uid and $local_user_gid at routing time.
6292 48. Exim crashed when trying to verify a sender address that was being
6295 49. Exim was recognizing only a space character after ".include". It now also
6296 recognizes a tab character.
6298 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6299 extracting the relevant information from the specification. The man page no
6300 longer contains scrambled data for the -d option, and I've added a section
6301 at the front about calling Exim under different names.
6303 51. Added "extra_headers" argument to the "mail" command in filter files.
6305 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6308 53. Installed eximstats 1.29.
6310 54. Added transport_filter_timeout as a generic transport option.
6312 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6313 Cc: header lines. This was required by RFC 822, but it not required by RFC
6316 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6317 incoming messages that did not have them. Now it does so only if the
6318 message originates locally, that is, if there is no associated remote host
6319 address. When Resent- header lines are present, this applies to the Resent-
6320 lines rather than the non-Resent- lines.
6322 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6323 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6325 58. Messages for configuration errors now include the name of the main
6326 configuration file - useful now that there may be more than one file in a
6327 list (.included file names were always shown).
6329 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6330 for those rare installations that do not start the daemon as root or run it
6331 setuid root. I've cut out the call to initgroups() if the daemon is not
6334 60. The Exim user and group can now be bound into the binary as text strings
6335 that are looked up at the start of Exim's processing.
6337 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6339 62. Added $mailstore_basename variable.
6341 63. Installed patch to sieve.c from Michael Haardt.
6343 64. When Exim failed to open the panic log after failing to open the main log,
6344 the original message it was trying to log was written to stderr and debug
6345 output, but if they were not available (the usual case in production), it
6346 was lost. Now it is written to syslog before the two lines that record the
6347 failures to open the logs.
6349 65. Users' Exim filters run in subprocesses under the user's uid. It is
6350 possible for a "deliver" command or an alias in a "personal" command to
6351 provoke an address rewrite. If logging of address rewriting is configured,
6352 this fails because the process is not running as root or exim. There may be
6353 a better way of dealing with this, but for the moment (because 4.30 needs
6354 to be released), I have disabled address rewrite logging when running a
6355 filter in a non-root, non-exim process.
6361 1. The buildconfig auxiliary program wasn't quoting the value set for
6362 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6363 not defined. This bug was masked in 4.22 by the effect that was fixed in
6366 2. Some messages that were rejected after a message id was allocated were
6367 shown as "incomplete" by exigrep. It no longer does this for messages that
6368 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6370 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6371 have allow_domain_literals set, the ID did not get logged in the <= line.
6372 Domain literals are now always recognized in Message-ID: header lines.
6374 4. The first argument for a ${extract expansion item is the key name or field
6375 number. Leading and trailing spaces in this item were not being ignored,
6376 causing some misleading effects.
6378 5. When deliver_drop_privilege was set, single queue runner processes started
6379 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6380 same command in the process it spins off) were not dropping privilege.
6382 6. When the daemon running as "exim" started a queue runner, it always
6383 re-executed Exim in the spun-off process. This is a waste of effort when
6384 deliver_drop_privilege is set. The new process now just calls the
6385 queue-runner function directly.
6391 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6394 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6395 ignored. Though the use of -oP was forcing the writing of a pid file, it
6396 was always written to the default place.
6398 3. If the message "no IP address found for host xxxx" is generated during
6399 incoming verification, it is now followed by identification of the incoming
6400 connection (so you can more easily find what provoked it).
6402 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6404 5. Added some features to "harden" Exim a bit more against certain attacks:
6406 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6407 be put in Local/Makefile. This is like the never_users runtime option,
6408 but it cannot be overridden. The default setting is "root".
6410 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6411 prefix string with which any file named in a -C command line option
6414 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6415 is retained for -C and -D only if the caller of Exim is root. Without
6416 it, the exim user may also use -C and -D and retain privilege.
6418 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6419 command line option is disabled.
6421 6. Macro names set by the -D option must start with an upper case letter, just
6422 like macro names defined in the configuration file.
6424 7. Added "dereference=" facility to LDAP.
6426 8. Two instances of the typo "uknown" in the source files are fixed.
6428 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6429 the Configure-Makefile script screwed up while processing it.
6431 10. Incorporated PCRE 4.4.
6433 11. The SMTP synchronization check was not operating right at the start of an
6434 SMTP session. For example, it could not catch a HELO sent before the client
6435 waited for the greeting. There is now a check for outstanding input at the
6436 point when the greeting is written. Because of the duplex, asynchronous
6437 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6438 way, but not yet received, when the check is performed.
6440 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6441 on TCP/IP sockets, because this apparently causes some broken clients to
6444 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6445 unchanged) from the Cygwin maintainer.
6447 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6448 was supported instead of testing for mbx. Effectively a typo.
6450 15. The spa authenticator server code was not checking that the input it
6451 received was valid base64.
6453 16. The debug output line for the "set" modifier in ACLs was not showing the
6454 name of the variable that was being set.
6456 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6458 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6459 lookup on its own. Something else was needed to trigger the lookup. For
6460 example, a match in host_lookup or the need for a host name in a host list.
6461 Now, if $sender_host_name is referenced and the host name has not yet been
6462 looked up, a lookup is performed. If the lookup fails, the variable remains
6463 empty, and $host_lookup_failed is set to "1".
6465 19. Added "eqi" as a case-independent comparison operator.
6467 20. The saslauthd authentication condition could segfault if neither service
6468 nor realm was specified.
6470 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6471 error message that was logged was misleading, and incoming SMTP
6472 connections were dropped. The message is now more accurate, and temporary
6473 errors are given to SMTP connections.
6475 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6476 (except RSET) with a 421 error, until QUIT is received. However, it was
6477 failing to send a response to QUIT.
6479 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6480 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6481 ACL is now run after the helo_try_verify_hosts code.
6483 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6484 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6485 case-independent, so other case variants are also recognized). Apparently
6486 some systems use these upper case variants.
6488 25. If more than two messages were waiting for the same host, and a transport
6489 filter was specified for the transport, Exim sent two messages over the
6490 same TCP/IP connection, and then failed with "socket operation on non-
6491 socket" when it tried to send the third.
6493 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6495 27. The extern definition of crypt16() in expand.c was not being excluded when
6496 the OS had its own crypt16() function.
6498 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6499 as a preferred synonym for return_size_limit, both as an option and as an
6502 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6504 30. Changed the default configuration ACL to relax the local part checking rule
6505 for addresses that are not in any local domains. For these addresses,
6506 slashes and pipe symbols are allowed within local parts, but the sequence
6507 /../ is explicitly forbidden.
6509 31. SPA server authentication was not clearing the challenge buffer before
6512 32. log_message in a "warn" ACL statement was writing to the reject log as
6513 well as to the main log, which contradicts the documentation and doesn't
6514 seem right (because no rejection is happening). So I have stopped it.
6516 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6517 However, I am unable to do any testing of this.
6519 34. Fixed an infelicity in the appendfile transport. When checking directories
6520 for a mailbox, to see if any needed to be created, it was accidentally
6521 using path names with one or more superfluous leading slashes; tracing
6522 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6524 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6525 contents are added to the log line that is written for every discarded
6526 recipient. (Previously a log_message setting was ignored.)
6528 36. The ${quote: operator now quotes the string if it is empty.
6530 37. The install script runs exim in order to find its version number. If for
6531 some reason other than non-existence or emptiness, which it checks, it
6532 could not run './exim', it was installing it with an empty version number,
6533 i.e. as "exim-". This error state is now caught, and the installation is
6536 38. An argument was missing from the function that creates an error message
6537 when Exim fails to connect to the socket for saslauthd authentication.
6538 This could cause Exim to crash, or give a corrupted message.
6540 39. Added isip, isip4, and isip6 to ${if conditions.
6542 40. The ACL variables $acl_xx are now saved with the message, and can be
6543 accessed later in routers, transports, and filters.
6545 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6546 strings in the file are not string-expanded.
6548 42. If a MAIL command specified a SIZE value that was too large to fit into an
6549 int variable, the check against message_size_limit failed. Such values are
6550 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6551 day this will have to be increased, but I don't think I want to be around
6552 when emails are that large.
6559 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6560 iconv() is not standard in FreeBSD.
6562 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6563 IPv6 enabled. The observed symptom was a segmentation fault on return from
6564 the function os_common_find_running_interfaces() in src/os.c.
6566 3. In the check_special_case() function in daemon.c I had used "errno" as an
6567 argument name, which causes warnings on some systems. This was basically a
6568 typo, since it was named "eno" in the comments!
6570 4. The code that waits for the clock to tick (at a resolution of some fraction
6571 of a second) so as to ensure message-id uniqueness was always waiting for
6572 at least one whole tick, when it could have waited for less. [This is
6573 almost certainly not relevant at current processor speeds, where it is
6574 unlikely to ever wait at all. But we try to future-proof.]
6576 5. The function that sleeps for a time interval that includes fractions of a
6577 second contained a race. It did not block SIGALRM between setting the
6578 timer, and suspending (a couple of lines later). If the interval was short
6579 and the sigsuspend() was delayed until after it had expired, the suspension
6580 never ended. On busy systems this could lead to processes getting stuck for
6583 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6584 process, before it forks any delivery processes. The open lookup caching
6585 mechanism meant that the open file or database connection was passed into
6586 the delivery process. The problem was that delivery processes always tidy
6587 up cached lookup data. This could cause a problem for the next delivery
6588 process started by the queue runner, because the external queue runner
6589 process does not know about the closure. So the next delivery process
6590 still has data in the lookup cache. In the case of a file lookup, there was
6591 no problem because closing a file descriptor in a subprocess doesn't affect
6592 the parent. However, if the lookup was caching a connection to a database,
6593 the connection was closed, and the second delivery process was likely to
6594 see errors such as "PGSQL: query failed: server closed the connection
6595 unexpectedly". The problem has been fixed by closing all cached lookups
6596 in a queue runner before running a delivery process.
6598 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6599 seem to have the "const" qualifier which it has on other OS. I've
6602 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6603 *of the same type* (client or server) with the same public name that an
6604 error should be diagnosed.
6606 9. When Exim looked up a host name for an IP address, but failed to find the
6607 original IP address when looking up the host name (a safety check), it
6608 output the message "<ip address> does not match any IP for NULL", which was
6609 confusing, to say the least. The bug was that the host name should have
6610 appeared instead of "NULL".
6612 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6613 user that is built into the binary, and the -C or -D options is used, root
6614 privilege is dropped before the configuration file is read. In addition,
6615 logging is switched to stderr instead of the normal log files. If the
6616 configuration then re-defines the Exim user, the unprivileged environment
6617 is probably not what is expected, so Exim logs a panic warning message (but
6620 However, if deliver_drop_privilege is set, the unprivileged state may well
6621 be exactly what is intended, so the warning has been cut out in that case,
6622 and Exim is allowed to try to write to its normal log files.
6628 1. smtp_return_error_details was not giving details for temporary sender
6629 or receiver verification errors.
6631 2. Diagnose a configuration error if two authenticators have the same public
6634 3. Exim used not to create the message log file for a message until the first
6635 delivery attempt. This could be confusing when incoming messages were held
6636 for policy or load reasons. The message log file is now created at the time
6637 the message is received, and an initial "Received" line is written to it.
6639 4. The automatically generated man page for command line options had a minor
6640 bug that caused no ill effects; however, a more serious problem was that
6641 the procedure for building the man page automatically didn't always
6642 operate. Consequently, release 4.20 contains an out-of-date version. This
6643 shouldn't happen again.
6645 5. When building Exim with embedded Perl support, the script that builds the
6646 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6647 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6649 6. The freeze_tell option was not being used for messages that were frozen on
6650 arrival, either by an ACL or by local_scan().
6652 7. Added the smtp_incomplete_transaction log selector.
6654 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6655 was accepting AUTH without a new EHLO.
6657 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6658 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6661 10. Logging of TCP/IP connections (when configured) now happens in the main
6662 daemon process instead of the child process, so that the TCP/IP connection
6663 count is more accurate (but it can never be perfect).
6665 11. The use of "drop" in a nested ACL was not being handled correctly in the
6666 outer ACL. Now, if condition failure induced by the nested "drop" causes
6667 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6668 or "require"), the connection is dropped.
6670 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6671 that yield "discard" can now be used with an "accept" or a "discard" verb,
6672 but an error is generated for any others (because I can't see a useful way
6673 to define what should happen).
6675 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6676 are now processed in the same way as lines in the Exim configuration file.
6677 In particular, continuation lines are supported.
6679 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6681 15. Added -ti meaning -t -i.
6683 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6684 domains, and warn by logging if others are found.
6686 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6687 returned from GIFCONF when Exim is trying to find the list of interfaces on
6688 a host. The code in os.c has been modified to copy each ifreq to an aligned
6689 structure in all cases.
6691 Also, in some cases, the returned ifreq's were being copied to a 'struct
6692 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6693 means the last couple of bytes of an IPv6 address could be chopped if the
6694 ifreq contained only a normal sockaddr (14 bytes storage).
6696 18. Named domain lists were not supported in the hosts_treat_as_local option.
6697 An entry such as +xxxx was not recognized, and was treated as a literal
6700 19. Ensure that header lines added by a DATA ACL are included in the reject log
6701 if the ACL subsequently rejects the message.
6703 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6704 MD5 (which is deprecated).
6706 21. When testing a filter file using -bf, Exim was writing a message when it
6707 took the sender from a "From " line in the message, but it was not doing so
6708 when it took $return_path from a Return-Path: header line. It now does.
6710 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6711 with a valid header line field name (a series of printing characters
6712 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6714 23. Changed "disc" in the source to "disk" to conform to the documentation and
6715 the book and for uniformity.
6717 24. Ignore Sendmail's -Ooption=value command line item.
6719 25. When execve() failed while trying to run a command in a pipe transport,
6720 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6721 could be confused with a return value of 69 from the command itself. This
6722 has been changed to 127, the value the shell returns if it is asked to run
6723 a non-existent command. The wording for the related log line suggests a
6724 non-existent command as the problem.
6726 26. If received_header_text expands to an empty string, do not add a Received:
6727 header line to the message. (Well, it adds a token one on the spool, but
6728 marks it "old" so that it doesn't get used or transmitted.)
6730 27. Installed eximstats 1.28 (addition of -nt option).
6732 28. There was no check for failure on the call to getsockname() in the daemon
6733 code. This can fail if there is a shortage of resources on the system, with
6734 ENOMEM, for example. A temporary error is now given on failure.
6736 29. Contrary to the C standard, it seems that in some environments, the
6737 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6738 program. Exim now does this explicitly; it affects the formatting of
6739 timestamps using strftime().
6741 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6742 complaints. I've now initialized all the variables, to avoid this.
6744 32. Header lines added by a system filter were not being "seen" during
6745 transport-time rewrites.
6747 33. The info_callback() function passed to OpenSSL is set up with type void
6748 (*)(SSL *, int, int), as described somewhere. However, when calling the
6749 function (actually a macro) that sets it up, the type void(*)() is
6750 expected. I've put in a cast to prevent warnings from picky compilers.
6752 34. If a DNS black list lookup found a CNAME record, but there were no A
6753 records associated with the domain it pointed at, Exim crashed.
6755 35. If a DNS black list lookup returned more than one A record, Exim ignored
6756 all but the first. It now scans all returned addresses if a particular IP
6757 value is being sought. In this situation, the contents of the
6758 $dnslist_value variable are a list of all the addresses, separated by a
6761 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6762 to accept a host name and all its aliases if the forward lookup for any of
6763 them yielded the IP address of the incoming connection. Now it accepts only
6764 those names whose forward lookup yields the correct IP address. Any other
6765 names are discarded. This closes a loophole whereby a rogue DNS
6766 administrator could create reverse DNS records to break through a
6767 wildcarded host restriction in an ACL.
6769 37. If a user filter or a system filter that ran in a subprocess used any of
6770 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6771 the wrong values were passed to the pipe command ($thisaddress had the
6772 value of $0, $0 had the value of $1, etc). This bug was introduced by
6773 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6775 38. Improved the line breaking for long SMTP error messages from ACLs.
6776 Previously, if there was no break point between 40 and 75 characters, Exim
6777 left the rest of the message alone. Two changes have been made: (a) I've
6778 reduced the minimum length to 35 characters; (b) if it can't find a break
6779 point between 35 and 75 characters, it looks ahead and uses the first one
6780 that it finds. This may give the occasional overlong line, but at least the
6781 remaining text gets split now.
6783 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6784 file descriptors might be low, and that setting 1000 would always raise it.
6785 It turns out that in some environments, the limit is already over 1000 and
6786 that lowering it causes trouble. So now Exim takes care not to decrease it.
6788 40. When delivering a message, the value of $return_path is set to $sender_
6789 address at the start of routing (routers may change the value). By an
6790 oversight, this default was not being set up when an address was tested by
6791 -bt or -bv, which affected the outcome if any router or filter referred to
6794 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6795 log only when "message" or "log_message" are set. However, if one of the
6796 conditions was an address verification, or a call to a nested ACL, the
6797 messages generated by the underlying test were being passed through. This
6798 no longer happens. The underlying message is available in $acl_verify_
6799 message for both "message" and "log_message" expansions, so it can be
6800 passed through if needed.
6802 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6803 new expansion $bh_ to give the encoded byte string without charset
6804 translation. Translation happens only if iconv() is available; HAVE_ICONV
6805 indicates this at build time. HEADERS_CHARSET gives the charset to
6806 translate to; headers_charset can change it in the configuration, and
6807 "headers charset" can change it in an individual filter file.
6809 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6810 that creates RFC 2047 encoded "words" labels them as that charset instead
6811 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6812 expansion operator; (ii) when Exim creates a From: line for a local
6813 message; (iii) when a header line is rewritten to include a "phrase" part.
6815 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6816 buggy, causing it to skip the first lines of messages whose message ID
6817 ended in 'D'. This would not have bitten before Exim release 4.14, because
6818 message IDs were unlikely to end in 'D' before then. The effect was to have
6819 incorrect size information for certain domains.
6821 45. #include "config.h" was missing at the start of the crypt16.c module. This
6822 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6825 46. If there was a timeout during a "random" callout check, Exim treated it as
6826 a failure of the random address, and carried on sending RSET and the real
6827 address. If the delay was just some slowness somewhere, the response to the
6828 original RCPT would be taken as a response to RSET and so on, causing
6829 mayhem of various kinds.
6831 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6832 when I implemented it. It didn't allow for the fact that some option values
6833 may legitimately be negative (e.g. size_addition), and it didn't even do
6834 the right test for positive values.
6836 48. Domain names in DNS records are case-independent. Exim always looks them up
6837 in lower case. Some resolvers return domain names in exactly the case they
6838 appear in the zone file, that is, they may contain uppercase letters. Not
6839 all resolvers do this - some return always lower case. Exim was treating a
6840 change of case by a resolver as a change of domain, similar to a widening
6841 of a domain abbreviation. This triggered its re-routing code and so it was
6842 trying to route what was effectively the same domain again. This normally
6843 caused routing to fail (because the router wouldn't handle the domain
6844 twice). Now Exim checks for this case specially, and just changes the
6845 casing of the domain that it ultimately uses when it transmits the message
6848 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6851 50. If a filter generated a file delivery with a non-absolute name (possible if
6852 no home directory exists for the router), the forbid_file option was not
6855 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6856 the existing equality matching.
6858 52. Exim was using ints instead of ino_t variables in some places where it was
6859 dealing with inode numbers.
6861 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6862 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6863 called TMPDIR, and if it finds it is different, it changes its value.
6865 54. The smtp_printf() function is now made available to local_scan() so
6866 additional output lines can be written before returning. There is also an
6867 smtp_fflush() function to enable the detection of a dropped connection.
6868 The variables smtp_input and smtp_batched_input are exported to
6871 55. Changed the default runtime configuration: the message "Unknown user"
6872 has been removed from the ACL, and instead placed on the localuser router,
6873 using the cannot_route_message feature. This means that any verification
6874 failures that generate their own messages won't get overridden. Similarly,
6875 the "Unrouteable address" message that was in the ACL for unverifiable
6876 relay addresses has also been removed.
6878 56. Added hosts_avoid_esmtp to the smtp transport.
6880 57. The exicyclog script was not checking for the esoteric option
6881 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6882 will work only if exicyclog is run under the appropriate euid.
6884 58. Following a discussion on the list, the rules by which Exim recognises line
6885 endings on incoming messages have been changed. The -dropcr and drop_cr
6886 options are now no-ops, retained only for backwards compatibility. The
6887 following line terminators are recognized: LF CRLF CR. However, special
6888 processing applies to CR:
6890 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6891 nor a local message in the state where . is a terminator.
6893 (ii) If a bare CR is encountered in a header line, an extra space is added
6894 after the line terminator so as not to end the header. The reasoning
6895 behind this is that bare CRs in header lines are most likely either
6896 to be mistakes, or people trying to play silly games.
6898 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6899 was being incorrectly given as 18 bytes larger than it should have been.
6900 This is a VOB (very old bug).
6902 60. This may never have affected anything current, but just in case it has:
6903 When the local host is found other than at the start of a list of hosts,
6904 the local host, those with the same MX, and any that follow, are discarded.
6905 When the list in question was part of a longer list of hosts, the following
6906 hosts (not currently being processed) were also being discarded. This no
6907 longer happens. I'm not sure if this situation could ever has previously
6910 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6913 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6914 header. This is something that is recommended in a new Internet Draft, and
6915 is something that is documented as being done by Sendmail. There are two
6916 possible values. For messages generated by the autoreply transport, Exim
6919 Auto-Submitted: auto-replied
6921 whereas for all other generated messages (e.g. bounces) it adds
6923 Auto-Submitted: auto-generated
6925 63. The "personal" condition in filters now includes a test for the
6926 Auto-Submitted: header. If it contains the string "auto-" the message it
6927 not considered personal.
6929 64. Added rcpt_include_affixes as a generic transport option.
6931 65. Added queue_only_override (default true).
6933 66. Added the syslog_duplication option.
6935 67. If what should have been the first header line of a message consisted of
6936 a space followed by a colon, Exim was mis-interpreting it as a header line.
6937 It isn't of course - it is syntactically invalid and should therefore be
6938 treated as the start of the message body. The misbehaviour could have
6939 caused a number of strange effects, including loss of data in subsequent
6940 header lines, and spool format errors.
6942 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6943 client host had authenticated. This control can now be exercised by an ACL
6944 for more flexibility.
6946 69. By default, callouts do not happen when testing with -bh. There is now a
6947 variant, -bhc, which does actually run the callout code, including
6948 consulting and updating the callout cache.
6950 70. Added support for saslauthd authentication, courtesy of Alexander
6953 71. If statvfs() failed on the spool or log directories while checking their
6954 size for availability, Exim confusingly gave the error "space shortage".
6955 Furthermore, in debugging mode it crashed with a floating point exception.
6956 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6957 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6958 serious problem, Exim now writes to the main and panic logs when this
6959 happens, with details of the failure. It then refuses to accept the
6960 incoming message, giving the message "spool directory problem" or "log
6961 directory problem" with a 421 code for SMTP messages.
6963 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6964 0, 1, and 2 exist, because some OS complain for execs without them (see
6965 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6966 descriptors. However, the code omitted to check that the open succeeded,
6967 causing mysterious errors if for some reason the permissions on /dev/null
6968 got screwed. Now Exim writes a message to the main and panic logs, and
6969 bombs out if it can't open /dev/null.
6971 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6972 interact so that it is all more flexible. It is supposed to remain
6973 backwards compatible. Also added extra_local_interfaces.
6975 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6976 to bomb out with an assertion failure - to the client this appears as a
6977 connection drop. This problem occurs in the part of the code that was taken
6978 from the Samba project. Fortunately, the assertion is in a very simple
6979 function, so I have fixed this by reproducing the function inline in the
6980 one place where it is called, and arranging for authentication to fail
6981 instead of killing the process with assert().
6983 75. The SPA client code was not working when the server requested OEM rather
6984 than Unicode encoding.
6986 76. Added code to make require_files with a specific uid setting more usable in
6987 the case where statting the file as root fails - usually a non-root-mounted
6988 NFS file system. When this happens and the failure is EACCES, Exim now
6989 forks a subprocess and does the per-uid checking as the relevant uid.
6991 77. Added process_log_path.
6993 78. If log_file_path was not explicitly set, a setting of check_log_space or
6994 check_log_inodes was ignored.
6996 79. If a space check for the spool or log partitions fails, the incident is now
6997 logged. Of course, in the latter case the data may get lost...
6999 80. Added the %p formatting code to string_format() so that it can be used to
7000 print addresses in debug_print(). Adjusted all the address printing in the
7001 debugging in store.c to use %p rather than %d.
7003 81. There was a concern that a line of code in smtp_in.c could overflow a
7004 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7005 initially expressed, the concern was not well-founded, because trailing
7006 spaces are removed early. However, if the trailing spaces were followed by
7007 a NULL, they did not get removed, so the overflow was possible. Two fixes
7010 (a) I re-wrote the offending code in a cleaner fashion.
7011 (b) If an incoming SMTP command contains a NULL character, it is rejected
7014 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7015 runs initgroups(), so that if the Exim user is in any additional groups,
7016 they will be used during message reception.
7022 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob