1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Rework error string handling in TLS interface so that the caller in
12 more cases is responsible for logging. This permits library-sourced
13 string to be attached to addresses during delivery, and collapses
14 pairs of long lines into single ones.
16 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
17 during configuration. Wildcards are allowed and expanded.
19 JH/02 Rework error string handling in DKIM to pass more info back to callers.
20 This permits better logging.
22 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
23 do not close it down and have the child transport start it up again on
24 the passed-on TCP connection. Instead, proxy the child (and any
25 subsequent ones) for TLS via a unix-domain socket channel. Logging is
26 affected: the continued delivery log lines do not have any DNSSEC, TLS
27 Certificate or OCSP information. TLS cipher information is still logged.
29 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
30 identical IP addresses on different listening ports. Will also affect
33 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
34 add noisy ifdef guards to special-case this sillyness.
35 Patch from Bernd Kuhls.
37 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
38 than 255 are no longer allowed.
40 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
41 Disable the session-cache too, which might reduce our load. Since we
42 currrectly use a new context for every connection, both as server and
43 client, there is no benefit for these.
44 GnuTLS appears to not support tickets server-side by default (we don't
45 call gnutls_session_ticket_enable_server()) but client side is enabled
46 by default on recent versions (3.1.3 +) unless the PFS priority string
49 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
50 <https://reproducible-builds.org/specs/source-date-epoch/>.
52 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
53 the check for any unsuccessful recipients did not notice the limit, and
54 erroneously found still-pending ones.
56 JH/08 Pipeline CHUNKING command and data together, on kernels that support
57 MSG_MORE. Only in-clear (not on TLS connections).
59 JH/09 Avoid using a temporary file during transport using dkim. Unless a
60 transport-filter is involved we can buffer the headers in memory for
61 creating the signature, and read the spool data file once for the
62 signature and again for transmission.
64 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
65 4.77 as the kernel support then wasn't solid, having issues in 64bit
66 mode. Now, it's been long enough. Add support for FreeBSD also.
68 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
69 case where the routing stage had gathered several addresses to send to
70 a host before calling the transport for the first, we previously failed
71 to close down TLS in the old transport process before passing the TCP
72 connection to the new process. The new one sent a STARTTLS command
73 which naturally failed, giving a failed delivery and bloating the retry
74 database. Investigation and fix prototype from Wolfgang Breyha.
76 JH/12 Fix check on SMTP command input synchronisation. Previously there were
77 false-negatives in the check that the sender had not preempted a response
78 or prompt from Exim (running as a server), due to that code's lack of
79 awareness of the SMTP input buffering.
81 PP/04 Add commandline_checks_require_admin option.
82 Exim drops privileges sanely, various checks such as -be aren't a
83 security problem, as long as you trust local users with access to their
84 own account. When invoked by services which pass untrusted data to
85 Exim, this might be an issue. Set this option in main configuration
86 AND make fixes to the calling application, such as using `--` to stop
89 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
90 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
91 responses to those, into a single TLS record each way (this usually means
92 a single packet). As a side issue, smtp_enforce_sync now works on TLS
95 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
96 affects you only if you're dancing at the edge of the param size limits.
97 If you are, and this message makes sense to you, then: raise the
98 configured limit or use OpenSSL 1.1. Nothing we can do for older
101 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
102 cmdline element to get the documented default one. Previously it was
105 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
106 are unsuable for non-message-specific reasons - eg. connection timeout,
107 banner-time rejection.
109 JH/16 Fix logging of delivery remote port, when specified by router, under
112 PP/06 Repair manualroute's ability to take options in any order, even if one
113 is the name of a transport.
116 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
118 JH/17 Change the list-building routines interface to use the expanding-string
119 triplet model, for better allocation and copying behaviour.
121 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
122 Previously it was constructed the first time a possibly-matching string
123 was met in the configuration file input during startup; now it is done
126 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
127 compatible one, to avoid the (poorly documented) possibility of a config
128 file in the working directory redirecting the DB files, possibly correpting
129 some existing file. CVE-2017-10140 assigned for BDB.
131 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
132 cache-hot. Previously, although the result was properly cached, the
133 initial verify call returned a defer.
135 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
136 the main verify for receipient in uncached-mode.
138 JH/22 Retire historical build files to an "unsupported" subdir. These are
139 defined as "ones for which we have no current evidence of testing".
141 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
142 if present. Previously it was ignored.
144 JH/24 Start using specified-initialisers in C structure init coding. This is
145 a C99 feature (it's 2017, so now considered safe).
147 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
148 if was a fixed-sized field and bitmask ops via macros; it is now more
151 PP/07 GitHub PR 56: Apply MariaDB build fix.
152 Patch provided by Jaroslav Škarvada.
154 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
155 during Coverity cleanups [4.87 JH/47]
156 Diagnosis and fix provided by Michael Fischer v. Mollard.
158 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
159 the right size to place the terminating semicolon on its own folded
160 line, the header hash was calculated to an incorrect value thanks to
161 the (relaxed) space the fold became.
163 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
164 and confused the parent.
166 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
167 which could crash as a result. This could lead to undeliverable messages.
169 JH/28 Logging: "next input sent too soon" now shows where input was truncated
172 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
173 matters on fast-turnover and PID-randomising systems, which were getting
174 out-of-order delivery.
176 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
177 a possibly-overlapping copy. The symptom was that "Remote host closed
178 connection in response to HELO" was logged instead of the actual 4xx
181 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
182 Previously only that bufferd was discarded, resulting in SYMTP command
185 JH/32 DKIM: when a message has multiple signatures matching an identity given
186 in dkim_verify_signers, run the dkim acl once for each. Previously only
187 one run was done. Bug 2189.
189 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
190 "panic the current process" to "deliberately defer". The panic log is
191 still written with the problem list name; the mail and reject logs now
192 get a temp-reject line for the message that was being handled, saying
193 something like "domains check lookup or other defer". The SMTP 451
194 message is still "Temporary local problem".
196 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
197 A crafted sequence of BDAT commands could result in in-use memory beeing
198 freed. CVE-2017-16943.
200 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
201 from SMTP input. Previously it was always done; now only done for DATA
202 and not BDAT commands. CVE-2017-16944.
204 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
205 to the message (such as an overlong header line). Previously this was
206 not done and we did not exit BDAT mode. Followon from the previous item
207 though a different problem.
213 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
214 than -2003 did; needs libidn2 in addition to libidn.
216 JH/02 The path option on a pipe transport is now expanded before use.
218 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
219 Patch provided by "Björn", documentation fix added too.
221 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
222 missing a wire-to-host endian conversion.
224 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
225 close after a BDAT command line could be taken as a following command,
226 giving a synch failure. Fix by only checking for synch immediately
227 before acknowledging the chunk.
229 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
230 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
231 macro. Patches provided by Josh Soref.
233 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
234 Previously we did not; the RFC seems ambiguous and VRFY is not listed
235 by IANA as a service extension. However, John Klensin suggests that we
238 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
239 the dkim code may be unix-mode line endings rather than smtp wire-format
240 CRLF, so prepend a CR to any bare LF.
242 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
243 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
245 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
246 the first were themselves being wrongly included in the feed into dkim
247 processing; with most chunk sizes in use this resulted in an incorrect
248 body hash calculated value.
250 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
251 DKIM signature block, for verification. Although advised against by
252 standards it is specifically not ruled illegal.
254 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
256 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
257 missing a body hash (the bh= tag).
259 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
260 It seems that HAProxy sends the Proxy Protocol information in clear and
261 only then does a TLS startup, so do the same.
263 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
264 TCP connections (such as for Spamd) unless the daemon successfully set
265 Fast Open mode on its listening sockets. This fixes breakage seen on
266 too-old kernels or those not configured for Fast Open, at the cost of
267 requiring both directions being enabled for TFO, and TFO never being used
268 by non-daemon-related Exim processes.
270 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
271 endings, at least on the first header line. Try to canonify any that get
272 past that check, despite the cost.
274 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
275 now limited to an arbitrary five deep, while parsing addresses with the
276 strip_excess_angle_brackets option enabled.
278 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
279 instead leave the unprompted TLS handshake in socket buffer for the
280 TLS library to consume.
282 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
284 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
286 JH/16 Drop variables when they go out of scope. Memory management drops a whole
287 region in one operation, for speed, and this leaves assigned pointers
288 dangling. Add checks run only under the testsuite which checks all
289 variables at a store-reset and panics on a dangling pointer; add code
290 explicitly nulling out all the variables discovered. Fixes one known
291 bug: a transport crash, where a dangling pointer for $sending_ip_address
292 originally assigned in a verify callout, is re-used.
294 PP/06 Drop '.' from @INC in various Perl scripts.
296 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
298 PP/08 Reduce a number of compilation warnings under clang; building with
299 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
300 should be warning-free.
302 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
304 HS/01 Fix portability problems introduced by PP/08 for platforms where
305 realloc(NULL) is not equivalent to malloc() [SunOS et al].
307 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
308 chunk. This allows us to accept broken chunked messages. We need a more
309 general solution here.
311 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
312 already-broken messages in the queue.
314 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
316 JH/19 Fix reference counting bug in routing-generated-address tracking.
322 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
323 supports it and a size is available (ie. the sending peer gave us one).
325 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
326 years of being deprecated). The replacements are the ACLs
327 acl_smtp_mime and acl_not_smtp_mime.
329 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
330 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
331 or even an in-clear connection were permitted. Now, if the host lookup
332 was dnssec and dane was requested then the host is only used if the
333 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
334 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
335 if one fails this test.
336 This means that a poorly-configured remote DNS will make it incommunicado;
337 but it protects against a DNS-interception attack on it.
339 JH/04 Bug 1810: make continued-use of an open smtp transport connection
340 non-noisy when a race steals the message being considered.
342 JH/05 If main configuration option tls_certificate is unset, generate a
343 self-signed certificate for inbound TLS connections.
345 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
346 in rewrites and routers.
348 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
349 and logged a warning sing 4.83; now they are a configuration file error.
351 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
352 (lacking @domain). Apply the same qualification processing as RCPT.
354 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
356 JH/10 Support ${sha256:} applied to a string (as well as the previous
359 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
360 a cutthrough deliver is pending, as we always want to make a connection.
361 This also avoids re-routing the message when later placing the cutthrough
362 connection after a verify cache hit.
363 Do not update it with the verify result either.
365 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
366 when routing results in more than one destination address.
368 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
369 signing (which inhibits the cutthrough capability). Previously only
370 the presence of an option was tested; now an expansion evaluating as
371 empty is permissible (obviously it should depend only on data available
372 when the cutthrough connection is made).
374 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
375 the relevant preceding SMTP command did not note the pipelining mode.
377 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
378 Previously they were not counted.
380 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
381 as one having no matching records. Previously we deferred the message
382 that needed the lookup.
384 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
385 distinguished as "(=".
387 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
388 for missing MX records. Previously it only worked for missing A records.
390 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
392 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
393 after the data-go-ahead and data-ack. Patch from Jason Betts.
395 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
396 even for a "none" policy. Patch from Tony Meyer.
398 JH/22 Fix continued use of a connection for further deliveries. If a port was
399 specified by a router, it must also match for the delivery to be
402 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
403 When one of the recipients of a message was unsuitable for the connection
404 (has no matching addresses), we lost track of needing to mark it
405 deferred. As a result mail would be lost.
407 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
409 JH/25 Decoding ACL controls is now done using a binary search; the source code
410 takes up less space and should be simpler to maintain. Merge the ACL
411 condition decode tables also, with similar effect.
413 JH/26 Fix problem with one_time used on a redirect router which returned the
414 parent address unchanged. A retry would see the parent address marked as
415 delivered, so not attempt the (identical) child. As a result mail would
418 JH/27 Fix a possible security hole, wherein a process operating with the Exim
419 UID can gain a root shell. Credit to http://www.halfdog.net/ for
420 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
423 JH/28 Enable {spool,log} filesystem space and inode checks as default.
424 Main config options check_{log,spool}_{inodes,space} are now
425 100 inodes, 10MB unless set otherwise in the configuration.
427 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
428 Previously it only applied to the main-section connection policy
431 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
433 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
434 by me. Added RFC7919 DH primes as an alternative.
436 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
437 are not in the system include path.
439 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
440 GnuTLS, when a session startup failed (eg because the client disconnected)
441 Exim did stdio operations after fclose. This was exposed by a recent
442 change which nulled out the file handle after the fclose.
444 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
445 signed directly by the cert-signing cert, rather than an intermediate
446 OCSP-signing cert. This is the model used by LetsEncrypt.
448 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
450 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
451 an incoming connection.
453 HS/02 Bug 1802: Do not half-close the connection after sending a request
456 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
457 fallback to "prime256v1".
459 JH/34 SECURITY: Use proper copy of DATA command in error message.
460 Could leak key material. Remotely exploitable. CVE-2016-9963.
466 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
467 and 3.4.4 - once the server is enabled to respond to an OCSP request
468 it does even when not requested, resulting in a stapling non-aware
469 client dropping the TLS connection.
471 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
472 support variable-length bit vectors. No functional change.
474 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
475 The I= interface field on outgoing lines is now after the H= remote
476 host field, same as incoming lines. There is a separate
477 outgoing_interface log selector which allows you to disable the
480 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
481 If not running log_selector +smtp_connection the mainlog would be held
482 open indefinitely after a "too many connections" event, including to a
483 deleted file after a log rotate. Leave the per net connection logging
484 leaving it open for efficiency as that will be quickly detected by the
485 check on the next write.
487 HS/01 Bug 1671: Fix post transport crash.
488 Processing the wait-<transport> messages could crash the delivery
489 process if the message IDs didn't exist for some reason. When
490 using 'split_spool_directory=yes' the construction of the spool
491 file name failed already, exposing the same netto behaviour.
493 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
494 mime_regex ACL conditions.
496 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
497 to DSN fail messages (bounces): remote IP, remote greeting, remote response
498 to HELO, local diagnostic string.
500 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
501 log line to debug. Even when configured with a tls authenticator many
502 client connections are expected to not authenticate in this way, so
503 an authenticate fail is not an error.
505 HS/02 Add the Exim version string to the process info. This way exiwhat
506 gives some more detail about the running daemon.
508 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
509 matter for fast-change records such as DNSBLs.
511 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
512 retry record, even if constant. There may be multiple transports with
513 different interface settings and the retry behaviour needs to be kept
516 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
518 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
520 JH/10 Bug 840: fix log_defer_output option of pipe transport
522 JH/11 Bug 830: use same host for all RCPTS of a message, even under
523 hosts_randomize. This matters a lot when combined with mua_wrapper.
525 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
526 ${quote_pgsql:<string>} operator.
528 JH/13 Bug 1708: avoid misaligned access in cached lookup.
530 JH/14 Change header file name for freeradius-client. Relevant if compiling
531 with Radius support; from the Gentoo tree and checked under Fedora.
533 JH/15 Bug 1712: Introduce $prdr_requested flag variable
535 JH/16 Bug 1714: Permit an empty string as expansion result for transport
536 option transport_filter, meaning no filtering.
538 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
540 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
541 defaults to "*" (all hosts). The variable is now available when not built
542 with TLS, default unset, mainly to enable keeping the testsuite sane.
543 If a server certificate is not supplied (via tls_certificate) an error is
544 logged, and clients will find TLS connections fail on startup. Presumably
545 they will retry in-clear.
546 Packagers of Exim are strongly encouraged to create a server certificate
547 at installation time.
549 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
550 with the $config_file variable.
552 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
553 in transport context, after the attempt, and per-recipient. The latter type
554 is per host attempted. The event data is the error message, and the errno
555 information encodes the lookup type (A vs. MX) used for the (first) host,
556 and the trailing two digits of the smtp 4xx response.
558 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
559 to write to mainlog (or rejectlog, paniclog) in the window between file
560 creation and permissions/ownership being changed. Particularly affects
561 installations where exicyclog is run as root, rather than exim user;
562 result is that the running daemon panics and dies.
564 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
566 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
567 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
568 "pri" and "weight". Note that the previous implicit priority given by the
569 list order is no longer honoured.
571 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
574 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
575 by defining SUPPORT_SOCKS.
577 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
578 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
579 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
580 variables are renamed to proxy_{local,external}_{address,port}.
582 JH/27 Move Internationalisation support from Experimental to mainline, enabled
583 for a build by defining SUPPORT_I18N
585 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
586 of the query string, and make ${quote_redis:} do that quoting.
588 JH/29 Move Events support from Experimental to mainline, enabled by default
589 and removable for a build by defining DISABLE_EVENT.
591 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
593 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
594 cached by the daemon.
596 JH/32 Move Redis support from Experimental to mainline, enabled for a build
597 by defining LOOKUP_REDIS. The libhiredis library is required.
599 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
600 keys are given for lookup.
602 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
603 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
604 only supported when built with TLS support. The PolarSSL SHA routines
605 are still used when the TLS library is too old for convenient support.
607 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
608 openssl_options), for security. OpenSSL forces this from version 1.1.0
609 server-side so match that on older versions.
611 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
612 allocation for $value could be released as the expansion processing
613 concluded, but leaving the global pointer active for it.
615 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
616 and to use the domains and local_parts ACL conditions.
618 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
619 incorrectly not doubled on cutthrough transmission, hence seen as a
620 body-termination at the receiving system - resulting in truncated mails.
621 Commonly the sender saw a TCP-level error, and retransmitted the message
622 via the normal store-and-forward channel. This could result in duplicates
623 received - but deduplicating mailstores were liable to retain only the
624 initial truncated version.
626 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
628 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
630 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
631 we're in there, support oversigning also; bug 1309.
633 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
635 HS/04 Add support for keep_environment and add_environment options.
637 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
638 either intentional arithmetic overflow during PRNG, or testing config-
641 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
642 delivery resulted in actual delivery. Cancel cutthrough before DATA
645 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
646 rejects a recipient: pass the reject to the originator.
648 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
649 Many were false-positives and ignorable, but it's worth fixing the
652 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
653 for the new environment-manipulation done at startup. Move the routines
654 from being local to tls.c to being global via the os.c file.
656 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
657 an extract embedded as result-arg for a map, the first arg for extract
658 is unavailable so we cannot tell if this is a numbered or keyed
659 extraction. Accept either.
665 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
668 JH/02 The smtp transport option "multi_domain" is now expanded.
670 JH/03 The smtp transport now requests PRDR by default, if the server offers
673 JH/04 Certificate name checking on server certificates, when exim is a client,
674 is now done by default. The transport option tls_verify_cert_hostnames
675 can be used to disable this per-host. The build option
676 EXPERIMENTAL_CERTNAMES is withdrawn.
678 JH/05 The value of the tls_verify_certificates smtp transport and main options
679 default to the word "system" to access the system default CA bundle.
680 For GnuTLS, only version 3.0.20 or later.
682 JH/06 Verification of the server certificate for a TLS connection is now tried
683 (but not required) by default. The verification status is now logged by
684 default, for both outbound TLS and client-certificate supplying inbound
687 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
690 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
691 Status Notification (bounce) messages are now MIME format per RFC 3464.
692 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
693 under the control of the dsn_advertise_hosts option, and routers may
694 have a dsn_lasthop option.
696 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
697 default, modifiable by a malware= option. The list separator for
698 the options can now be changed in the usual way. Bug 68.
700 JH/10 The smtp_receive_timeout main option is now expanded before use.
702 JH/11 The incoming_interface log option now also enables logging of the
703 local interface on delivery outgoing connections.
705 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
706 if the interface and destination host and port all match.
708 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
711 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
712 Patch from Andrew Lewis.
714 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
715 now supports optional time-restrictions, weighting, and priority
716 modifiers per server. Patch originally by <rommer@active.by>.
718 JH/16 The spamd_address main option now supports a mixed list of local
719 and remote servers. Remote servers can be IPv6 addresses, and
720 specify a port-range.
722 JH/17 Bug 68: The spamd_address main option now supports an optional
723 timeout value per server.
725 JH/18 Bug 1581: Router and transport options headers_add/remove can
726 now have the list separator specified.
728 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
731 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
734 JH/21 Support for the A6 type of dns record is withdrawn.
736 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
737 rather than the verbs used.
739 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
740 from 255 to 1024 chars.
742 JH/24 Verification callouts now attempt to use TLS by default.
744 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
745 are generic router options now. The defaults didn't change.
747 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
748 Original patch from Alexander Shikoff, worked over by JH.
750 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
751 files if ZCAT_COMMAND is not executable.
753 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
755 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
757 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
758 Normally benign, it bites when the pair was led to by a CNAME;
759 modern usage is to not canonicalize the domain to a CNAME target
760 (and we were inconsistent anyway for A-only vs AAAA+A).
762 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
764 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
765 when evaluating $sender_host_dnssec.
767 JH/31 Check the HELO verification lookup for DNSSEC, adding new
768 $sender_helo_dnssec variable.
770 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
772 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
774 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
776 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
777 documented as working, but never had. Support all but $spam_report.
779 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
780 added for tls authenticator.
782 HS/03 Add perl_taintmode main config option
788 TL/01 When running the test suite, the README says that variables such as
789 no_msglog_check are global and can be placed anywhere in a specific
790 test's script, however it was observed that placement needed to be near
791 the beginning for it to behave that way. Changed the runtest perl
792 script to read through the entire script once to detect and set these
793 variables, reset to the beginning of the script, and then run through
794 the script parsing/test process like normal.
796 TL/02 The BSD's have an arc4random API. One of the functions to induce
797 adding randomness was arc4random_stir(), but it has been removed in
798 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
799 function when detected.
801 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
802 cause callback expansion.
804 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
805 syntax errors in an expansion can be treated as a string instead of
806 logging or causing an error, due to the internal use of bool_lax
807 instead of bool when processing it.
809 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
810 server certificates when making smtp deliveries.
812 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
814 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
816 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
818 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
819 Merged patch from Sebastian Wiedenroth.
821 JH/05 Fix results-pipe from transport process. Several recipients, combined
822 with certificate use, exposed issues where response data items split
823 over buffer boundaries were not parsed properly. This eventually
824 resulted in duplicates being sent. This issue only became common enough
825 to notice due to the introduction of connection certificate information,
826 the item size being so much larger. Found and fixed by Wolfgang Breyha.
828 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
829 size buffer was used, resulting in syntax errors when an expansion
832 JH/07 Add support for directories of certificates when compiled with a GnuTLS
833 version 3.3.6 or later.
835 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
836 is EXPERIMENTAL_EVENT, the main-configuration and transport options
837 both become "event_action", the variables become $event_name, $event_data
838 and $event_defer_errno. There is a new variable $verify_mode, usable in
839 routers, transports and related events. The tls:cert event is now also
840 raised for inbound connections, if the main configuration event_action
843 TL/06 In test suite, disable OCSP for old versions of openssl which contained
844 early OCSP support, but no stapling (appears to be less than 1.0.0).
846 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
847 server certificate names available under the smtp transport option
848 "tls_verify_cert_hostname" now do not permit multi-component wildcard
851 JH/10 Time-related extraction expansions from certificates now use the main
852 option "timezone" setting for output formatting, and are consistent
853 between OpenSSL and GnuTLS compilations. Bug 1541.
855 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
856 encoded parameter in the incoming message. Bug 1558.
858 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
859 include certificate info, eximon was claiming there were spoolfile
862 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
864 JH/14 Log delivery-related information more consistently, using the sequence
865 "H=<name> [<ip>]" wherever possible.
867 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
868 are problematic for Debian distribution, omit them from the release
871 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
873 JH/16 Fix string representation of time values on 64bit time_t architectures.
876 JH/17 Fix a null-indirection in certextract expansions when a nondefault
877 output list separator was used.
882 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
883 checkers that were complaining about end of non-void function with no
886 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
887 This was a regression introduced in 4.83 by another bugfix.
889 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
891 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
892 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
898 TF/01 Correctly close the server side of TLS when forking for delivery.
900 When a message was received over SMTP with TLS, Exim failed to clear up
901 the incoming connection properly after forking off the child process to
902 deliver the message. In some situations the subsequent outgoing
903 delivery connection happened to have the same fd number as the incoming
904 connection previously had. Exim would try to use TLS and fail, logging
905 a "Bad file descriptor" error.
907 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
908 utilities have not been installed.
910 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
911 temporary space as the ACL may create new global variables.
913 TL/01 LDAP support uses per connection or global context settings, depending
914 upon the detected version of the libraries at build time.
916 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
917 to extract and use the src ip:port in logging and expansions as if it
918 were a direct connection from the outside internet. PPv2 support was
919 updated based on HAProxy spec change in May 2014.
921 JH/02 Add ${listextract {number}{list}{success}{fail}}.
923 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
924 Properly escape header and check for NULL return.
926 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
929 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
931 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
932 characters in header names, implemented as a verify condition.
933 Contributed by Michael Fischer v. Mollard.
935 TL/05 Rename SPF condition results err_perm and err_temp to standardized
936 results permerror and temperror. Previous values are deprecated but
937 still accepted. In a future release, err_perm and err_temp will be
938 completely removed, which will be a backward incompatibility if the
939 ACL tests for either of these two old results. Patch contributed by
940 user bes-internal on the mailing list.
942 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
944 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
945 selectors, in both main and reject logs.
947 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
950 JH/07 Add malware type "sock" for talking to simple daemon.
952 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
954 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
955 routers/transports under cutthrough routing.
957 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
958 numbers. Touch up "bool" conditional to keep the same definition.
960 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
962 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
964 JH/12 Expand items in router/transport headers_add or headers_remove lists
965 individually rather than the list as a whole. Bug 1452.
967 Required for reasonable handling of multiple headers_ options when
968 they may be empty; requires that headers_remove items with embedded
969 colons must have them doubled (or the list-separator changed).
971 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
972 view the policy declared in the DMARC record. Currently, $dmarc_status
973 is a combined value of both the record presence and the result of the
976 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
978 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
979 dnslookup router and the smtp transport (applying to the forward
982 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
983 of ldap servers used for a specific lookup. Patch provided by Heiko
986 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
987 New variable $lookup_dnssec_authenticated for observability.
989 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
990 Patch submitted by Lars Timman.
992 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
994 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
995 Requires trusted mode and valid format message id, aborts otherwise.
996 Patch contributed by Heiko Schlichting.
998 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
999 certextract with support for various fields. Bug 1358.
1001 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1002 is requested by default, modifiable by smtp transport option
1005 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1006 operate on certificate variables to give certificate fingerprints
1007 Also new ${sha256:cert_variable}.
1009 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1011 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1012 Christian Aistleitner.
1014 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1016 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1017 file. Patch from Wolfgang Breyha.
1019 JH/25 Expand the coverage of the delivery $host and $host_address to
1020 client authenticators run in verify callout. Bug 1476.
1022 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1023 align with daemon_smtp_ports. Bug 72.
1025 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1026 support and error reporting did not work properly.
1028 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1029 and is readable. Patch from Andrew Colin Kissa.
1031 TL/14 Enhance documentation of ${run expansion and how it parses the
1032 commandline after expansion, particularly in the case when an
1033 unquoted variable expansion results in an empty value.
1035 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1037 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1038 by renaming away from it. Addresses GH issue 15, reported by
1041 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1043 TL/15 SECURITY: prevent double expansion in math comparison functions
1044 (can expand unsanitized data). Not remotely exploitable.
1051 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1053 PP/02 Make -n do something, by making it not do something.
1054 When combined with -bP, the name of an option is not output.
1056 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1059 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1060 $sender_host_name and config options to manage this, and basic check
1063 PP/05 DSCP support for outbound connections and control modifier for inbound.
1065 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1066 (Only plugin which currently uses this is kerberos4, which nobody should
1067 be using, but we should make it available and other future plugins might
1068 conceivably use it, even though it would break NAT; stuff *should* be
1069 using channel bindings instead).
1071 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1072 name; added for Sendmail compatibility; requires admin caller.
1073 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1074 just ignore it); requires trusted caller.
1075 Also parse but ignore: -Ac -Am -X<logfile>
1078 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1080 TL/02 Add +smtp_confirmation as a default logging option.
1082 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1083 Patch by Magnus Holmgren from 2007-02-20.
1085 TL/04 Bugzilla 1281 - Spec typo.
1086 Bugzilla 1283 - Spec typo.
1087 Bugzilla 1290 - Spec grammar fixes.
1089 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1091 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1093 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1094 mailing list by by Dmitry Isaikin.
1096 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1098 JH/02 Support "G" suffix to numbers in ${if comparisons.
1100 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1102 NM/01 Bugzilla 1197 - Spec typo
1103 Bugzilla 1196 - Spec examples corrections
1105 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1107 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1108 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1111 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1112 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1114 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1115 "acl {{name}{arg}...}", and optional args on acl condition
1118 JH/05 Permit multiple router/transport headers_add/remove lines.
1120 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1122 JH/07 Avoid using a waiting database for a single-message-only transport.
1123 Performance patch from Paul Fisher. Bugzilla 1262.
1125 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1128 JH/09 Add $headers_added variable, with content from use of ACL modifier
1129 add_header (but not yet added to the message). Bugzilla 199.
1131 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1132 Pulled from Bugzilla 817 by Wolfgang Breyha.
1134 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1136 (nb: this is the same fix as in Exim 4.80.1)
1138 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1141 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1144 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1146 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1147 advertise SMTP AUTH mechanism to us, instead of a generic
1148 protocol violation error. Also, make Exim more robust to bad
1149 data from the Dovecot auth socket.
1151 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1153 When a queue runner is handling a message, Exim first routes the
1154 recipient addresses, during which it prunes them based on the retry
1155 hints database. After that it attempts to deliver the message to
1156 any remaining recipients. It then updates the hints database using
1159 So if a recipient address works intermittently, it can get repeatedly
1160 deferred at routing time. The retry hints record remains fresh so the
1161 address never reaches the final cutoff time.
1163 This is a fairly common occurrence when a user is bumping up against
1164 their storage quota. Exim had some logic in its local delivery code
1165 to deal with this. However it did not apply to per-recipient defers
1166 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1168 This change adds a proper retry rule check during routing so that the
1169 final cutoff time is checked against the message's age. We only do
1170 this check if there is an address retry record and there is not a
1171 domain retry record; this implies that previous attempts to handle
1172 the address had the retry_use_local_parts option turned on. We use
1173 this as an approximation for the destination being like a local
1174 delivery, as in LMTP.
1176 I suspect this new check makes the old local delivery cutoff check
1177 redundant, but I have not verified this so I left the code in place.
1179 TF/02 Correct gecos expansion when From: is a prefix of the username.
1181 Test 0254 submits a message to Exim with the header
1185 When I ran the test suite under the user fanf2, Exim expanded
1186 the header to contain my full name, whereas it should have added
1187 a Resent-Sender: header. It erroneously treats any prefix of the
1188 username as equal to the username.
1190 This change corrects that bug.
1192 GF/01 DCC debug and logging tidyup
1193 Error conditions log to paniclog rather than rejectlog.
1194 Debug lines prefixed by "DCC: " to remove any ambiguity.
1196 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1198 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1199 Bug spotted by Jeremy Harris; was flawed since initial commit.
1200 Would have resulted in OCSP responses post-SNI triggering an Exim
1201 NULL dereference and crash.
1203 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1205 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1206 Bug detection, analysis and fix by Samuel Thibault.
1207 Bugzilla 1331, Debian bug #698092.
1209 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1211 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1212 Server implementation by Todd Lyons, client by JH.
1213 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1214 config variable "prdr_enable" controls whether the server
1215 advertises the facility. If the client requests PRDR a new
1216 acl_data_smtp_prdr ACL is called once for each recipient, after
1217 the body content is received and before the acl_smtp_data ACL.
1218 The client is controlled by both of: a hosts_try_prdr option
1219 on the smtp transport, and the server advertisement.
1220 Default client logging of deliveries and rejections involving
1221 PRDR are flagged with the string "PRDR".
1223 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1224 fclose(). Diagnosis by Todd Lyons.
1226 PP/17 Update configure.default to handle IPv6 localhost better.
1227 Patch by Alain Williams (plus minor tweaks).
1230 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1231 This is now consistent with GnuTLS, and is now documented: the
1232 previous undocumented portable approach to treating the option as
1233 unset was to force an expansion failure. That still works, and
1234 an empty string is now equivalent.
1236 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1237 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1238 not performing validation itself.
1240 PP/20 Added force_command boolean option to pipe transport.
1241 Patch from Nick Koston, of cPanel Inc.
1243 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1246 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1248 PP/21 Fix eximon continuous updating with timestamped log-files.
1249 Broken in a format-string cleanup in 4.80, missed when I repaired the
1250 other false fix of the same issue.
1251 Report and fix from Heiko Schlichting.
1254 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1255 Report from Prashanth Katuri.
1257 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1258 It's SecureTransport, so affects any MacOS clients which use the
1259 system-integrated TLS libraries, including email clients.
1261 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1262 using a MIME ACL for non-SMTP local injection.
1263 Report and assistance in diagnosis by Warren Baker.
1265 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1267 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1269 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1270 last id that failed so it may be referenced in subsequent ACL's.
1272 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1275 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1276 ldap library initialization, allowing self-signed CA's to be
1277 used. Also properly sets require_cert option later in code by
1278 using NULL (global ldap config) instead of ldap handle (per
1279 session). Bug diagnosis and testing by alxgomz.
1281 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1282 the src/util/ subdirectory.
1284 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1285 renamed to Transport Post Delivery Action by Jeremy Harris, as
1288 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1289 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1290 redis_servers = needs to be configured which will be used by the redis
1291 lookup. Patch from Warren Baker, of The Packet Hub.
1293 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1295 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1296 hostname or reverse DNS when processing a host list. Used suggestions
1297 from multiple comments on this bug.
1299 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1301 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1302 Missed a few lines, added it to make the runtest require no keyboard
1305 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1306 contains upper case chars. Make router use caseful_local_part.
1308 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1309 support when GnuTLS has been built with p11-kit.
1315 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1317 This, or similar/improved, will also be change PP/11 of 4.82.
1323 PP/01 Handle short writes when writing local log-files.
1324 In practice, only affects FreeBSD (8 onwards).
1325 Bugzilla 1053, with thanks to Dmitry Isaikin.
1327 NM/01 Bugzilla 949 - Documentation tweak
1329 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1332 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1334 PP/02 Implemented gsasl authenticator.
1336 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1338 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1339 `pkg-config foo` for cflags/libs.
1341 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1342 with rest of GSASL and with heimdal_gssapi.
1344 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1345 `pkg-config foo` for cflags/libs for the TLS implementation.
1347 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1348 properties get this fed in as external SSF. A number of robustness
1349 and debugging improvements to the cyrus_sasl authenticator.
1351 PP/08 cyrus_sasl server now expands the server_realm option.
1353 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1354 Patch by Jeremy Harris.
1356 PP/10 Added dbmjz lookup type.
1358 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1360 PP/12 MAIL args handles TAB as well as SP, for better interop with
1361 non-compliant senders.
1362 Analysis and variant patch by Todd Lyons.
1364 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1365 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1366 Patch from Dirk Mueller <dmueller@suse.com>
1368 PP/13 tls_peerdn now print-escaped for spool files.
1369 Observed some $tls_peerdn in wild which contained \n, which resulted
1370 in spool file corruption.
1372 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1373 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1374 or write after TLS renegotiation, which otherwise led to messages
1377 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1378 as a tracking header (ie: a signed header comes before the signature).
1379 Patch from Wolfgang Breyha.
1381 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1382 comma-sep list; embedded commas doubled.
1384 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1386 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1388 Report and patch from Dmitry Banschikov.
1390 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1391 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1392 failures. We appear to now support TLS1.1+ with Exim.
1394 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1395 lets Exim select keys and certificates based upon TLS SNI from client.
1396 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1397 before an outbound SMTP session. New log_selector, +tls_sni.
1399 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1400 NULL dereference. Report and patch from Alun Jones.
1402 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1403 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1404 Not seeing resolver debug output on NetBSD, but suspect this is a
1405 resolver implementation change.
1407 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1408 Left warnings. Added "eximon gdb" invocation mode.
1410 PP/21 Defaulting "accept_8bitmime" to true, not false.
1412 PP/22 Added -bw for inetd wait mode support.
1414 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1415 locate the relevant includes and libraries. Made this the default.
1417 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1418 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1420 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1421 This may cause build issues on older platforms.
1423 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1424 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1425 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1426 Added SNI support via GnuTLS too.
1427 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1429 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1431 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1432 Applied second patch from Janne, implementing suggestion to default
1433 multiple-strings-in-record handling to match SPF spec.
1435 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1437 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1438 read-only, out of scope).
1439 Patch from Wolfgang Breyha, report from Stuart Northfield.
1441 PP/29 Fix three issues highlighted by clang analyser static analysis.
1442 Only crash-plausible issue would require the Cambridge-specific
1443 iplookup router and a misconfiguration.
1444 Report from Marcin Mirosław.
1446 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1448 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1449 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1450 As part of this, removing so much warning spew let me fix some minor
1451 real issues in debug logging.
1453 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1454 assignment on my part. Fixed.
1456 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1457 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1458 Janne Snabb (who went above and beyond: thank you).
1460 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1461 string otherwise requires a connection and a bunch more work and it's
1462 relatively easy to get wrong. Should also expose TLS library linkage
1465 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1466 64-bit ${eval} (JH/03).
1468 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1469 GNU libc to support some of the 64-bit stuff, should not lead to
1470 conflicts. Defined before os.h is pulled in, so if a given platform
1471 needs to override this, it can.
1473 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1474 protection layer was required, which is not implemented.
1475 Bugzilla 1254, patch from Wolfgang Breyha.
1477 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1478 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1479 tls_dhparam take prime identifiers. Also unbreak combination of
1480 OpenSSL+DH_params+TLSSNI.
1482 PP/39 Disable SSLv2 by default in OpenSSL support.
1488 PP/01 Solaris build fix for Oracle's LDAP libraries.
1489 Bugzilla 1109, patch from Stephen Usher.
1491 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1493 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1496 TF/02 Fix a couple more cases where we did not log the error message
1497 when unlink() failed. See also change 4.74-TF/03.
1499 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1500 lock up or crash if it happened to be inside a call to libc when it
1501 got a SIGUSR1 from exiwhat.
1503 The SIGUSR1 handler appends the current process status to the process
1504 log which is later printed by exiwhat. It used to use the general
1505 purpose logging code to do this, but several functions it calls are
1506 not safe for signals.
1508 The new output code in the SIGUSR1 handler is specific to the process
1509 log, and simple enough that it's easy to inspect for signal safety.
1510 Removing some special cases also simplifies the general logging code.
1511 Removing the spurious timestamps from the process log simplifies
1514 TF/04 Improved ratelimit ACL condition.
1516 The /noupdate option has been deprecated in favour of /readonly which
1517 has clearer semantics. The /leaky, /strict, and /readonly update modes
1518 are mutually exclusive. The update mode is no longer included in the
1519 database key; it just determines when the database is updated. (This
1520 means that when you upgrade Exim will forget old rate measurements.)
1522 Exim now checks that the per_* options are used with an update mode that
1523 makes sense for the current ACL. For example, when Exim is processing a
1524 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1525 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1526 must specify per_mail/readonly. If you omit the update mode it defaults to
1527 /leaky where that makes sense (as before) or /readonly where required.
1529 The /noupdate option is now undocumented but still supported for
1530 backwards compatibility. It is equivalent to /readonly except that in
1531 ACLs where /readonly is required you may specify /leaky/noupdate or
1532 /strict/noupdate which are treated the same as /readonly.
1534 A useful new feature is the /count= option. This is a generalization
1535 of the per_byte option, so that you can measure the throughput of other
1536 aggregate values. For example, the per_byte option is now equivalent
1537 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1539 The per_rcpt option has been generalized using the /count= mechanism
1540 (though it's more complicated than the per_byte equivalence). When it is
1541 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1542 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1543 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1544 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1545 Note that using per_rcpt with a non-readonly update mode in more than
1546 one ACL will cause the recipients to be double-counted. (The per_mail
1547 and per_byte options don't have this problem.)
1549 The handling of very low rates has changed slightly. If the computed rate
1550 is less than the event's count (usually one) then this event is the first
1551 after a long gap. In this case the rate is set to the same as this event's
1552 count, so that the first message of a spam run is counted properly.
1554 The major new feature is a mechanism for counting the rate of unique
1555 events. The new per_addr option counts the number of different
1556 recipients that someone has sent messages to in the last time period. It
1557 behaves like per_rcpt if all the recipient addresses are different, but
1558 duplicate recipient addresses do not increase the measured rate. Like
1559 the /count= option this is a general mechanism, so the per_addr option
1560 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1561 example, measure the rate that a client uses different sender addresses
1562 with the options per_mail/unique=$sender_address. There are further
1563 details in the main documentation.
1565 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1567 TF/06 Removed a few PCRE remnants.
1569 TF/07 Automatically extract Exim's version number from tags in the git
1570 repository when doing development or release builds.
1572 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1573 Bugzilla 879. Patch from Paul Fisher.
1575 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1576 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1579 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1581 PP/05 Variable $av_failed, true if the AV scanner deferred.
1582 Bugzilla 1078. Patch from John Horne.
1584 PP/06 Stop make process more reliably on build failure.
1585 Bugzilla 1087. Patch from Heiko Schlittermann.
1587 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1588 Bugzilla 1089. Patch from Heiko Schlittermann.
1590 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1591 Bugzilla 1131. Patch from Holger Weiß.
1593 PP/09 Handle IPv6 addresses with SPF.
1594 Bugzilla 860. Patch from Wolfgang Breyha.
1596 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1598 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1601 PP/11 match_* no longer expand right-hand-side by default.
1602 New compile-time build option, EXPAND_LISTMATCH_RHS.
1603 New expansion conditions, "inlist", "inlisti".
1605 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1607 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1609 PP/14 fix log_write() format string regression from TF/03.
1610 Bugzilla 1152. Patch from Dmitry Isaikin.
1616 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1618 PP/02 Harmonised TLS library version reporting; only show if debugging.
1619 Layout now matches that introduced for other libraries in 4.74 PP/03.
1621 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1623 PP/04 New "dns_use_edns0" global option.
1625 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1628 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1629 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1631 TK/01 Updated PolarSSL code to 0.14.2.
1632 Bugzilla 1097. Patch from Andreas Metzler.
1634 PP/07 Catch divide-by-zero in ${eval:...}.
1635 Fixes bugzilla 1102.
1637 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1640 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1641 format-string attack -- SECURITY: remote arbitrary code execution.
1643 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1644 time unintentionally subject to list matching rules, letting the header
1645 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1646 arbitrary string expansion). This allowed for information disclosure.
1648 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1649 INT_MIN/-1 -- value coerced to INT_MAX.
1655 NM/01 Workaround for PCRE version dependency in version reporting
1658 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1659 This fixes portability to compilers other than gcc, notably
1660 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1662 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1663 makefiles for portability to HP-UX and POSIX correctness.
1665 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1666 Also via indented variable definition in the Makefile.
1667 (Debugging by Oliver Heesakkers).
1669 PP/02 Restore caching of spamd results with expanded spamd_address.
1670 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1672 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1673 Improves build reliability. Fix from: Frank Elsner
1675 NM/02 Fix wide character breakage in the rfc2047 coding
1676 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1678 NM/03 Allow underscore in dnslist lookups
1679 Fixes bug 1026. Patch from Graeme Fowler
1681 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1682 Code patches from Adam Ciarcinski of NetBSD.
1684 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1687 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1688 is logged, to avoid truncation. Patch from John Horne.
1690 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1691 Patch from Jakob Hirsch.
1693 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1694 SQL string expansion failure details.
1695 Patch from Andrey Oktyabrski.
1697 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1698 Patch from Simon Arlott.
1700 PP/09 New lookups functionality failed to compile on old gcc which rejects
1701 extern declarations in function scope.
1702 Patch from Oliver Fleischmann
1704 PP/10 Use sig_atomic_t for flags set from signal handlers.
1705 Check getgroups() return and improve debugging.
1706 Fixed developed for diagnosis in bug 927 (which turned out to be
1709 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1710 Patch from Mark Zealey.
1712 PP/12 Bugzilla 1056: Improved spamd server selection.
1713 Patch from Mark Zealey.
1715 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1716 Based on patch from Heiko Schlittermann.
1718 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1719 Patch from Uwe Doering, sign-off by Michael Haardt.
1721 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1722 variable declaration deep within a block. Bug and patch from
1725 PP/15 lookups-Makefile IRIX compatibility coercion.
1727 PP/16 Make DISABLE_DKIM build knob functional.
1729 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1730 Patch by Simon Arlott
1732 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1733 variable argument macros. Our copy now differs from upstream.
1739 TF/01 Failure to get a lock on a hints database can have serious
1740 consequences so log it to the panic log.
1742 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1743 controlled using the smtp_confirmation log selector.
1745 TF/03 Include the error message when we fail to unlink a spool file.
1747 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1748 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1749 for maintaining out-of-tree patches for some time.
1751 PP/01 Bugzilla 139: Documentation and portability issues.
1752 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1753 Handle per-OS dynamic-module compilation flags.
1755 PP/02 Let /dev/null have normal permissions.
1756 The 4.73 fixes were a little too stringent and complained about the
1757 permissions on /dev/null. Exempt it from some checks.
1758 Reported by Andreas M. Kirchwitz.
1760 PP/03 Report version information for many libraries, including
1761 Exim version information for dynamically loaded libraries. Created
1762 version.h, now support a version extension string for distributors
1763 who patch heavily. Dynamic module ABI change.
1765 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1766 privilege escalation vulnerability whereby the Exim run-time user
1767 can cause root to append content of the attacker's choosing to
1770 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1773 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1774 If dropping privileges for untrusted macros, we disabled normal logging
1775 on the basis that it would fail; for the Exim run-time user, this is not
1776 the case, and it resulted in successful deliveries going unlogged.
1777 Fixed. Reported by Andreas Metzler.
1783 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1784 only prepend for the Resent-* case. Fixes regression introduced in
1785 Exim 4.70 by NM/22 for Bugzilla 607.
1787 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1788 increasing numbers of administrators be bitten by this.
1790 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1792 PP/03 Bugzilla 994: added openssl_options main configuration option.
1794 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1796 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1798 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1800 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1801 without a peer certificate, leading to a segfault because of an
1802 assumption that peers always have certificates. Be a little more
1803 paranoid. Problem reported by Martin Tscholak.
1805 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1806 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1807 NB: ClamAV planning to remove STREAM in "middle of 2010".
1808 CL also introduces -bmalware, various -d+acl logging additions and
1809 more caution in buffer sizes.
1811 PP/09 Implemented reverse_ip expansion operator.
1813 PP/10 Bugzilla 937: provide a "debug" ACL control.
1815 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1817 PP/12 Bugzilla 973: Implement --version.
1819 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1821 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1823 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1825 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1826 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1827 ignore trailing whitespace.
1829 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1831 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1834 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1835 Notification from Dr Andrew Aitchison.
1837 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1838 ExtendedDetectionInfo response format.
1839 Notification from John Horne.
1841 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1844 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1845 XSL and documented dependency on system catalogs, with examples of how
1848 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1851 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1852 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1853 configuration file which is writeable by the Exim user or group.
1855 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1856 of configuration files to cover files specified with the -C option if
1857 they are going to be used with root privileges, not just the default
1860 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1861 option (effectively making it always true).
1863 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1864 files to be used while preserving root privileges.
1866 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1867 that rogue child processes cannot use them.
1869 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1870 run-time user, instead of root.
1872 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1873 Exim run-time user without dropping privileges.
1875 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1876 result string, instead of calling string_vformat() twice with the same
1879 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1880 for other users. Others should always drop root privileges if they use
1881 -C on the command line, even for a whitelisted configure file.
1883 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1885 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1891 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1892 $data_path, and $header_path variables; fixed documentation bugs and
1895 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1896 exipick to access non-standard spools, including the "frozen" queue
1899 NM/01 Bugzilla 965: Support mysql stored procedures.
1900 Patch from Alain Williams
1902 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1904 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1905 Patch from Andreas Metzler
1907 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1908 Patch from Kirill Miazine
1910 NM/05 Bugzilla 671: Added umask to procmail example.
1912 JJ/03 installed exipick 20100323.0, fixing doc bug
1914 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1915 directory. Notification and patch from Dan Rosenberg.
1917 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1919 TK/02 Improve log output when DKIM signing operation fails.
1921 MH/01 Treat the transport option dkim_domain as a colon separated
1922 list, not as a single string, and sign the message with each element,
1923 omitting multiple occurences of the same signer.
1925 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1926 Bugzilla 985, 986. Patch by Simon Arlott
1928 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1929 Patch by Simon Arlott
1931 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1932 MBX locking. Notification from Dan Rosenberg.
1938 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1940 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1942 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1944 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1946 NM/04 Bugzilla 915: Flag broken perl installation during build.
1952 TK/01 Added patch by Johannes Berg that expands the main option
1953 "spamd_address" if it starts with a dollar sign.
1955 TK/02 Write list of recipients to X-Envelope-Sender header when building
1956 the mbox-format spool file for content scanning (suggested by Jakob
1959 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1960 (http://www.dcc-servers.net/) support via dccifd. Activated by
1961 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1963 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1964 by Mark Daniel Reidel <mr@df.eu>.
1966 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1967 When building exim an external PCRE library is now needed -
1968 PCRE is a system library on the majority of modern systems.
1969 See entry on PCRE_LIBS in EDITME file.
1971 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1972 conversation. Added nologin parameter to request.
1973 Patch contributed by Kirill Miazine.
1975 TF/01 Do not log submission mode rewrites if they do not change the address.
1977 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1979 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1980 log files in place. Contributed by Roberto Lima.
1982 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1984 TF/03 Bugzilla 615: When checking the local_parts router precondition
1985 after a local_part_suffix or local_part_prefix option, Exim now
1986 does not use the address's named list lookup cache, since this
1987 contains cached lookups for the whole local part.
1989 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1990 Robert Millan. Documentation is in experimental-spec.txt.
1992 TF/04 Bugzilla 668: Fix parallel build (make -j).
1994 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1996 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1997 Patch provided by Jan Srzednicki.
1999 TF/05 Leading white space used to be stripped from $spam_report which
2000 wrecked the formatting. Now it is preserved.
2002 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2003 that they are available at delivery time.
2005 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2007 TF/08 TLS error reporting now respects the incoming_interface and
2008 incoming_port log selectors.
2010 TF/09 Produce a more useful error message if an SMTP transport's hosts
2011 setting expands to an empty string.
2013 NM/06 Bugzilla 744: EXPN did not work under TLS.
2014 Patch provided by Phil Pennock.
2016 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2017 Patch provided by Richard Godbee.
2019 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2020 acl_smtp_notquit, added index entry.
2022 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2023 Patch provided by Eugene Bujak.
2025 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2026 accept(). Patch provided by Maxim Dounin.
2028 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2029 Patch provided by Phil Pennock.
2031 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2033 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2034 Patch provided by Brad "anomie" Jorsch.
2036 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2037 Patch provided by Dean Brooks.
2039 TK/05 Add native DKIM support (does not depend on external libraries).
2041 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2042 Patch provided by Graeme Fowler.
2044 NM/16 Bugzilla 851: Documentation example syntax fix.
2046 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2048 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2051 NM/19 Bugzilla 745: TLS version reporting.
2052 Patch provided by Phil Pennock.
2054 NM/20 Bugzilla 167: bool: condition support.
2055 Patch provided by Phil Pennock.
2057 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2058 clients. Patch provided by Phil Pennock.
2060 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2061 Patch provided by Brad "anomie" Jorsch.
2063 NM/23 Bugzilla 687: Fix misparses in eximstats.
2064 Patch provided by Heiko Schlittermann.
2066 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2067 Patch provided by Heiko Schlittermann.
2069 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2070 plus update to original patch.
2072 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2074 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2075 Patch provided by David Brownlee.
2077 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2079 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2081 NM/30 Bugzilla 888: TLS documentation bugfixes.
2083 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2085 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2086 Unlike the original bugzilla I have changed all shell scripts in src tree.
2088 NM/33 Bugzilla 898: Transport filter timeout fix.
2089 Patch by Todd Rinaldo.
2091 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2092 Patch by Serge Demonchaux.
2094 NM/35 Bugzilla 39: Base64 decode bug fixes.
2095 Patch by Jakob Hirsch.
2097 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2099 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2101 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2103 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2109 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2110 ALT-N's libdkim that I have put here:
2111 http://duncanthrax.net/exim-experimental/
2113 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2114 were called 'true' and it seems that C99 defines that as a reserved
2115 keyword to be used with 'bool' variable types. That means you could
2116 not include C99-style headers which use bools without triggering
2117 build errors in sieve.c.
2119 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2120 as mailq or other aliases. Changed the --help handling significantly
2121 to do whats expected. exim_usage() emits usage/help information.
2123 SC/01 Added the -bylocaldomain option to eximstats.
2125 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2127 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2129 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2135 PH/01 Another patch from the Sieve maintainer.
2137 PH/02 When an IPv6 address is converted to a string for single-key lookup
2138 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2139 dots are used instead of colons so that keys in lsearch files need not
2140 contain colons. This was done some time before quoting was made available
2141 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2142 (notated using the quote facility) so as to distinguish them from IPv4
2143 keys. This meant that lookups for IP addresses in host lists did not work
2144 for iplsearch lookups.
2146 This has been fixed by arranging for IPv6 addresses to be expressed with
2147 colons if the lookup type is iplsearch. This is not incompatible, because
2148 previously such lookups could never work.
2150 The situation is now rather anomalous, since one *can* have colons in
2151 ordinary lsearch keys. However, making the change in all cases is
2152 incompatible and would probably break a number of configurations.
2154 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2157 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2158 conversion specification without a maximum field width, thereby enabling
2159 a rogue spamd server to cause a buffer overflow. While nobody in their
2160 right mind would setup Exim to query an untrusted spamd server, an
2161 attacker that gains access to a server running spamd could potentially
2162 exploit this vulnerability to run arbitrary code as the Exim user.
2164 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2165 $primary_hostname instead of what libspf2 thinks the hosts name is.
2167 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2168 a directory entry by the name of the lookup key. Previously, if a
2169 symlink pointed to a non-existing file or a file in a directory that
2170 Exim lacked permissions to read, a lookup for a key matching that
2171 symlink would fail. Now it is enough that a matching directory entry
2172 exists, symlink or not. (Bugzilla 503.)
2174 PH/03 The body_linecount and body_zerocount variables are now exported in the
2177 PH/04 Added the $dnslist_matched variable.
2179 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2180 This means they are set thereafter only if the connection becomes
2183 PH/06 Added the client_condition to authenticators so that some can be skipped
2184 by clients under certain conditions.
2186 PH/07 The error message for a badly-placed control=no_multiline_responses left
2187 "_responses" off the end of the name.
2189 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2191 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2192 (without spaces) instead of just copying the configuration text.
2194 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2196 PH/11 Added $max_received_linelength.
2198 PH/12 Added +ignore_defer and +include_defer to host lists.
2200 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2201 way in which PCRE > 7.0 is built.
2203 PH/14 Implemented queue_only_load_latch.
2205 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2206 MAIL command. The effect was to mangle the value on 64-bit systems.
2208 PH/16 Another patch from the Sieve maintainer.
2210 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2212 PH/18 If a system quota error occurred while trying to create the file for
2213 a maildir delivery, the message "Mailbox is full" was not appended to the
2214 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2215 only to a quota excession during the actual writing of the file.
2217 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2218 characters?) which causes problems in log lines. The DN values are now
2219 passed through string_printing() before being added to log lines.
2221 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2222 and InterBase are left for another time.)
2224 PH/21 Added message_body_newlines option.
2226 PH/22 Guard against possible overflow in moan_check_errorcopy().
2228 PH/23 POSIX allows open() to be a macro; guard against that.
2230 PH/24 If the recipient of an error message contained an @ in the local part
2231 (suitably quoted, of course), incorrect values were put in $domain and
2232 $local_part during the evaluation of errors_copy.
2238 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2239 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2242 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2243 issue a MAIL command.
2245 PH/02 In an ACL statement such as
2247 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2249 if a client was not listed at all, or was listed with a value other than
2250 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2251 the condition was not true (as it should be), so access was not denied.
2252 The bug was that the ! inversion was incorrectly passed on to the second
2253 item. This has been fixed.
2255 PH/03 Added additional dnslists conditions == and =& which are different from
2256 = and & when the dns lookup returns more than one IP address.
2258 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2259 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2261 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2262 FSYNC, which compiles an option called disable_fsync that allows for
2263 bypassing fsync(). The documentation is heavily laced with warnings.
2265 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2267 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2268 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2269 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2270 including adding "make clean"; (3) Added -fPIC when compiling the test
2271 dynamically loaded module, to get rid of a warning.
2273 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2274 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2275 = 0s. The bug is otherwise harmless.
2277 PH/07 There was a bug in the dovecot authenticator such that the value of
2278 $auth1 could be overwritten, and so not correctly preserved, after a
2279 successful authentication. This usually meant that the value preserved by
2280 the server_setid option was incorrect.
2282 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2284 PH/09 Installed PCRE release 7.0.
2286 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2287 run for batched SMTP input. It is now run at the start of every message
2288 in the batch. While fixing this I discovered that the process information
2289 (output by running exiwhat) was not always getting set for -bs and -bS
2290 input. This is fixed, and it now also says "batched" for BSMTP.
2292 PH/11 Added control=no_pipelining.
2294 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2295 patch, slightly modified), and move the expansion of helo_data till after
2296 the connection is made in the smtp transport (so it can use these
2299 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2301 PH/14 Added log_selector = +pid.
2303 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2305 PH/16 Add ${if forany and ${if forall.
2307 PH/17 Added dsn_from option to vary the From: line in DSNs.
2309 PH/18 Flush SMTP output before performing a callout, unless control =
2310 no_callout_flush is set.
2312 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2313 was true (the default) a successful delivery failed to delete the retry
2314 item, thus causing premature timeout of the address. The bug is now
2317 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2319 PH/21 Long custom messages for fakedefer and fakereject are now split up
2320 into multiline responses in the same way that messages for "deny" and
2321 other ACL rejections are.
2323 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2324 with slight modification.
2326 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2327 draft, changing the syntax and factoring some duplicate code".
2329 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2330 for deliveries of the second and subsequent messages over the same SMTP
2333 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2334 ${reduce, with only minor "tidies".
2336 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2338 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2339 expansion side effects.
2341 PH/27 When a message times out after an over-quota error from an Exim-imposed
2342 quota, the bounce message says "mailbox is full". This message was not
2343 being given when it was a system quota that was exceeded. It now should
2346 MH/03 Made $recipients available in local_scan(). local_scan() already has
2347 better access to the recipient list through recipients_list[], but
2348 $recipients can be useful in postmaster-provided expansion strings.
2350 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2351 in the case of a MAIL command with additional options following the
2352 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2353 were accidentally chopped off.
2355 PH/29 SMTP synchronization checks are implemented when a command is read -
2356 there is a check that no more input is waiting when there shouldn't be
2357 any. However, for some commands, a delay in an ACL can mean that it is
2358 some time before the response is written. In this time, more input might
2359 arrive, invalidly. So now there are extra checks after an ACL has run for
2360 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2361 pipelining has not been advertised.
2363 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2365 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2366 returned in the SMTP response when smtp_return_error_details was false.
2367 This has been fixed.
2369 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2370 instead of the C I/O that was originally supplied, because problems were
2371 reported on Solaris.
2373 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2374 Exim which did not show up earlier: it was assuming that a call to
2375 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2376 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2377 was a macro that became an assignment, so it seemed to work. This has
2378 changed to a proper function call with a void return, hence the compile
2379 error. Exim's code has been fixed.
2381 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2384 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2386 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2388 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2389 because it turns out to be too big or there is a timeout) there is no
2390 "Completed" line in the log. When some messages of this type were
2391 selected by exigrep, they were listed as "not completed". Others were
2392 picked up by some special patterns. I have improved the selection
2393 criteria to be more general.
2395 PH/38 The host_find_failed option in the manualroute router can now be set
2396 to "ignore", to completely ignore a host whose IP address cannot be
2397 found. If all hosts are ignored, the behaviour is controlled by the new
2398 host_all_ignored option.
2400 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2401 homing or because of multiple MX records with /mx) generated more than
2402 one IP address, and the following item turned out to be the local host,
2403 all the secondary addresses of the first item were incorrectly removed
2404 from the list, along with the local host and any following hosts (which
2405 is what is supposed to happen).
2407 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2408 whoever called Exim into the -H file. In the case of the daemon it was
2409 behaving confusingly. When first started, it used values for whoever
2410 started the daemon, but after a SIGHUP it used the Exim user (because it
2411 calls itself on a restart). I have changed the code so that it now always
2414 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2415 message are rejected with the same error (e.g. no authentication or bad
2416 sender address), and a DATA command is nevertheless sent (as can happen
2417 with PIPELINING or a stupid MUA), the error message that was given to the
2418 RCPT commands is included in the rejection of the DATA command. This is
2419 intended to be helpful for MUAs that show only the final error to their
2422 PH/42 Another patch from the Sieve maintainer.
2424 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2425 Eximstats - Fixed some broken HTML links and added missing column headers
2427 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2428 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2430 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2432 PH/43 Yet another patch from the Sieve maintainer.
2434 PH/44 I found a way to check for a TCP/IP connection going away before sending
2435 the response to the final '.' that terminates a message, but only in the
2436 case where the client has not sent further data following the '.'
2437 (unfortunately, this is allowed). However, in many cases there won't be
2438 any further data because there won't be any more messages to send. A call
2439 to select() can be used: if it shows that the input is "ready", there is
2440 either input waiting, or the socket has been closed. An attempt to read
2441 the next input character can distinguish the two cases. Previously, Exim
2442 would have sent an OK response which the client would never have see.
2443 This could lead to message repetition. This fix should cure that, at
2444 least in a lot of common cases.
2446 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2447 advertised in response to EHLO.
2453 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2454 fixed by 4.65/MH/01 (is this a record?) are fixed:
2456 (i) An empty string was always treated as zero by the numeric comparison
2457 operators. This behaviour has been restored.
2459 (ii) It is documented that the numeric comparison operators always treat
2460 their arguments as decimal numbers. This was broken in that numbers
2461 starting with 0 were being interpreted as octal.
2463 While fixing these problems I realized that there was another issue that
2464 hadn't been noticed. Values of message_size_limit (both the global option
2465 and the transport option) were treated as octal if they started with 0.
2466 The documentation was vague. These values are now always treated as
2467 decimal, and I will make that clear in the documentation.
2473 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2474 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2477 MH/01 Don't check that the operands of numeric comparison operators are
2478 integers when their expansion is in "skipping" mode (fixes bug
2479 introduced by 4.64-PH/07).
2481 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2482 child addresses, Exim now panics and dies. Previously, because the count
2483 is held in a short int, deliveries were likely to be lost. As such a
2484 large number of recipients for a single message is ridiculous
2485 (performance will be very, very poor), I have chosen to impose a limit
2486 rather than extend the field.
2492 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2493 leftover -K file (the existence of which was triggered by #402).
2494 While we were at it, introduced process PID as part of the -K
2495 filename. This should rule out race conditions when creating
2498 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2499 processing considerably. Previous code took too long for large mails,
2500 triggering a timeout which in turn triggers #401.
2502 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2503 in the DK code in transports.c. sendfile() is not really portable,
2504 hence the _LINUX specificness.
2506 TF/01 In the add_headers option to the mail command in an Exim filter,
2507 there was a bug that Exim would claim a syntax error in any
2508 header after the first one which had an odd number of characters
2511 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2512 callout verification, Exim cached a "reject" for the entire domain. This
2513 is correct for most verifications, but it is not correct for a recipient
2514 verification with use_sender or use_postmaster set, because in that case
2515 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2516 case of MAIL FROM:<> rejection from other early rejections (e.g.
2517 rejection of HELO). When verifying a recipient using a non-null MAIL
2518 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2519 Whatever the result of the callout, the value of the domain cache is
2520 left unchanged (for any other kind of callout, getting as far as trying
2521 RCPT means that the domain itself is ok).
2523 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2526 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2527 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2528 session, the daemon ignores ECONNECT errors and logs others; it now
2529 ignores EPIPE as well.
2531 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2532 (quoted-printable decoding).
2534 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2535 later the small subsequent patch to fix an introduced bug.
2537 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2539 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2541 PH/08 An error is now given if message_size_limit is specified negative.
2543 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2544 to be given (somewhat) arbitrary names.
2546 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2549 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2550 miscellaneous code fixes
2552 PH/10 Added the log_reject_target ACL modifier to specify where to log
2555 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2556 hostname. This is wrong, because it relates to the incoming message (and
2557 probably the interface on which it is arriving) and not to the outgoing
2558 callout (which could be using a different interface). This has been
2559 changed to use the value of the helo_data option from the smtp transport
2560 instead - this is what is used when a message is actually being sent. If
2561 there is no remote transport (possible with a router that sets up host
2562 addresses), $smtp_active_hostname is used.
2564 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2565 tweaks were necessary in order to get it to work (see also 21 below):
2566 (a) The code assumed that strncpy() returns a negative number on buffer
2567 overflow, which isn't the case. Replaced with Exim's string_format()
2569 (b) There were several signed/unsigned issues. I just did the minimum
2570 hacking in of casts. There is scope for a larger refactoring.
2571 (c) The code used strcasecmp() which is not a standard C function.
2572 Replaced with Exim's strcmpic() function.
2573 (d) The code set only $1; it now sets $auth1 as well.
2574 (e) A simple test gave the error "authentication client didn't specify
2575 service in request". It would seem that Dovecot has changed its
2576 interface. Fortunately there's a specification; I followed it and
2577 changed what the client sends and it appears to be working now.
2579 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2582 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2583 address A is aliased to B and C, where B exists and C does not. Without
2584 -v the output is "A verified" because verification stops after a
2585 successful redirection if more than one address is generated. However,
2586 with -v the child addresses are also verified. Exim was outputting "A
2587 failed to verify" and then showing the successful verification for C,
2588 with its parentage. It now outputs "B failed to verify", showing B's
2589 parentage before showing the successful verification of C.
2591 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2592 look up a TXT record in a specific list after matching in a combined
2595 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2596 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2597 they consult the DNS. I had assumed they would set it the way they
2598 wanted; and indeed my experiments on Linux seem to show that in some
2599 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2600 To be on the safe side, however, I have now made the interface to
2601 host_find_byname() similar to host_find_bydns(), with an argument
2602 containing the DNS resolver options. The host_find_byname() function now
2603 sets these options at its start, just as host_find_bydns() does. The smtp
2604 transport options dns_qualify_single and dns_search_parents are passed to
2605 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2606 of host_find_byname() use the default settings of RES_DEFNAMES
2607 (qualify_single) but not RES_DNSRCH (search_parents).
2609 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2610 spool_read_header() do less string testing, by means of a preliminary
2611 switch on the second character of optional "-foo" lines. (This is
2612 overdue, caused by the large number of possibilities that now exist.
2613 Originally there were few.) While I was there, I also converted the
2614 str(n)cmp tests so they don't re-test the leading "-" and the first
2615 character, in the hope this might squeeze out yet more improvement.
2617 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2618 flag allowing group syntax was set by the header_syntax check but not
2619 turned off, possible causing trouble later; (2) The flag was not being
2620 set at all for the header_verify test, causing "group"-style headers to
2621 be rejected. I have now set it in this case, and also caused header_
2622 verify to ignore an empty address taken from a group. While doing this, I
2623 came across some other cases where the code for allowing group syntax
2624 while scanning a header line wasn't quite right (mostly, not resetting
2625 the flag correctly in the right place). These bugs could have caused
2626 trouble for malformed header lines. I hope it is now all correct.
2628 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2629 with the "reply" argument non-NULL. The code, however (which originally
2630 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2631 but it didn't always do it. This confused somebody who was copying the
2632 code for some other use. I have removed all the tests.
2634 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2635 feature that was used to support insecure browsers during the U.S. crypto
2636 embargo. It requires special client support, and Exim is probably the
2637 only MTA that supported it -- and would never use it because real RSA is
2638 always available. This code has been removed, because it had the bad
2639 effect of slowing Exim down by computing (never used) parameters for the
2640 RSA_EXPORT functionality.
2642 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2643 authenticator to fail if there's a tab character in the incoming data
2644 (there should never be unless someone is messing about, as it's supposed
2645 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2646 if the connection is using TLS or if the remote IP is the same as the
2647 local IP, and the "valid-client-cert option" if a client certificate has
2650 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2651 authenticators. This can be used for authorization after authentication
2652 succeeds. (In the case of plaintext, it servers for both authentication
2655 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2656 if any retry times were supplied.
2658 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2659 connection, where there is no client IP address to check. In this
2660 situation, the verify now always succeeds.
2662 PH/25 Applied John Jetmore's -Mset patch.
2664 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2666 PH/27 In a string expansion for a processed (not raw) header when multiple
2667 headers of the same name were present, leading whitespace was being
2668 removed from all of them, but trailing whitespace was being removed only
2669 from the last one. Now trailing whitespace is removed from each header
2670 before concatenation. Completely empty headers in a concatenation (as
2671 before) are ignored.
2673 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2674 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2676 PH/29 [Removed. This was a change that I later backed out, and forgot to
2677 correct the ChangeLog entry (that I had efficiently created) before
2678 committing the later change.]
2680 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2681 address errors (4xx response to RCPT) over the same connection as other
2682 messages routed to the same hosts. Such deliveries are always "forced",
2683 so retry times are not inspected. This resulted in far too many retries
2684 for the affected addresses. The effect occurred only when there were more
2685 hosts than the hosts_max_try setting in the smtp transport when it had
2686 the 4xx errors. Those hosts that it had tried were not added to the list
2687 of hosts for which the message was waiting, so if all were tried, there
2688 was no problem. Two fixes have been applied:
2690 (i) If there are any address or message errors in an SMTP delivery, none
2691 of the hosts (tried or untried) are now added to the list of hosts
2692 for which the message is waiting, so the message should not be a
2693 candidate for sending over the same connection that was used for a
2694 successful delivery of some other message. This seems entirely
2695 reasonable: after all the message is NOT "waiting for some host".
2696 This is so "obvious" that I'm not sure why it wasn't done
2697 previously. Hope I haven't missed anything, but it can't do any
2698 harm, as the worst effect is to miss an optimization.
2700 (ii) If, despite (i), such a delivery is accidentally attempted, the
2701 routing retry time is respected, so at least it doesn't keep
2702 hammering the server.
2704 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2705 in ${readsocket because some servers need this prod.
2707 PH/32 Added some extra debug output when updating a wait-xxx database.
2709 PH/33 The hint "could be header name not terminated by colon", which has been
2710 given for certain expansion errors for a long time, was not being given
2711 for the ${if def:h_colon_omitted{... case.
2713 PH/34 The spec says: "With one important exception, whenever a domain list is
2714 being scanned, $domain contains the subject domain." There was at least
2715 one case where this was not true.
2717 PH/35 The error "getsockname() failed: connection reset by peer" was being
2718 written to the panic log as well as the main log, but it isn't really
2719 panic-worthy as it just means the connection died rather early on. I have
2720 removed the panic log writing for the ECONNRESET error when getsockname()
2723 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2724 runs only) independently of the message's sender address. This meant
2725 that, if the 4xx error was in fact related to the sender, a different
2726 message to the same recipient with a different sender could confuse
2727 things. In particular, this can happen when sending to a greylisting
2728 server, but other circumstances could also provoke similar problems.
2729 I have changed the default so that the retry time for these errors is now
2730 based a combination of the sender and recipient addresses. This change
2731 can be overridden by setting address_retry_include_sender=false in the
2734 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2735 remote server are returned as part of bounce messages. This was not
2736 happening for LMTP over a pipe (the lmtp transport), but now it is the
2737 same for both kinds of LMTP.
2739 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2740 in header lines that were in fact CNAMEs. This is no longer the case.
2742 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2743 and queue runs started by the daemon processed all messages. This has
2744 been fixed so that -R and -S can now usefully be given with -q<time>.
2746 PH/40 Import PCRE release 6.7 (fixes some bugs).
2748 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2750 PH/42 Give an error if -q is specified more than once.
2752 PH/43 Renamed the variables $interface_address and $interface_port as
2753 $received_ip_address and $received_port, to make it clear that these
2754 values apply to message reception, and not to the outgoing interface when
2755 a message is delivered. (The old names remain recognized, of course.)
2757 PH/44 There was no timeout on the connect() call when using a Unix domain
2758 socket in the ${readsocket expansion. There now is.
2760 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2761 be meaningful with "accept".
2763 SC/01 Eximstats V1.43
2764 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2766 SC/02 Eximstats V1.44
2767 Use a glob alias rather than an array ref in the generated
2768 parser. This improves both readability and performance.
2770 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2771 Collect SpamAssassin and rejection statistics.
2772 Don't display local sender or destination tables unless
2773 there is data to show.
2774 Added average volumes into the top table text output.
2776 SC/04 Eximstats V1.46
2777 Collect data on the number of addresses (recipients)
2778 as well as the number of messages.
2780 SC/05 Eximstats V1.47
2781 Added 'Message too big' to the list of mail rejection
2782 reasons (thanks to Marco Gaiarin).
2784 SC/06 Eximstats V1.48
2785 Mainlog lines which have GMT offsets and are too short to
2786 have a flag are now skipped.
2788 SC/07 Eximstats V1.49 (Alain Williams)
2789 Added the -emptyok flag.
2791 SC/08 Eximstats V1.50
2792 Fixes for obtaining the IP address from reject messages.
2794 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2795 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2796 whitespace changes from 4.64-PH/27
2798 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2801 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2802 are found, allow negative numbers in numeric criteria)
2804 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2806 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2807 to match changes made in 4.64-PH/43
2809 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2811 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2812 "domain" is actually an IP address, and force a failure. This locks out
2813 those revolvers/nameservers that support "A-for-A" lookups, in
2814 contravention of the specifications.
2816 PH/48 When a host name was looked up from an IP address, and the subsequent
2817 forward lookup of the name timed out, the host name was left in
2818 $sender_host_name, contrary to the specification.
2820 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2821 restricted to single-key lookups, Exim was not diagnosing an error if
2822 * or *@ was used with a query-style lookup.
2824 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2826 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2827 long ago, but noone interested enough thought of it. Let's just say that
2828 the "1.1" means that there are some new functions that weren't there at
2829 some point in the past.
2831 PH/51 Error processing for expansion failure of helo_data from an smtp
2832 transport during callout processing was broken.
2834 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2835 tested/used via the -bh/-bhc/-bs options.
2837 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2838 bug, fixed in subsequent PCRE releases).
2840 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2841 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2843 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2849 SC/01 Use a glob alias rather than an array ref in eximstats generated
2850 parser. This improves both readability and performance.
2852 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2853 Don't display local sender or destination tables in eximstats unless
2854 there is data to show.
2855 Added average volumes into the eximstats top table text output.
2857 SC/03 Collect data on the number of addresses (recipients) as well
2858 as the number of messages in eximstats.
2860 TF/01 Correct an error in the documentation for the redirect router. Exim
2861 does (usually) call initgroups() when daemonizing.
2863 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2864 with consistent privilege compared to when running as a daemon.
2866 TF/03 Note in the spec that $authenticated_id is not set for local
2867 submissions from trusted users.
2869 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2870 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2872 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2873 by adding some example configuration directives to the default
2874 configuration file. A little bit of work is required to uncomment the
2875 directives and define how usernames and passwords are checked, but
2876 there is now a framework to start from.
2878 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2879 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2880 without this. I don't know how relevant this is to other LDAP libraries.
2882 PH/02 Add the verb name to the "unknown ACL verb" error.
2884 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2886 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2888 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2889 directory not expanded when it should be if an expanded home directory
2890 was set for the address (which is overridden by the transport).
2892 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2895 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2896 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2897 because it is too late at that time, and has no effect.
2899 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2900 security issue with \' (bugzilla #107). I could not use the
2901 PQescapeStringConn() function, because it needs a PGconn value as one of
2904 PH/08 When testing addresses using -bt, indicate those final addresses that
2905 are duplicates that would not cause an additional delivery. At least one
2906 person was confused, thinking that -bt output corresponded to deliveries.
2907 (Suppressing duplicates isn't a good idea as you lose the information
2908 about possibly different redirections that led to the duplicates.)
2910 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2911 systems where poll() doesn't work, in particular OS X.
2913 PH/10 Added more information to debugging output for retry time not reached.
2915 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2916 operations in malware.c.
2918 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2921 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2922 syslog_duplication set false, log lines that would normally be written
2923 both the the main log and to the reject log were not written to syslog at
2926 PH/14 In the default configuration, change the use of "message" in ACL warn
2927 statements to "add_header".
2929 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2930 not followed by a command (e.g. "seen endif").
2932 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2933 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2936 PH/17 Added extra conditions to the default value of delay_warning_condition
2940 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2941 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2942 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2945 The Auto-Submitted: and various List- headers are standardised, whereas I
2946 don't think Precedence: ever was.
2948 PH/18 Refactored debugging code in route_finduser() to show more information,
2949 in particular, the error code if getpwnam() issues one.
2951 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2952 This is apparently needed in addition to the PH/07 change above to avoid
2953 any possible encoding problems.
2955 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2956 but not after initializing Perl.
2958 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2959 output them only if debugging. By default they are written stderr,
2960 apparently, which is not desirable.
2962 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2965 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2968 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2970 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2971 authenticated or an ident call has been made. Suppress the default
2972 values for $authenticated_id and $authenticated_sender (but permit -oMai
2973 and -oMas) when testing with -bh.
2975 PH/24 Re-jigged the order of the tests in the default configuration so that the
2976 tests for valid domains and recipients precede the DNS black list and CSA
2977 tests, on the grounds that those ones are more expensive.
2979 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2980 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2981 This bug exists in every version of Exim that I still have, right back to
2984 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2985 However, an attempt to turn on case-sensitivity in a regex key by
2986 including (?-i) didn't work because the subject string was already
2987 lowercased, and the effects were non-intuitive. It turns out that a
2988 one-line patch can be used to allow (?-i) to work as expected.
2994 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2995 other effects) broke the use of negated acl sub-conditions.
2997 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3000 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3001 "Deny" causes Exim to reject the incoming connection with a 554 error.
3002 Unfortunately, if there is a major crisis, such as a disk failure,
3003 tcp-wrappers gives "deny", whereas what one would like would be some
3004 kind of temporary error. A kludge has been added to help with this.
3005 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3006 554 error is used if errno is still zero or contains ENOENT (which occurs
3007 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3010 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3012 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3013 errors. Otherwise a message that provokes a temporary error (when other
3014 messages do not) can cause a whole host to time out.
3016 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3017 addresses were routed directly to files or pipes from a redirect router.
3018 File deliveries just didn't batch; pipe deliveries might have suffered
3021 PH/06 A failure to get a lock for a hints database would erroneously always say
3022 "Failed to get write lock", even when it was really a read lock.
3024 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3025 of 0600. This has been changed to use the value of the lockfile_mode
3026 option (which defaults to 0600).
3028 PH/08 Applied small patch from the Sieve maintainer.
3030 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3031 folder from quota calculations, a direct delivery into this folder messed
3032 up the contents of the maildirsize file. This was because the regex was
3033 used only to exclude .Trash (or whatever) when the size of the mailbox
3034 was calculated. There was no check that a delivery was happening into an
3035 excluded directory. This bug has been fixed by ignoring all quota
3036 processing for deliveries into excluded directories.
3038 PH/10 Added the maildirfolder_create_regex option to appendfile.
3044 PH/01 The code for finding all the local interface addresses on a FreeBSD
3045 system running IPv6 was broken. This may well have applied to all BSD
3046 systems, as well as to others that have similar system calls. The broken
3047 code found IPv4 interfaces correctly, but gave incorrect values for the
3048 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3049 that it would not match correctly against @[] and not recognize the IPv6
3052 PH/02 The ipliteral router was not recognizing addresses of the form user@
3053 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3055 PH/03 Added disable_ipv6.
3057 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3058 decoded form, because it is most often used to construct To: headers
3059 lines in autoreplies, and the decoded form may well be syntactically
3060 invalid. However, $reply_address has leading white space removed, and all
3061 newlines turned into spaces so that the autoreply transport does not
3064 PH/05 If group was specified without a user on a router, and no group or user
3065 was specified on a transport, the group from the router was ignored.
3067 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3068 for visible compile-time settings that can be used to change these
3069 numbers, for those that want even more. Backwards compatibility with old
3070 spool files has been maintained. However, going back to a previous Exim
3071 release will lost any variables that are in spool files.
3073 PH/07 Two small changes when running in the test harness: increase delay when
3074 passing a TCP/IP connection to a new process, in case the original
3075 process has to generate a bounce, and remove special handling of
3076 127.0.0.2 (sic), which is no longer necessary.
3078 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3079 be the same on different OS.
3081 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3084 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3085 whether --show-vars was specified or not
3087 JJ/02 exipick: Added support for new ACL variable spool format introduced
3090 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3091 syntactically invalid From: or Reply-to: line, and a filter used this to
3092 generate an autoreply, and therefore failed to obtain an address for the
3093 autoreply, Exim could try to deliver to a non-existent relative file
3094 name, causing unrelated and misleading errors. What now happens is that
3095 it logs this as a hard delivery error, but does not attempt to create a
3098 PH/11 The exinext utility has a -C option for testing purposes, but although
3099 the given file was scanned by exinext itself; it wasn't being passed on
3100 when Exim was called.
3102 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3103 an end-of-file indication when reading a command response.
3105 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3106 compiled. In many other places in Exim, IPv6 addresses are always
3107 recognized, so I have changed this. It also means that IPv4 domain
3108 literals of the form [IPV4:n.n.n.n] are now always recognized.
3110 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3111 used if the router is not running as root, for example, when verifying at
3112 ACL time, or when using -bh. The debugging output from this situation was
3113 non-existent - all you got was a failure to exec. I have made two
3116 (a) Failures to set uid/gid, the current directory, or a process leader
3117 in a subprocess such as that created by queryprogram now generate
3118 suitable debugging output when -d is set.
3120 (b) The queryprogram router detects when it is not running as root,
3121 outputs suitable debugging information if -d is set, and then runs
3122 the subprocess without attempting to change uid/gid.
3124 PH/15 Minor change to Makefile for building test_host (undocumented testing
3127 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3128 additional section of a DNS packet that returns MX or SRV records.
3129 Instead, it always explicitly searches for A/AAAA records. This avoids
3130 major problems that occur when a DNS server includes only records of one
3131 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3132 fixed another bug: if SRV records were looked up and the corresponding
3133 address records were *not* found in the additional section, the port
3134 values from the SRV records were lost.
3136 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3137 using the correct key (the original address) when searching the retry
3138 rules in order to find which one to use for generating the retry hint.
3140 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3141 adding the default one. Similarly, if it contains a Reply-To: header, the
3142 errors_reply_to option, if set, is not used.
3144 PH/19 When calculating a retry time, Exim used to measure the "time since
3145 failure" by looking at the "first failed" field in the retry record. Now
3146 it does not use this if it is later than than the arrival time of the
3147 message. Instead it uses the arrival time. This makes for better
3148 behaviour in cases where some deliveries succeed, thus re-setting the
3149 "first failed" field. An example is a quota failure for a huge message
3150 when small messages continue to be delivered. Without this change, the
3151 "time since failure" will always be short, possible causing more frequent
3152 delivery attempts for the huge message than are intended.
3153 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3155 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3156 $1, $2, $3) because the numerical variables can be reset during some
3157 expansion items (e.g. "match"), thereby losing the authentication data.
3159 PH/21 Make -bV show the size of off_t variables so that the test suite can
3160 decide whether to run tests for quotas > 2G.
3162 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3163 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3164 filecount value is greater than 2G or if a quota value is greater than 2G
3165 on a system where the size of off_t is not greater than 4, a panic error
3168 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3169 never match. The debug and -bh output now contains an explicit error
3170 message indicating a malformed IPv4 address or mask.
3172 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3173 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3174 PH/23 above applies.
3176 PH/25 Do not write to syslog when running in the test harness. The only
3177 occasion when this arises is a failure to open the main or panic logs
3178 (for which there is an explicit test).
3180 PH/26 Added the /no_tell option to "control=freeze".
3182 PH/27 If a host name lookup failed very early in a connection, for example, if
3183 the IP address matched host_lookup and the reverse lookup yielded a name
3184 that did not have a forward lookup, an error message of the form "no IP
3185 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3186 could be logged. Now it outputs the IP address instead of "NULL".
3188 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3189 allows the sender and the authenticated sender to be set when
3190 submitting a message from within Exim. Since child_open_exim() is
3191 documented for local_scan(), the new function should be too.
3193 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3194 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3195 results in an empty string is now treated as unset.
3197 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3199 PH/31 Added sender_verify_fail logging option.
3201 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3202 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3203 tidied the source and removed it altogether.
3205 PH/33 When a queue run was abandoned because the load average was too high, a
3206 log line was always written; now it is written only if the queue_run log
3207 selector is set. In addition, the log line for abandonment now contains
3208 information about the queue run such as the pid. This is always present
3209 in "start" and "stop" lines but was omitted from the "abandon" line.
3211 PH/34 Omit spaces between a header name and the colon in the error message that
3212 is given when verify = headers_syntax fails (if there are lots of them,
3213 the message gets confusing).
3215 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3216 names, as there are now some PTR records that contain slashes. This check
3217 is only to protect against broken name servers that fall over on strange
3218 characters, so the fact that it applies to all lookups doesn't matter.
3220 PH/36 Now that the new test suite is complete, we can remove some of the
3221 special code in Exim that was needed for the old test suite. For example,
3222 sorting DNS records because real resolvers return them in an arbitrary
3223 order. The new test suite's fake resolver always returns records in the
3226 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3227 bounces) except when queue_only is set, to avoid logging races between
3228 the different processes.
3230 PH/38 Panic-die if .include specifies a non-absolute path.
3232 PH/39 A tweak to the "H" retry rule from its user.
3234 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3235 a label. They prevented compilation on older perls.
3237 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3238 a warning to be raised on newish perls.
3240 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3241 on queue. Changes to match documented behaviour of showing count of
3242 messages matching specified criteria.
3244 PH/40 Changed the default ident timeout from 30s to 5s.
3246 PH/41 Added support for the use of login_cap features, on those BSD systems
3247 that have them, for controlling the resources used by pipe deliveries.
3249 PH/42 The content-scanning code uses fopen() to create files in which to put
3250 message data. Previously it was not paying any attention to the mode of
3251 the files. Exim runs with umask(0) because the rest of the code creates
3252 files with open(), and sets the required mode explicitly. Thus, these
3253 files were ending up world-writeable. This was not a big issue, because,
3254 being within the spool directory, they were not world-accessible. I have
3255 created a function called modefopen, which takes an additional mode
3256 argument. It sets umask(777), creates the file, chmods it to the required
3257 mode, then resets the umask. All the relevant calls to fopen() in the
3258 content scanning code have been changed to use this function.
3260 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3261 to 24 hours. This avoids potential overflow problems when processing G
3262 and H retry rules. I suspect nobody ever tinkers with this value.
3264 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3266 PH/45 When the plaintext authenticator is running as a client, the server's
3267 challenges are checked to ensure they are valid base64 strings. By
3268 default, the authentication attempt is cancelled if an invalid string is
3269 received. Setting client_ignore_invalid_base64 true ignores these errors.
3270 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3271 they are received. Thus, the responses can be made to depend on the
3272 challenges. If an invalid string is ignored, an empty string is placed in
3275 PH/46 Messages that are created by the autoreply transport now contains a
3276 References: header, in accordance with RFCs 2822 and 3834.
3278 PH/47 Added authenticated_sender_force to the smtp transport.
3280 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3282 PH/49 Installed latest patch from the Sieve maintainer.
3284 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3285 was also set, the appendfile transport was unnecessarily scanning a
3286 directory of message files (e.g. for maildir delivery) to find the count
3287 of files (along with the size), even though it did not need this
3288 information. It now does the scan only if it needs to find either the
3289 size of the count of files.
3291 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3293 PH/52 Two bugs concerned with error handling when the smtp transport is
3296 (i) Exim was not creating retry information for temporary errors given
3297 for individual recipients after the DATA command when the smtp transport
3298 was used in LMTP mode. This meant that they could be retried too
3299 frequently, and not timed out correctly.
3301 (ii) Exim was setting the flag that allows error details to be returned
3302 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3303 recipients that were returned after the DATA command.
3305 PH/53 This is related to PH/52, but is more general: for any failing address,
3306 when detailed error information was permitted to be returned to the
3307 sender, but the error was temporary, then after the final timeout, only
3308 "retry timeout exceeded" was returned. Now it returns the full error as
3309 well as "retry timeout exceeded".
3311 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3312 do this, and (what is worse) MTAs that accept it.
3314 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3315 will now be deprecated.
3317 PH/56 New os.c-cygwin from the Cygwin maintainer.
3319 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3320 formats (previously only available in exim formats via -bpr, -bpru,
3321 and -bpra. Now also available in native and exiqgrep formats)
3323 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3324 with very large, slow to parse queues
3326 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3328 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3330 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3331 responses to authentication challenges, though it was showing the
3332 challenges; (ii) I've removed the CR characters from the debug output for
3335 PH/58 Allow for the insertion of a newline as well as a space when a string
3336 is turned into more than one encoded-word during RFC 2047 encoding. The
3337 Sieve code now uses this.
3339 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3340 data_4xx, lost_connection, tls_required.
3342 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3343 message was being sent as an SMTP response.
3345 PH/61 Add -l and -k options to exicyclog.
3347 PH/62 When verifying, if an address was redirected to one new address, so that
3348 verification continued, and the new address failed or deferred after
3349 having set something in $address_data, the value of $address_data was not
3350 passed back to the ACL. This was different to the case when no
3351 redirection occurred. The value is now passed back in both cases.
3353 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3354 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3355 use login_cap.h, so on its own it isn't the distinguishing feature. The
3356 new name refers directly to the setclassresources() function.
3358 PH/65 Added configuration files for NetBSD3.
3360 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3362 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3363 is preferred over IPv4.
3365 PH/68 The bounce_return_message and bounce_return_body options were not being
3366 honoured for bounces generated during the reception of non-SMTP messages.
3367 In particular, this applied to messages rejected by the ACL. This bug has
3368 been fixed. However, if bounce_return_message is true and bounce_return_
3369 body is false, the headers that are returned for a non-SMTP message
3370 include only those that have been read before the error was detected.
3371 (In the case of an ACL rejection, they have all been read.)
3373 PH/69 The HTML version of the specification is now built in a directory called
3374 spec_html instead of spec.html, because the latter looks like a path with
3375 a MIME-type, and this confuses some software.
3377 PH/70 Catch two compiler warnings in sieve.c.
3379 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3380 function verify_get_ident() calls ip_connect() to connect a socket, but
3381 if the "connect()" function timed out, ip_connect() used to close the
3382 socket. However, verify_get_ident() also closes the socket later, and in
3383 between Exim writes to the log, which may get opened at this point. When
3384 the socket was closed in ip_connect(), the log could get the same file
3385 descriptor number as the socket. This naturally causes chaos. The fix is
3386 not to close the socket in ip_connect(); the socket should be closed by
3387 the function that creates it. There was only one place in the code where
3388 this was missing, in the iplookup router, which I don't think anybody now
3389 uses, but I've fixed it anyway.
3391 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3392 well as to direct DNS lookups. Otherwise the handling of names in host
3393 lists is inconsistent and therefore confusing.
3399 PH/01 Two changes to the default runtime configuration:
3401 (1) Move the checks for relay_from_hosts and authenticated clients from
3402 after to before the (commented out) DNS black list checks.
3404 (2) Add control=submission to the relay_from_hosts and authenticated
3405 clients checks, on the grounds that messages accepted by these
3406 statements are most likely to be submissions.
3408 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3410 (1) Generate an error if the third argument for the ${prvs expansion is
3413 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3416 (3) Reset the variables that are obtained from the first argument of
3417 ${prvscheck and used in the second argument before leaving the code,
3418 because their memory is reclaimed, so using them afterwards may do
3421 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3422 one (it's much easier than Tom thought :-).
3424 (5) Because of (4), we can now allow for the use of $prvscheck_result
3425 inside the third argument.
3427 PH/03 For some reason, the default setting of PATH when running a command from
3428 a pipe transport was just "/usr/bin". I have changed it to
3431 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3432 anything to be listed in the output from -bV.
3434 PH/05 When a filter generated an autoreply, the entire To: header line was
3435 quoted in the delivery log line, like this:
3437 => >A.N.Other <ano@some.domain> <original@ddress> ...
3439 This has been changed so that it extracts the operative address. There
3440 may be more than one such address. If so, they are comma-separated, like
3443 => >ano@some.domain,ona@other.domain <original@ddress> ...
3445 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3446 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3447 not being looked up in the rDNS to get a host name, Exim was showing the
3448 IP address twice in Received: lines, even though the IP addresses were
3449 identical. For example:
3451 Received: from [1.2.3.4] (helo=[1.2.3.4])
3453 However, if the real host name was known, it was omitting the HELO data
3454 if it matched the actual IP address. This has been tidied up so that it
3455 doesn't show the same IP address twice.
3457 PH/07 When both +timestamp and +memory debugging was on, the value given by
3458 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3459 called by the debug printing, thereby overwriting the timestamp buffer.
3460 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3462 PH/08 When the original message was included in an autoreply transport, it
3463 always said "this is a copy of the message, including all the headers",
3464 even if body_only or headers_only was set. It now gives an appropriate
3467 PH/09 Applied a patch from the Sieve maintainer which:
3469 o fixes some comments
3470 o adds the (disabled) notify extension core
3471 o adds some debug output for the result of if/elsif tests
3472 o points to the current vacation draft in the documentation
3473 and documents the missing references header update
3477 o fixes a bug in processing the envelope test (when testing
3478 multiple envelope elements, the last element determined the
3481 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3482 Electronic Mail") by including:
3484 Auto-submitted: auto-generated
3486 in the messages that it generates (bounce messages and others, such as
3487 warnings). In the case of bounce messages for non-SMTP messages, there was
3488 also a typo: it was using "Auto_submitted" (underscore instead of
3489 hyphen). Since every message generated by Exim is necessarily in response
3490 to another message, thes have all been changed to:
3492 Auto-Submitted: auto-replied
3494 in accordance with these statements in the RFC:
3496 The auto-replied keyword:
3498 - SHOULD be used on messages sent in direct response to another
3499 message by an automatic process,
3501 - MUST NOT be used on manually-generated messages,
3503 - MAY be used on Delivery Status Notifications (DSNs) and Message
3504 Disposition Notifications (MDNs),
3506 - MUST NOT be used on messages generated by automatic or periodic
3507 processes, except for messages which are automatic responses to
3510 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3511 to the default Received: header definition.
3513 PH/12 Added log selector acl_warn_skipped (default on).
3515 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3516 variables because (a) they are in the wrong storage pool and (b) even if
3517 they were copied, it wouldn't work properly because of the caching.
3519 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3520 checking when decoding. Apparently there are clients that generate
3521 overlong encoded strings. Why am I not surprised?
3523 PH/15 If the first argument of "${if match_address" was not empty, but did not
3524 contain an "@" character, Exim crashed. Now it writes a panic log message
3525 and treats the condition as false.
3527 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3529 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3530 extension "envelope-auth". The code is finished and in agreement with
3531 other implementations, but there is no documentation so far and in fact,
3532 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3533 not changing the active code.
3535 Print executed "if" and "elsif" statements when debugging is used. This
3536 helps a great deal to understand what a filter does.
3538 Document more things not specified clearly in RFC3028. I had all this
3539 sorted out, when out of a sudden new issues came to my mind. Oops."
3541 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3544 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3545 canonical form (as documented). However, after a host name lookup from
3546 the IP address, check_host() was doing a simple string comparison with
3547 addresses acquired from the DNS when checking that the found name did
3548 have the original IP as one of its addresses. Since any found IPv6
3549 addresses are likely to be in abbreviated form, the comparison could
3550 fail. Luckily, there already exists a function for doing the comparison
3551 by converting both addresses to binary, so now that is used instead of
3552 the text comparison.
3554 PH/20 There was another similar case to PH/19, when a complete host name was
3555 given in a host list; looking up its IP address could give an abbreviated
3556 form, whereas the current host's name might or might not be abbreviated.
3557 The same fix has been applied.
3563 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3564 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3567 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3568 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3570 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3572 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3573 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3574 Also, for "Auto-Submitted", treat anything other than "no" as
3575 non-personal, in accordance with RFC 3834. (Previously it treated
3576 anything starting "auto-" as non-personal.)
3578 TF/01 The control=submission/name=... option had a problem with syntax
3579 errors if the name included a slash character. The /name= option
3580 now slurps the rest of the string, so it can include any characters
3581 but it must come last in the list of options (after /sender_retain
3584 PH/05 Some modifications to the interface to the fake nameserver for the new
3592 TK/01 Added the "success_on_redirect" address verification option. See
3593 NewStuff for rationale and an example.
3595 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3597 PH/02 Patch to exigrep to allow it to work on syslog lines.
3599 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3600 fread() to skip over the body file's header line, because in Cygwin the
3601 header line is locked and is inaccessible.
3603 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3604 co-exist for some time) to make it clear that it is the Exim ID that is
3605 referenced, not the Message-ID: header line.
3607 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3608 string_format() function, because snprintf() does not exist on all
3611 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3612 ${stat: expansion item.
3614 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3615 protocol synchronization error", to keep the pedants happy.
3617 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3618 well as for IRIX systems, when gcc is being used. See the host.c source
3621 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3623 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3626 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3627 transport and to the smtp transport in LMTP mode.
3629 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3631 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3632 run a filter in a subprocess. This could lead to confusion in subsequent
3633 lookups in the parent process. There should also be a search_tidyup() at
3634 the end of the subprocess.
3636 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3637 only if the host matched helo_try_verify_hosts, which caused the
3638 verification to occur when the EHLO/HELO command was issued. The ACL just
3639 tested the remembered result. Now, if a previous verification attempt has
3640 not happened, "verify = helo" does it there and then.
3642 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3644 TK/03 Fix log output including CR from clamd.
3646 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3647 exist provoked a memory error which could cause a segfault.
3649 PH/15 Installed PCRE 6.2
3651 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3653 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3654 of the problem. Specifically, suggested +O2 rather than +O1 for the
3657 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3659 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3660 record, the port was not correctly logged when the outgoing_port log
3661 selector was set (it logged the transort's default port).
3663 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3664 fallback_hosts, and "hosts" in the smtp transport.
3666 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3667 host errors such as "Connection refused".
3669 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3670 authentication with radiusclient 0.4.9:
3672 - Error returned from rc_read_config was caught wrongly
3673 - Username/password not passed on to radius server due to wrong length.
3675 The presumption is that some radiusclient API changes for 4.51/PH/17
3676 were not taken care of correctly. The code is still untested by me (my
3677 Linux distribution still has 0.3.2 of radiusclient), but it was
3678 contributed by a Radius user.
3680 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3681 expanding the "port" option of the smtp transport.
3683 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3684 while reading a MIME header. Thanks to Tom Hughes for a patch.
3686 PH/24 Include config.h inside local_scan.h so that configuration settings are
3689 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3690 that in an ACL for RCPT (for example), you can examine exactly what was
3693 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3694 commands, but it was not correctly comparing the address with the actual
3695 client host address. Thus, it would show the EHLO address in Received:
3696 header lines when this was not necessary.
3698 PH/27 Added the % operator to ${eval:}.
3700 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3701 it should be ignoring failures (because with -C, for example, it has lost
3702 privilege). It wasn't ignoring creation failures other than "already
3705 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3708 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3709 because an input line was too long, either on its own, or by virtue of
3710 too many continuations, the temporary file was not being removed, and the
3711 return code was incorrect.
3713 PH/31 Missing "BOOL" in function definition in filtertest.c.
3715 PH/32 Applied Sieve patches from the maintainer.
3717 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3719 PH/33 Added "verify = not_blind".
3721 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3722 Local/Makefile (with some defaults set). These are used in built scripts
3723 such as exicyclog, but they have never been used in the exim_install
3724 script (though there are many overriding facilities there). I have
3725 arranged that the exim_install script now takes note of these two
3728 PH/35 Installed configuration files for Dragonfly.
3730 PH/36 When a locally submitted message by a trusted user did not contain a
3731 From: header, and the sender address was obtained from -f or from an SMTP
3732 MAIL command, and the trusted user did not use -F to supply a sender
3733 name, $originator_name was incorrectly used when constructing a From:
3734 header. Furthermore, $originator_name was used for submission mode
3735 messages from external hosts without From: headers in a similar way,
3736 which is clearly wrong.
3738 PH/37 Added control=suppress_local_fixups.
3740 PH/38 When log_selector = +received_sender was set, and the addition of the
3741 sender made the log line's construction buffer exactly full, or one byte
3742 less than full, an overflow happened when the terminating "\n" was
3745 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3746 when the result of a list match is failure because a DNS lookup failed.
3748 PH/40 RM_COMMAND is now used in the building process.
3750 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3751 the "build-* directories that it finds.
3753 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3754 address in a domain literal was a prefix of an interface address.
3756 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3757 when verifying a sender address, unless rewrite_headers is false.
3759 PH/44 Wrote a long comment about why errors_to addresses are verified as
3760 recipients, not senders.
3762 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3763 the ratelimit ACL was added.
3765 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3767 PH/46 Added extra information about PostgreSQL errors to the error string.
3769 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3770 avoiding the need to install special zones in a real server. This is
3771 backwards compatible; if it can't find the fake resolver, it drops back.
3772 Thus, both old and new test suites can be run.
3774 TF/02 Added util/ratelimit.pl
3776 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3777 clock is set back in time.
3779 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3780 Candler <B.Candler@pobox.com>.
3782 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3783 OK for addresses that are the result of redirections.
3785 PH/48 A number of further additions for the benefit of the new test suite,
3786 including a fake gethostbyname() that interfaces to the fake DNS resolver
3789 TF/06 The fix for widen_domains has also been applied to qualify_single and
3790 search_parents which are the other dnslookup options that can cause
3793 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3796 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3798 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3799 service" if it encountered an MX record with an empty target hostname.
3800 The message is now "an MX or SRV record indicated no SMTP service".
3802 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3803 if the DNS of the sending site is misconfigured. This is quite a
3804 common situation. This change restores the behaviour of treating a
3805 helo verification defer as a failure.
3807 PH/51 If self=fail was set on a router, the bounce message did not include the
3808 actual error message.
3814 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3816 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3817 command itself ended in error, the underlying message about the transport
3818 filter timeout was being overwritten with the pipe command error. Now the
3819 underlying error message should be appended to the second error message.
3821 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3823 PH/02 Reduce the amount of output that "make" produces by default. Full output
3824 can still be requested.
3826 PH/03 The warning log line about a condition test deferring for a "warn" verb
3827 was being output only once per connection, rather than after each
3828 occurrence (because it was using the same function as for successful
3829 "warn" verbs). This seems wrong, so I have changed it.
3831 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3832 it should not have, which might have caused a crash in the right
3833 circumstances, but probably never did.
3835 PH/04 Installed a modified version of Tony Finch's patch to make submission
3836 mode fix the return path as well as the Sender: header line, and to
3837 add a /name= option so that you can make the user's friendly name appear
3840 TF/03 Added the control = fakedefer ACL modifier.
3842 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3843 Mark Lowes for thorough testing.
3845 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3847 TK/03 Merged latest SRS patch from Miles Wilton.
3849 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3850 with the definition in sysexits.h (which is #included earlier).
3851 Fortunately, Exim does not actually use EX_OK. The code used to try to
3852 preserve the sysexits.h value, by assuming that macro definitions were
3853 scanned for macro replacements. I have been disabused of this notion,
3854 so now the code just undefines EX_OK before #including unistd.h.
3856 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3857 in the smtp transport. When a block could not be written in a single
3858 write() function, the timeout was being re-applied to each part-write.
3859 This seems wrong - if the receiver was accepting one byte at a time it
3860 would take for ever. The timeout is now adjusted when this happens. It
3861 doesn't have to be particularly precise.
3863 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3864 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3866 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3867 without a domain if the check to <postmaster@domain> fails.
3869 SC/01 Eximstats: added -xls and the ability to specify output files
3870 (patch written by Frank Heydlauf).
3872 SC/02 Eximstats: use FileHandles for outputting results.
3874 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3876 SC/04 Eximstats: fixed display of large numbers with -nvr option
3878 SC/05 Eximstats: fixed merging of reports with empty tables.
3880 SC/06 Eximstats: added the -include_original_destination flag
3882 SC/07 Eximstats: removed tabs and trailing whitespace.
3884 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3886 TK/06 MBOX spool code: Add real "From " MBOX separator line
3887 so the .eml file is really in mbox format (even though
3888 most programs do not really care). Patch from Alex Miller.
3890 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3891 The latter is generated from $received_to and is only set if the
3892 message has one envelope recipient. SA can use these headers,
3893 obviously out-of-the-box. Patch from Alex Miller.
3895 PH/08 The ${def test on a variable was returning false if the variable's
3896 value was "0", contrary to what the specification has always said!
3897 The result should be true unless the variable is empty.
3899 PH/09 The syntax error of a character other than { following "${if
3900 def:variable_name" (after optional whitespace) was not being diagnosed.
3901 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3902 accidental colon was present, for example, could give incorrect results.
3904 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3905 result is used (not including appendfile, where other changes are about
3908 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3909 This involved changing a lot of size variables from int to off_t. It
3910 should work with maildirs and everything.
3912 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3913 spamd dying while we are connected to it.
3915 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3918 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3919 function for BDB 4.3.
3921 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3923 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3924 the sources for PCRE have been re-arranged and more files are now
3927 PH/15 The code I had for printing potentially long long variables in PH/11
3928 above was not the best (it lost precision). The length of off_t variables
3929 is now inspected at build time, and an appropriate printing format (%ld
3930 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3931 to be "long long int" or "long int". This is needed for the internal
3932 formatting function string_vformat().
3934 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3935 the configuration file to be ":syslog", then the script "guesses" where
3936 the logs files are, rather than using the compiled in default. In our
3937 case the guess is not the same as the compiled default, so the script
3938 suddenly stopped working when I started to use syslog. The patch checks
3939 to see if log_file_path is "". If so, it attempts to read it from exim
3940 with no configuration file to get the compiled in version, before it
3941 falls back to the previous guessing code."
3943 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3944 implementing BATV in an Exim configuration. See NewStuff for the gory
3947 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3948 Makefile that are specific to HP-UX.
3950 PH/18 If the "use_postmaster" option was set for a recipient callout together
3951 with the "random" option, the postmaster address was used as the MAIL
3952 FROM address for the random test, but not for the subsequent recipient
3953 test. It is now used for both.
3955 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3956 patch removes a few documentation additions to RFC 3028, because the
3957 latest draft now contains them. It adds the new en;ascii-case comparator
3958 and a new error check for 8bit text in MIME parts. Comparator and
3959 require names are now matched exactly. I enabled the subaddress
3960 extension, but it is not well tested yet (read: it works for me)."
3962 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3963 rework some of the code of TK/09 above to avoid the hardwired use of
3964 "%lld" and "long long". Replaced the call to snprintf() with a call to
3967 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3968 records point to non-existent hosts", "retry timeout exceeded", and
3969 "retry time not reached for any host after a long failure period".
3971 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3972 experimental DomainKeys support:
3974 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3975 (2) On an error such as an illegally used "control", the wrong name for
3976 the control was given.
3978 These problems did NOT occur unless DomainKeys support was compiled.
3980 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3982 PH/24 Added ${if match_ip condition.
3984 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3985 databases so that it will be absolutely obvious if a crash occurs in the
3986 DB library. This is a regular occurrence (often caused by mis-matched
3989 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3990 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3991 picked up on a user's system that detects such things. There doesn't seem
3992 to be a gcc warning option for this - only an attribute that has to be
3993 put on the function's prototype. It seems that in Fedora Core 4 they have
3994 set this on a number of new functions. No doubt there will be more in due
3997 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3998 specify a transport. However, if an address that was verified by such a
3999 router was the subject of a callout, Exim crashed because it tried to
4000 read the rcpt_include_affixes from the non-existent transport. Now it
4001 just assumes that the setting of that option is false. This bug was
4002 introduced by 4.51/PH/31.
4004 PH/28 Changed -d+all to exclude +memory, because that information is very
4005 rarely of interest, but it makes the output a lot bigger. People tend to
4006 do -d+all out of habit.
4008 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4009 code in os-type was giving problems when libc.so lives in lib64, like on
4012 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4013 aren't the modern standard, and it seems that some systems' include files
4014 don't always have them. Exim was already checking for some of the newer
4015 ones like T_AAAA, and defining it itself. I've added checks for all the
4016 record types that Exim uses.
4018 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4019 not automatically generating a new one, as it is supposed to. This
4020 prevented TLS from working. If the file did exist, but contained invalid
4021 data, a new version was generated, as expected. It was only the case of a
4022 non-existent file that was broken.
4024 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4025 with a change in libdomainkeys > 0.64.
4027 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4028 from DNS. If the selector record carries the flag, it now has
4029 precedence over the domain-wide flag.
4031 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4033 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4034 the use of a port name that isn't defined in /etc/services) occurred, the
4035 message was deferred as in a normal delivery, and thus remained on the
4036 spool, instead of being failed because of the mua_wrapper setting. This
4037 is now fixed, and I tidied up some of the mua_wrapper messages at the
4040 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4041 the messages in a hash of arrays rather than using individual hashes.
4042 This is a bit cleaner and results in dramatic memory savings, albeit
4043 at a slight CPU cost.
4045 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4046 as requested by Marc Sherman.
4048 SC/10 Eximstats: added histograms for user specified patterns as requested
4051 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4053 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4054 fopen() in the content-scanning modules that did not already have it.
4060 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4061 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4063 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4065 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4067 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4068 to test Sieve filters that use "vacation".
4070 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4071 that changes the way the GnuTLS parameters are stored in the cache file.
4072 The new format can be generated externally. For backward compatibility,
4073 if the data in the cache doesn't make sense, Exim assumes it has read an
4074 old-format file, and it generates new data and writes a new file. This
4075 means that you can't go back to an older release without removing the
4078 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4079 work if there are any delivery delays because "one_time" forces the
4080 parent to be marked "delivered", so its unseen clone is never tried
4081 again. For this reason, Exim now forbids the simultaneous setting of
4084 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4085 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4086 entry if you want to know the details. The fix, however, neglected to
4087 consider the case when local delivery batching is involved. The test for
4088 "previously delivered" was not happening when checking to see if an
4089 address could be batched with a previous (undelivered) one; under
4090 certain circumstances this could lead to multiple deliveries to the same
4093 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4094 in its include files, and this causes problems building Exim.
4096 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4097 header_syntax) but Exim was just ignoring anything given after a slash.
4098 In particular, this caused confusion with an attempt to use "verify =
4099 reverse_host_lookup/defer_ok". An error is now given when options are
4100 supplied for verify items that do not have them. (Maybe reverse_host_
4101 lookup should have a defer_ok option, but that's a different point.)
4103 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4104 defined by RFC 821) to 2048, because there were problems with some AUTH
4105 commands, and RFC 1869 says the size should be increased for extended
4106 SMTP commands that take arguments.
4108 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4111 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4112 "unknown" error; now it says that the functionality isn't in the binary.
4114 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4115 an address' error message when a string expansion fails (syntax or
4116 whatever). Otherwise the password may appear in the log. Following change
4117 PH/42 below, there is no longer a chance of it appearing in a bounce
4120 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4122 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4123 of its addresses was ever tried. (Bugzilla bug #2.)
4125 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4126 the result incorrectly in the debug output. (It correctly added a newline
4127 to what was transported.)
4129 TF/01 Added $received_time.
4131 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4132 commented out examples of how to interface to a virus scanner and to
4133 SpamAssassin. Also added commented examples of av_scanner and
4134 spamd_address settings.
4136 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4137 and controls are allowed in which ACLs. There were a couple of minor
4138 errors. Some of the entries in the conditions table (which is a table of
4139 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4140 negation of where the condition IS allowed.
4142 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4144 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4145 header file does not have a version number, so I've had to invent a new
4146 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4147 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4148 radiusclient), but it was contributed by a Radius user.
4150 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4151 files or directories, for OpenSSL.
4153 PH/19 When an Exim process that is running as root has to create an Exim log
4154 file, it does so in a subprocess that runs as exim:exim so as to get the
4155 ownership right at creation (otherwise, other Exim processes might see
4156 the file with the wrong ownership). There was no test for failure of this
4157 fork() call, which would lead to the process getting stuck as it waited
4158 for a non-existent subprocess. Forks do occasionally fail when resources
4159 run out. I reviewed all the other calls to fork(); they all seem to check
4162 PH/20 When checking for unexpected SMTP input at connect time (before writing
4163 the banner), Exim was not dealing correctly with a non-positive return
4164 from the read() function. If the client had disconnected by this time,
4165 the result was a log entry for a synchronization error with an empty
4166 string after "input=" when read() returned zero. If read() returned -1
4167 (an event I could not check), uninitialized data bytes were printed.
4168 There were reports of junk text (parts of files, etc) appearing after
4171 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4173 PH/22 Added support for macro redefinition, and (re)definition in between
4174 driver and ACL definitions.
4176 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4177 forgetting to use the resulting value; it was using the unexpanded value.
4179 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4180 hadn't been configured. The fix is from Juergen Kreileder, who
4181 understands it better than I do:
4183 "Here's what I see happening with three configured cyrus_sasl
4184 authenticators configured (plain, login, cram-md5):
4186 On startup auth_cyrus_sasl_init() gets called for each of these.
4187 This means three calls to sasl_listmech() without a specified mech_list.
4188 => SASL tests which mechs of all available mechs actually work
4189 => three warnings about OTP not working
4190 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4192 With the patch, sasl_listmech() also gets called three times. But now
4193 SASL's mech_list option is set to the server_mech specified in the the
4194 authenticator. Or in other words, the answer from sasl_listmech()
4195 gets limited to just the mech you're testing for (which is different
4197 => the return list contains just 'plain' or 'login', 'cram-md5' or
4198 nothing depending on the value of ob->server_mech.
4200 I've just tested the patch: Authentication still works fine,
4201 unavailable mechs specified in the exim configuration are still
4202 caught, and the auth.log warnings about OTP are gone."
4204 PH/25 When debugging is enabled, the contents of the command line are added
4205 to the debugging output, even when log_selector=+arguments is not
4208 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4209 answer is "GNU", and only if the return is "GNU/something" is the answer
4212 PH/27 $acl_verify_message is now set immediately after the failure of a
4213 verification in an ACL, and so is available in subsequent modifiers. In
4214 particular, the message can be preserved by coding like this:
4216 warn !verify = sender
4217 set acl_m0 = $acl_verify_message
4219 Previously, $acl_verify_message was set only while expanding "message"
4220 and "log_message" when a very denied access.
4222 PH/28 Modified OS/os.c-Linux with
4224 -#ifndef OS_LOAD_AVERAGE
4225 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4227 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4228 nomenclature these days.)
4230 PH/29 Installed patch from the Sieve maintainer that adds the options
4231 sieve_useraddress and sieve_subaddress to the redirect router.
4233 PH/30 In these circumstances:
4234 . Two addresses routed to the same list of hosts;
4235 . First host does not offer TLS;
4236 . First host accepts first address;
4237 . First host gives temporary error to second address;
4238 . Second host offers TLS and a TLS session is established;
4239 . Second host accepts second address.
4240 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4241 and peerdn, if requested) that were in fact used only for the second
4244 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4245 attention to any local part prefix or suffix that was matched by the
4246 router that accepted the address. It now behaves in the same way as it
4247 does for delivery: the affixes are removed from the local part unless
4248 rcpt_include_affixes is set on the transport.
4250 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4251 timeout during the DATA phase of an incoming message.
4253 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4254 applied a patch sanctioned by the Sieve maintainer.
4256 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4257 the uid or gid is negative. A case of a negative gid caused this to be
4258 noticed. The fix allows for either to be negative.
4260 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4261 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4264 PH/36 The change PH/12 above was broken. Fixed it.
4266 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4267 the grounds that routing the same address twice would always produce the
4268 same answer. This might have been true once, but it is certainly no
4269 longer true now. Routing a child address may depend on the previous
4270 routing that produced that child. Some complicated redirection strategies
4271 went wrong when messages had multiple recipients, and made Exim's
4272 behaviour dependent on the order in which the addresses were given.
4274 I have moved the duplicate checking until after the routing is complete.
4275 Exim scans the addresses that are assigned to local and remote
4276 transports, and removes any duplicates. This means that more work will be
4277 done, as duplicates will always all be routed, but duplicates are
4278 presumably rare, so I don't expect this is of any significance.
4280 For deliveries to pipes, files, and autoreplies, the duplicate checking
4281 still happens during the routing process, since they are not going to be
4284 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4285 It corrects a timeout issue with spamd. This is Ian's comment: "The
4286 background is that sometimes spamd either never reads data from a
4287 connection it has accepted, or it never writes response data. The exiscan
4288 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4289 blindly assumes that writes won't block so it may never time out."
4291 PH/39 Allow G after quota size as well as K and M.
4293 PH/40 The value set for $authenticated_id in an authenticator may not contain
4294 binary zeroes or newlines because the value is written to log lines and
4295 to spool files. There was no check on this. Now the value is run through
4296 the string_printing() function so that such characters are converted to
4297 printable escape sequences.
4299 PH/41 $message_linecount is a new variable that contains the total number of
4300 lines in the message. Compare $body_linecount, which is the count for the
4303 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4304 bounce and delay warning messages, except in certain special cases, which
4307 (a) An SMTP error message from a remote host;
4308 (b) A message specified in a :fail: redirection;
4309 (c) A message specified in a "fail" command in a system filter;
4310 (d) A message specified in a FAIL return from the queryprogram router;
4311 (e) A message specified by the cannot_route_message router option.
4313 In these cases only, Exim does include the error details in bounce and
4314 warning messages. There are also a few cases where bland messages such
4315 as "unrouteable address" or "local delivery error" are given.
4317 PH/43 $value is now also set for the "else" part of a ${run expansion.
4319 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4320 being worked on, but at least Exim now implements the latest version to
4323 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4324 process to complete was treated as a delivery failure, a timeout while
4325 writing the message to the pipe was logged, but erroneously treated as a
4326 successful delivery. Such timeouts include transport filter timeouts. For
4327 consistency with the overall process timeout, these timeouts are now
4328 treated as errors, giving rise to delivery failures by default. However,
4329 there is now a new Boolean option for the pipe transport called
4330 timeout_defer, which, if set TRUE, converts the failures into defers for
4331 both kinds of timeout. A transport filter timeout is now identified in
4334 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4335 systems where "make" and "gmake" are different, calling "gmake" at top
4336 level broke things. I've arranged for the value of $(MAKE) to be passed
4337 from the Makefile to this script so that it can call the same version of
4341 A note about Exim versions 4.44 and 4.50
4342 ----------------------------------------
4344 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4345 changes of various kinds. As a consequence, a big documentation update was
4346 needed. This delayed the release for rather longer than seemed good, especially
4347 in the light of a couple of (minor) security issues. Therefore, the changes
4348 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4349 release. So 4.44 and 4.50 are in effect two different branches that both start
4352 I have left the 4.50 change log unchanged; it contains all the changes since
4353 4.43. The change log for 4.44 is below; many of its items are identical to
4354 those for 4.50. This seems to be the most sensible way to preserve the
4355 historical information.
4361 1. Minor wording change to the doc/README.SIEVE file.
4363 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4364 computation of the current number of files was incorrect.
4366 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4367 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4370 4. Give more explanation in the error message when the command for a transport
4371 filter fails to execute.
4373 5. There are several places where Exim runs a non-Exim command in a
4374 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4375 was being done only for the command run by the queryprogram router. It is
4376 now done for all such subprocesses. The other cases are: ${run, transport
4377 filters, and the commands run by the lmtp and pipe transports.
4379 6. Added CONFIGURE_GROUP build-time option.
4381 7. Some older OS have a limit of 256 on the maximum number of file
4382 descriptors. Exim was using setrlimit() to set 1000 as a large value
4383 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4384 systems. I've change it so that if it can't get 1000, it tries for 256.
4386 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4387 was an oversight, and furthermore, ever since the addition of extra
4388 controls (e.g. 4.43/32), the checks on when to allow different forms of
4389 "control" were broken. There should now be diagnostics for all cases when a
4390 control that does not make sense is encountered.
4392 9. Added the /retain_sender option to "control=submission".
4394 10. $recipients is now available in the predata ACL (oversight).
4396 11. Tidy the search cache before the fork to do a delivery from a message
4397 received from the command line. Otherwise the child will trigger a lookup
4398 failure and thereby defer the delivery if it tries to use (for example) a
4399 cached ldap connection that the parent has called unbind on.
4401 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4402 of $address_data from the recipient verification was clobbered by the
4403 sender verification.
4405 13. The value of address_data from a sender verification is now available in
4406 $sender_address_data in subsequent conditions in the ACL statement.
4408 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4410 15. Added a new option "connect=<time>" to callout options, to set a different
4413 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4414 was its contents. (It was OK if the option was not defined at all.)
4416 17. A "Completed" log line is now written for messages that are removed from
4417 the spool by the -Mrm option.
4419 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4420 information about exactly what failed.
4422 19. Added -dd to debug only the daemon process.
4424 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4425 handles timeouts, both on the server side and network timeouts. Renamed the
4426 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4428 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4429 It is now set to "smtps".
4431 22. $host_address is now set to the target address during the checking of
4432 ignore_target_hosts.
4434 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4435 being passed; this would have caused $sender_host_name to have been used if
4436 matching the list had actually called for a host name (not very likely,
4437 since this list is usually IP addresses). A host name is now passed as
4440 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4441 code that specifies a non-restarting handler (typically sigaction() in
4442 modern systems) in an attempt to fix a rare and obscure crash bug.
4444 25. Narrowed the window for a race in the daemon that could cause it to ignore
4445 SIGCHLD signals. This is not a major problem, because they are used only to
4446 wake it up if nothing else does.
4448 26. A malformed maildirsize file could cause Exim to calculate negative values
4449 for the mailbox size or file count. Odd effects could occur as a result.
4450 The maildirsize information is now recalculated if the size or filecount
4453 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4454 support for a long time. Removed HAVE_SYS_VFS_H.
4456 28. Installed the latest version of exipick from John Jetmore.
4458 29. In an address list, if the pattern was not a regular expression, an empty
4459 subject address (from a bounce message) matched only if the pattern was an
4460 empty string. Non-empty patterns were not even tested. This was the wrong
4461 because it is perfectly reasonable to use an empty address as part of a
4462 database query. An empty address is now tested by patterns that are
4463 lookups. However, all the other forms of pattern expect the subject to
4464 contain a local part and a domain, and therefore, for them, an empty
4465 address still always fails if the pattern is not itself empty.
4467 30. Exim went into a mad DNS loop when attempting to do a callout where the
4468 host was specified on an smtp transport, and looking it up yielded more
4469 than one IP address.
4471 31. Re-factored the code for checking spool and log partition space into a
4472 function that finds that data and another that does the check. The former
4473 is then used to implement four new variables: $spool_space, $log_space,
4474 $spool_inodes, and $log_inodes.
4476 32. The RFC2047 encoding function was originally intended for short strings
4477 such as real names; it was not keeping to the 75-character limit for
4478 encoded words that the RFC imposes. It now respects the limit, and
4479 generates multiple encoded words if necessary. To be on the safe side, I
4480 have increased the buffer size for the ${rfc2047: expansion operator from
4483 33. It is now permitted to omit both strings after an "if" condition; if the
4484 condition is true, the result is "true". As before, when the second string
4485 is omitted, a false condition yields an empty string. This makes it less
4486 cumbersome to write custom ACL and router conditions.
4488 34. Failure to deliver a bounce message always caused it to be frozen, even if
4489 there was an errors_to setting on the router. The errors_to setting is now
4492 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4493 canonical form (fully expanded) before being placed in
4494 $sender_host_address.
4496 36. The table in the code that translates DNS record types into text (T_A to
4497 "A" for instance) was missing entries for NS and CNAME. It is just possible
4498 that this could have caused confusion if both these types were looked up
4499 for the same domain, because the text type is used as part of Exim's
4500 per-process caching. But the chance of anyone hitting this buglet seems
4503 37. The dnsdb lookup has been extended in a number of ways.
4505 (1) There is a new type, "zns", which walks up the domain tree until it
4506 finds some nameserver records. It should be used with care.
4508 (2) There is a new type, "mxh", which is like "mx" except that it returns
4509 just the host names, not the priorities.
4511 (3) It is now possible to give a list of domains (or IP addresses) to be
4512 looked up. The behaviour when one of the lookups defers can be
4513 controlled by a keyword.
4515 (4) It is now possible to specify the separator character for use when
4516 multiple records are returned.
4518 38. The dnslists ACL condition has been extended: it is now possible to supply
4519 a list of IP addresses and/or domains to be looked up in a particular DNS
4522 39. Added log_selector=+queue_time_overall.
4524 40. When running the queue in the test harness, wait just a tad after forking a
4525 delivery process, to get repeatability of debugging output.
4527 41. Include certificate and key file names in error message when GnuTLS fails
4528 to set them up, because the GnuTLS error message doesn't include the name
4529 of the failing file when there is a problem reading it.
4531 42. Allow both -bf and -bF in the same test run.
4533 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4535 44. The "Exiscan patch" is now merged into the mainline Exim source.
4537 45. Sometimes the final signoff response after QUIT could fail to get
4538 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4539 < 0 before doing a fflush(). This bug looks as though it goes back to the
4540 introduction of TLS in release 3.20, but "sometimes" must have been rare
4541 because the tests only now provoked it.
4543 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4544 (this can affect the format of dates).
4546 47. exim_tidydb, when checking for the continued existence of a message for
4547 which it has found a message-specific retry record, was not finding
4548 messages that were in split spool directories. Consequently, it was
4549 deleting retry records that should have stayed in existence.
4551 48. Steve fixed some bugs in eximstats.
4553 49. The SPA authentication driver was not abandoning authentication and moving
4554 on to the next authenticator when an expansion was forced to fail,
4555 contradicting the general specification for all authenticators. Instead it
4556 was generating a temporary error. It now behaves as specified.
4558 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4559 (the order specifies the preference for clients). The order is now AES256,
4560 AES128, 3DES, ARCFOUR128.
4562 51. Small patch to Sieve code - explicitly set From: when generating an
4565 52. Exim crashed if a remote delivery caused a very long error message to be
4566 recorded - for instance if somebody sent an entire SpamAssassin report back
4567 as a large number of 550 error lines. This bug was coincidentally fixed by
4568 increasing the size of one of Exim's internal buffers (big_buffer) that
4569 happened as part of the Exiscan merge. However, to be on the safe side, I
4570 have made the code more robust (and fixed the comments that describe what
4573 53. Now that there can be additional text after "Completed" in log lines (if
4574 the queue_time_overall log selector is set), a one-byte patch to exigrep
4575 was needed to allow it to recognize "Completed" as not the last thing in
4578 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4579 patch that reportedly fixes this has been added. I am not expert enough to
4580 create a test for it. This is what the patch creator wrote:
4582 "I found a little strange behaviour of ldap code when working with
4583 Windows 2003 AD Domain, where users was placed in more than one
4584 Organization Units. When I tried to give exim partial DN, the exit code
4585 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4586 But simultaneously result of request was absolutely normal ldap result,
4587 so I produce this patch..."
4589 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4590 REFERENCE, so I have modified the code to exclude the patch when that macro
4593 55. Some experimental protocols are using DNS PTR records for new purposes. The
4594 keys for these records are domain names, not reversed IP addresses. The
4595 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4596 leaves it alone. Component reversal etc. now happens only for IP addresses.
4599 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4601 57. Double the size of the debug message buffer (to 2048) so that more of very
4602 long debug lines gets shown.
4604 58. The exicyclog utility now does better if the number of log files to keep
4605 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4607 59. Two changes related to the smtp_active_hostname option:
4609 (1) $smtp_active_hostname is now available as a variable.
4610 (2) The default for smtp_banner uses $smtp_active_hostname instead
4611 of $primary_hostname.
4613 60. The host_aton() function is supposed to be passed a string that is known
4614 to be a valid IP address. However, in the case of IPv6 addresses, it was
4615 not checking this. This is a hostage to fortune. Exim now panics and dies
4616 if the condition is not met. A case was found where this could be provoked
4617 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4618 components; fortuitously, this particular loophole had already been fixed
4619 by change 4.50/55 above.
4621 If there are any other similar loopholes, the new check in host_aton()
4622 itself should stop them being exploited. The report I received stated that
4623 data on the command line could provoke the exploit when Exim was running as
4624 exim, but did not say which command line option was involved. All I could
4625 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4626 running as the user.
4629 61. There was a buffer overflow vulnerability in the SPA authentication code
4630 (which came originally from the Samba project). I have added a test to the
4631 spa_base64_to_bits() function which I hope fixes it.
4634 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4635 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4637 63. The daemon start-up calls getloadavg() while still root for those OS that
4638 need the first call to be done as root, but it missed one case: when
4639 deliver_queue_load_max is set with deliver_drop_privilege. This is
4640 necessary for the benefit of the queue runner, because there is no re-exec
4641 when deliver_drop_privilege is set.
4643 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4644 This has been fixed.
4646 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4647 was in use, was not putting the data itself into the right store pool;
4648 consequently, it could be overwritten for a subsequent message in the same
4649 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4652 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4654 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4655 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4656 the function were treating the return as a boolean value, which happened to
4657 work because 0=false and not-0=true, but is not correct code.
4659 68. The host_aton() function was not handling scoped IPv6 addresses (those
4660 with, for example, "%eth0" on the end) correctly.
4662 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4663 negated items (that is, ~something) in unsigned ints. Some compilers
4664 apparently mutter when there is no cast.
4666 70. If an address verification called from an ACL failed, and did not produce a
4667 user-specific message (i.e. there was only a "system" message), nothing was
4668 put in $acl_verify_message. In this situation, it now puts the system
4671 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4672 session; change 4.31/43 added the unwanted input to the log line - except
4673 that it did not do this in the start of session case. It now does.
4675 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4676 This is wrong and can cause the other end to generate a synchronization
4677 error if it is another Exim or anything else that does the synchronization
4678 check. A QUIT command is no longer sent after a timeout.
4680 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4681 during host lookups.
4683 74. The defer_ok option of callout verification was not working if it was used
4684 when verifying addresses in header lines, that is, for this case:
4686 verify = header_sender/callout=defer_ok
4688 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4689 those file descriptors could be used for SMTP connections. If anything
4690 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4691 could be sent to the SMTP client, causing chaos. The daemon now opens
4692 stdin, stdout, and stderr to /dev/null when it puts itself into the
4695 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4696 log by default. The user can override this with suitable Perl magic.
4698 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4699 the log message when discard triggers, was not working for the DATA ACL or
4700 for the non-SMTP ACL.
4702 78. Error message wording change in sieve.c.
4704 79. If smtp_accept_max_per_host was set, the number of connections could be
4705 restricted to fewer than expected, because the daemon was trying to set up
4706 a new connection before checking whether the processes handling previous
4707 connections had finished. The check for completed processes is now done
4708 earlier. On busy systems, this bug wouldn't be noticed because something
4709 else would have woken the daemon, and it would have reaped the completed
4712 80. If a message was submitted locally by a user whose login name contained one
4713 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4714 It caused a spool format error. I have fixed the spool reading code. A
4715 related problem was that the "from" clause in the Received: line became
4716 illegal because of the space(s). It is now covered by ${quote_local_part.
4718 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4721 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4723 83. Patch from Sieve maintainer to fix unterminated string problem in
4724 "vacation" handling.
4726 84. Some minor changes to the Linux configuration files to help with other
4727 OS variants using glibc.
4729 85. One more patch for Sieve to update vacation handling to latest spec.
4732 ----------------------------------------------------
4733 See the note above about the 4.44 and 4.50 releases.
4734 ----------------------------------------------------
4740 1. Change 4.43/35 introduced a bug that caused file counts to be
4741 incorrectly computed when quota_filecount was set in an appendfile
4744 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4745 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4748 3. Give more explanation in the error message when the command for a transport
4749 filter fails to execute.
4751 4. There are several places where Exim runs a non-Exim command in a
4752 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4753 was being done only for the command run by the queryprogram router. It is
4754 now done for all such subprocesses. The other cases are: ${run, transport
4755 filters, and the commands run by the lmtp and pipe transports.
4757 5. Some older OS have a limit of 256 on the maximum number of file
4758 descriptors. Exim was using setrlimit() to set 1000 as a large value
4759 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4760 systems. I've change it so that if it can't get 1000, it tries for 256.
4762 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4763 was an oversight, and furthermore, ever since the addition of extra
4764 controls (e.g. 4.43/32), the checks on when to allow different forms of
4765 "control" were broken. There should now be diagnostics for all cases when a
4766 control that does not make sense is encountered.
4768 7. $recipients is now available in the predata ACL (oversight).
4770 8. Tidy the search cache before the fork to do a delivery from a message
4771 received from the command line. Otherwise the child will trigger a lookup
4772 failure and thereby defer the delivery if it tries to use (for example) a
4773 cached ldap connection that the parent has called unbind on.
4775 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4776 of $address_data from the recipient verification was clobbered by the
4777 sender verification.
4779 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4780 was its contents. (It was OK if the option was not defined at all.)
4782 11. A "Completed" log line is now written for messages that are removed from
4783 the spool by the -Mrm option.
4785 12. $host_address is now set to the target address during the checking of
4786 ignore_target_hosts.
4788 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4789 being passed; this would have caused $sender_host_name to have been used if
4790 matching the list had actually called for a host name (not very likely,
4791 since this list is usually IP addresses). A host name is now passed as
4794 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4795 code that specifies a non-restarting handler (typically sigaction() in
4796 modern systems) in an attempt to fix a rare and obscure crash bug.
4798 15. Narrowed the window for a race in the daemon that could cause it to ignore
4799 SIGCHLD signals. This is not a major problem, because they are used only to
4800 wake it up if nothing else does.
4802 16. A malformed maildirsize file could cause Exim to calculate negative values
4803 for the mailbox size or file count. Odd effects could occur as a result.
4804 The maildirsize information is now recalculated if the size or filecount
4807 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4808 support for a long time. Removed HAVE_SYS_VFS_H.
4810 18. Updated exipick to current release from John Jetmore.
4812 19. Allow an empty sender to be matched against a lookup in an address list.
4813 Previously the only cases considered were a regular expression, or an
4816 20. Exim went into a mad DNS lookup loop when doing a callout where the
4817 host was specified on the transport, if the DNS lookup yielded more than
4820 21. The RFC2047 encoding function was originally intended for short strings
4821 such as real names; it was not keeping to the 75-character limit for
4822 encoded words that the RFC imposes. It now respects the limit, and
4823 generates multiple encoded words if necessary. To be on the safe side, I
4824 have increased the buffer size for the ${rfc2047: expansion operator from
4827 22. Failure to deliver a bounce message always caused it to be frozen, even if
4828 there was an errors_to setting on the router. The errors_to setting is now
4831 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4832 canonical form (fully expanded) before being placed in
4833 $sender_host_address.
4835 24. Updated eximstats to version 1.33
4837 25. Include certificate and key file names in error message when GnuTLS fails
4838 to set them up, because the GnuTLS error message doesn't include the name
4839 of the failing file when there is a problem reading it.
4841 26. Expand error message when OpenSSL has problems setting up cert/key files.
4844 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4845 (this can affect the format of dates).
4847 28. exim_tidydb, when checking for the continued existence of a message for
4848 which it has found a message-specific retry record, was not finding
4849 messages that were in split spool directories. Consequently, it was
4850 deleting retry records that should have stayed in existence.
4852 29. eximstats updated to version 1.35
4853 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4854 1.35 - bugfix such that pie charts by volume are generated correctly
4856 30. The SPA authentication driver was not abandoning authentication and moving
4857 on to the next authenticator when an expansion was forced to fail,
4858 contradicting the general specification for all authenticators. Instead it
4859 was generating a temporary error. It now behaves as specified.
4861 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4862 (the order specifies the preference for clients). The order is now AES256,
4863 AES128, 3DES, ARCFOUR128.
4865 31. Small patch to Sieve code - explicitly set From: when generating an
4868 32. Exim crashed if a remote delivery caused a very long error message to be
4869 recorded - for instance if somebody sent an entire SpamAssassin report back
4870 as a large number of 550 error lines. This bug was coincidentally fixed by
4871 increasing the size of one of Exim's internal buffers (big_buffer) that
4872 happened as part of the Exiscan merge. However, to be on the safe side, I
4873 have made the code more robust (and fixed the comments that describe what
4876 33. Some experimental protocols are using DNS PTR records for new purposes. The
4877 keys for these records are domain names, not reversed IP addresses. The
4878 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4879 leaves it alone. Component reversal etc. now happens only for IP addresses.
4882 34. The host_aton() function is supposed to be passed a string that is known
4883 to be a valid IP address. However, in the case of IPv6 addresses, it was
4884 not checking this. This is a hostage to fortune. Exim now panics and dies
4885 if the condition is not met. A case was found where this could be provoked
4886 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4887 components; fortuitously, this particular loophole had already been fixed
4888 by change 4.50/55 or 4.44/33 above.
4890 If there are any other similar loopholes, the new check in host_aton()
4891 itself should stop them being exploited. The report I received stated that
4892 data on the command line could provoke the exploit when Exim was running as
4893 exim, but did not say which command line option was involved. All I could
4894 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4895 running as the user.
4898 35. There was a buffer overflow vulnerability in the SPA authentication code
4899 (which came originally from the Samba project). I have added a test to the
4900 spa_base64_to_bits() function which I hope fixes it.
4903 36. The daemon start-up calls getloadavg() while still root for those OS that
4904 need the first call to be done as root, but it missed one case: when
4905 deliver_queue_load_max is set with deliver_drop_privilege. This is
4906 necessary for the benefit of the queue runner, because there is no re-exec
4907 when deliver_drop_privilege is set.
4909 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4910 was in use, was not putting the data itself into the right store pool;
4911 consequently, it could be overwritten for a subsequent message in the same
4912 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4915 38. Sometimes the final signoff response after QUIT could fail to get
4916 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4917 < 0 before doing a fflush(). This bug looks as though it goes back to the
4918 introduction of TLS in release 3.20, but "sometimes" must have been rare
4919 because the tests only now provoked it.
4925 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4926 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4927 Now it returns an integer. A number of calls were still expecting a T/F
4928 return. Fortuitously, in all cases, the tests worked in OK situations,
4929 which is the norm. However, things would have gone wrong on any write
4930 failures on the smtp file descriptor. This function is used when sending
4931 messages over SMTP and also when doing verify callouts.
4933 2. When Exim is called to do synchronous delivery of a locally submitted
4934 message (the -odf or -odi options), it no longer closes stderr before doing
4937 3. Implemented the mua_wrapper option.
4939 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4941 5. Implemented the functions header_remove(), header_testname(),
4942 header_add_at_position(), and receive_remove_recipient(), and exported them
4945 6. If an ACL "warn" statement specified the addition of headers, Exim already
4946 inserted X-ACL-Warn: at the start if there was no header name. However, it
4947 was not making this test for the second and subsequent header lines if
4948 there were newlines in the string. This meant that an invalid header could
4949 be inserted if Exim was badly configured.
4951 7. Allow an ACL "warn" statement to add header lines at the start or after all
4952 the Received: headers, as well as at the end.
4954 8. Added the rcpt_4xx retry error code.
4956 9. Added postmaster_mailfrom=xxx to callout verification option.
4958 10. Added mailfrom=xxxx to the callout verification option, for verify=
4961 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4962 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4964 12. Inserted some casts to stop certain compilers warning when using pointer
4965 differences as field lengths or precisions in printf-type calls (mostly
4966 affecting debugging statements).
4968 13. Added optional readline() support for -be (dynamically loaded).
4970 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4971 same clock tick as a message's arrival, so that its received time was the
4972 same as the "first fail" time on the retry record, and that message
4973 remained on the queue past the ultimate address timeout, every queue runner
4974 would try a delivery (because it was past the ultimate address timeout) but
4975 after another failure, the ultimate address timeout, which should have then
4976 bounced the address, did not kick in. This was a "< instead of <=" error;
4977 in most cases the first failure would have been in the next clock tick
4978 after the received time, and all would be well.
4980 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4981 being recognized when the domain list was tested by the match_domain
4982 condition in an expansion string.
4984 16. Added the ${str2b64: operator.
4986 17. Exim was always calling setrlimit() to set a large limit for the number of
4987 processes, without checking whether the existing limit was already
4988 adequate. (It did check for the limit on file descriptors.) Furthermore,
4989 errors from getrlimit() and setrlimit() were being ignored. Now they are
4990 logged to the main and panic logs, but Exim does carry on, to try to do its
4991 job under whatever limits there are.
4993 18. Imported PCRE 5.0.
4995 19. Trivial typo in log message " temporarily refused connection" (the leading
4998 20. If the log selector return_path_on_delivery was set and an address was
4999 redirected to /dev/null, the delivery process crashed because it assumed
5000 that a return path would always be set for a "successful" delivery. In this
5001 case, the whole delivery is bypassed as an optimization, and therefore no
5004 21. Internal re-arrangement: the function for sending a challenge and reading
5005 a response while authentication was assuming a zero-terminated challenge
5006 string. It's now changed to take a pointer and a length, to allow for
5007 binary data in such strings.
5009 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5011 23. Exim was not respecting finduser_retries when seeking the login of the
5012 uid under which it was called; it was always trying 10 times. (The default
5013 setting of finduser_retries is zero.) Also, it was sleeping after the final
5014 failure, which is pointless.
5016 24. Implemented tls_on_connect_ports.
5018 25. Implemented acl_smtp_predata.
5020 26. If the domain in control=submission is set empty, Exim assumes that the
5021 authenticated id is a complete email address when it generates From: or
5022 Sender: header lines.
5024 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5025 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5026 chown and chgrp in /bin and hostname in /usr/bin.
5028 28. Exim was keeping the "process log" file open after each use, just as it
5029 does for the main log. This opens the possibility of it remaining open for
5030 long periods when the USR1 signal hits a daemon. Occasional processlog
5031 errors were reported, that could have been caused by this. Anyway, it seems
5032 much more sensible not to leave this file open at all, so that is what now
5035 29. The long-running daemon process does not normally write to the log once it
5036 has entered its main loop, and it closes the log before doing so. This is
5037 so that log files can straightforwardly be renamed and moved. However,
5038 there are a couple of unusual error situations where the daemon does write
5039 log entries, and I had neglected to close the log afterwards.
5041 30. The text of an SMTP error response that was received during a remote
5042 delivery was being truncated at 512 bytes. This is too short for some of
5043 the long messages that one sometimes sees. I've increased the limit to
5046 31. It is now possible to make retry rules that apply only when a message has a
5047 specific sender, in particular, an empty sender.
5049 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5050 it possible to be selective about when SMTP synchronization is enforced.
5052 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5054 32. Implemented hosts_connection_nolog.
5056 33. Added an ACL for QUIT.
5058 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5061 35. Added mailbox_size and mailbox_filecount to appendfile.
5063 36. Added control = no_multiline_responses to ACLs.
5065 37. There was a bug in the logic of the code that waits for the clock to tick
5066 in the case where the clock went backwards by a substantial amount such
5067 that the microsecond fraction of "now" was more than the microsecond
5068 fraction of "then" (but the whole seconds number was less).
5070 38. Added support for the libradius Radius client library this is found on
5071 FreeBSD (previously only the radiusclient library was supported).
5077 1. When certain lookups returned multiple values in the form name=value, the
5078 quoting of the values was not always being done properly. Specifically:
5079 (a) If the value started with a double quote, but contained no whitespace,
5081 (b) If the value contained whitespace other than a space character (i.e.
5082 tabs or newlines or carriage returns) it was not quoted.
5083 This fix has been applied to the mysql and pgsql lookups by writing a
5084 separate quoting function and calling it from the lookup code. The fix
5085 should probably also be applied to nisplus, ibase and oracle lookups, but
5086 since I cannot test any of those, I have not disturbed their existing code.
5088 2. A hit in the callout cache for a specific address caused a log line with no
5089 reason for rejecting RCPT. Now it says "Previous (cached) callout
5090 verification failure".
5092 3. There was an off-by-one bug in the queryprogram router. An over-long
5093 return line was truncated at 256 instead of 255 characters, thereby
5094 overflowing its buffer with the terminating zero. As well as fixing this, I
5095 have increased the buffer size to 1024 (and made a note to document this).
5097 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5098 when Exim is waiting for an SMTP response from a remote server, Exim
5099 restarts its select() call on the socket, thereby resetting its timeout.
5100 This is not a problem when such interrupts are rare. Somebody set up a cron
5101 job to run exiwhat every 2 minutes, which is less than the normal select()
5102 timeout (5 or 10 minutes). This meant that the select() timeout never
5103 kicked in because it was always reset. I have fixed this by comparing the
5104 time when an interrupt arrives with the time at the start of the first call
5105 to select(). If more time than the timeout has elapsed, the interrupt is
5106 treated as a timeout.
5108 5. Some internal re-factoring in preparation for the addition of Sieve
5109 extensions (by MH). In particular, the "personal" test is moved to a
5110 separate function, and given an option for scanning Cc: and Bcc: (which is
5111 not set for Exim filters).
5113 6. When Exim created an email address using the login of the caller as the
5114 local part (e.g. when creating a From: or Sender: header line), it was not
5115 quoting the local part when it contained special characters such as @.
5117 7. Installed new OpenBSD configuration files.
5119 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5120 try to make them clearer.
5122 9. Callout options, other than the timeout value, were being ignored when
5123 verifying sender addresses in header lines. For example, when using
5125 verify = header_sender/callout=no_cache
5127 the cache was (incorrectly) being used.
5129 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5130 only the Cygwin environment.
5132 11. When return_path_on_delivery was set as a log selector, if different remote
5133 addresses in the same message used different return paths and parallel
5134 remote delivery occurred, the wrong values would sometimes be logged.
5135 (Whenever a remote delivery process finished, the return path value from
5136 the most recently started remote delivery process was logged.)
5138 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5139 lines when AUTH and/or TLS are in use. This is the "received protocol"
5140 field. Exim used to use "asmtp" for authenticated SMTP, without any
5141 indication (in the protocol name) for TLS use. Now it follows the RFC and
5142 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5143 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5144 names appear in log lines as well as in Received: header lines.
5146 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5147 extensions, and comparison tests, and to fix some bugs.
5149 14. Changes to the "personal" filter test:
5151 (1) The test was buggy in that it was just doing the equivalent of
5152 "contains" tests on header lines. For example, if a user's address was
5153 anne@some.where, the "personal" test would incorrectly be true for
5155 To: susanne@some.where
5157 This test is now done by extracting each address from the header in turn,
5158 and checking the entire address. Other tests that are part of "personal"
5159 are now done using regular expressions (for example, to check local parts
5160 of addresses in From: header lines).
5162 (2) The list of non-personal local parts in From: addresses has been
5163 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5164 taken from the Sieve specification recommendations.
5166 (3) If the message contains any header line starting with "List-" it is
5167 treated as non-personal.
5169 (4) The test for "circular" in the Subject: header line has been removed
5170 because it now seems ill-conceived.
5172 15. Minor typos in src/EDITME comments corrected.
5174 16. Installed latest exipick from John Jetmore.
5176 17. If headers_add on a router specified a text string that was too long for
5177 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5178 of string_sprintf() is now avoided.
5180 18. $message_body_size was not set (it was always zero) when running the DATA
5181 ACL and the local_scan() function.
5183 19. For the "mail" command in an Exim filter, no default was being set for
5184 the once_repeat time, causing a random time value to be used if "once" was
5185 specified. (If the value happened to be <= 0, no repeat happened.) The
5186 default is now 0s, meaning "never repeat". The "vacation" command was OK
5187 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5188 (I found it when inspecting the code).
5190 20. There is now an overall timeout for performing a callout verification. It
5191 defaults to 4 times the callout timeout, which applies to individual SMTP
5192 commands during the callout. The overall timeout applies when there is more
5193 than one host that can be tried. The timeout is checked before trying the
5194 next host. This prevents very long delays if there are a large number of
5195 hosts and all are timing out (e.g. when the network connections are timing
5196 out). The value of the overall timeout can be changed by specifying an
5197 additional sub-option for "callout", called "maxwait". For example:
5199 verify = sender/callout=5s,maxwait=20s
5201 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5202 to the end before writing, but this should make it even safer).
5204 22. Exim was forgetting that it had advertised PIPELINING for the second and
5205 subsequent messages on an SMTP connection. It was also not resetting its
5206 memory on STARTTLS and an internal HELO.
5208 23. When Exim logs an SMTP synchronization error within a session, it now
5209 records whether PIPELINING has been advertised or not.
5211 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5212 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5213 rather than long int.
5215 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5217 26. Added the never_mail option to autoreply.
5223 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5224 crash if the getsockname() call failed; this can happen if a connection is
5225 closed very soon after it is established. The problem was simply in the
5226 order in which certain operations were done, causing Exim to try to write
5227 to the SMTP stream before it had set up the file descriptor. The bug has
5228 been fixed by making things happen in the correct order.
5234 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5235 before the connection was closed, thus losing the rejection response.
5237 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5238 some early Solaris releases, but causes trouble in current releases where
5239 socklen_t is defined.
5241 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5244 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5247 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5248 that releasing the top of it at the end releases what was used for sub-
5249 expansions (unless the block got too big). However, discard this block if
5250 the first thing is a variable or header, so that we can use its block when
5251 it is dynamic (useful for very large $message_headers, for example).
5253 6. Lookups now cache *every* query, not just the most recent. A new, separate
5254 store pool is used for this. It can be recovered when all lookup caches are
5255 flushed. Lookups now release memory at the end of their result strings.
5256 This has involved some general refactoring of the lookup sources.
5258 7. Some code has been added to the store_xxx() functions to reduce the amount
5259 of flapping under certain conditions.
5261 8. log_incoming_interface used to affect only the <= reception log lines. Now
5262 it causes the local interface and port to be added to several more SMTP log
5263 lines, for example "SMTP connection from", and rejection lines.
5265 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5267 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5269 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5270 could be overwritten at the end of the current message (or the start of a
5271 new message if it was set in a HELO ACL). The value is now preserved for
5272 the duration of the SMTP connection.
5274 12. If a transport had a headers_rewrite setting, and a matching header line
5275 contained an unqualified address, that address was qualified, even if it
5276 did not match any rewriting rules. The underlying bug was that the values
5277 of the flags that permit the existence of unqualified sender and recipient
5278 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5279 non-local messages, and by -bnq for local messages) were not being
5280 preserved with the message after it was received.
5282 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5283 "next input=" as part of the text comprising the host identity instead of
5284 the correct text. The code was using the same buffer for two different
5285 strings. However, depending on which order the printing function evaluated
5286 its arguments, the bug did not always show up. Under Linux, for example, my
5287 test suite worked just fine.
5289 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5290 doesn't work with some older versions of Perl. It has been changed to "my",
5291 which in any case is probably the better facility to use.
5293 15. A really picky compiler found some instances of statements for creating
5294 error messages that either had too many or two few arguments for the format
5297 16. The size of the buffer for calls to the DNS resolver has been increased
5298 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5299 for addresses that have a lot of PTR records. This alleviates a problem; it
5300 does not fully solve it.
5302 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5303 buffer now truncates the list and logs the incident, which is the same
5304 action as happens when Exim is looking up a host name and its aliases.
5305 Previously in this situation something unpredictable would happen;
5306 sometimes it was "internal error: store_reset failed".
5308 18. If a server dropped the connection unexpectedly when an Exim client was
5309 using GnuTLS and trying to read a response, the client delivery process
5310 crashed while trying to generate an error log message.
5312 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5313 string, for example:
5315 warn message = H1: something\nH2: something
5317 the text was added as a single header line from Exim's point of view
5318 though it ended up OK in the delivered message. However, searching for the
5319 second and subsequent header lines using $h_h2: did not work. This has been
5320 fixed. Similarly, if a system filter added multiple headers in this way,
5321 the routers could not see them.
5323 20. Expanded the error message when iplsearch is called with an invalid key to
5324 suggest using net-iplsearch in a host list.
5326 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5327 ACLs are no longer actually imposed (and a message to that effect is
5330 22. If a "gecos" field in a passwd entry contained escaped characters, in
5331 particular, if it contained a \" sequence, Exim got it wrong when building
5332 a From: or a Sender: header from that name. A second bug also caused
5333 incorrect handling when an unquoted " was present following a character
5334 that needed quoting.
5336 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5337 was not being matched caselessly.
5339 24. Arranged for all hyphens in the exim.8 source to be escaped with
5342 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5343 properly. Recipient callout cache records were still being keyed to include
5344 the sender, even when use_sender was set false. This led to far more
5345 callouts that were necessary. The sender is no longer included in the key
5346 when use_sender is false.
5348 26. Added "control = submission" modifier to ACLs.
5350 27. Added the ${base62d: operator to decode base 62 numbers.
5352 28. dnsdb lookups can now access SRV records.
5354 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5355 the configuration file.
5357 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5358 (-v) mode. This makes the output for a verbose queue run more intelligible.
5360 31. Added a use_postmaster feature to recipient callouts.
5362 32. Added the $body_zerocount variable, containing the number of binary zero
5363 bytes in the message body.
5365 33. The time of last modification of the "new" subdirectory is now used as the
5366 "mailbox time last read" when there is a quota error for a maildir
5369 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5371 35. Added +ignore_unknown as a special item in host lists.
5373 36. Code for decoding IPv6 addresses in host lists is now included, even if
5374 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5375 address was recognized as an IP address, but was then not correctly decoded
5376 into binary, causing unexpected and incorrect effects when compared with
5383 1. Very minor rewording of debugging text in manualroute to say "list of
5384 hosts" instead of "hostlist".
5386 2. If verify=header_syntax was set, and a header line with an unqualified
5387 address (no domain) and a large number of spaces between the end of the
5388 name and the colon was received, the reception process suffered a buffer
5389 overflow, and (when I tested it) crashed. This was caused by some obsolete
5390 code that should have been removed. The fix is to remove it!
5392 3. When running in the test harness, delay a bit after writing a bounce
5393 message to get a bit more predictability in the log output.
5395 4. Added a call to search_tidyup() just before forking a reception process. In
5396 theory, someone could use a lookup in the expansion of smtp_accept_max_
5397 per_host which, without the tidyup, could leave open a database connection.
5399 5. Added the variables $recipient_data and $sender_data which get set from a
5400 lookup success in an ACL "recipients" or "senders" condition, or a router
5401 "senders" option, similar to $domain_data and $local_part_data.
5403 6. Moved the writing of debug_print from before to after the "senders" test
5406 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5407 problems for message scanning, either using a data ACL, or using
5408 local_scan() because the Received: header was not generated till after they
5409 were called (in order to set the time as the time of reception completion).
5410 I have revised the way this works. The header is now generated after the
5411 body is received, but before the ACL or local_scan() are called. After they
5412 are run, the timestamp in the header is updated.
5418 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5419 before starting a queue runner without re-exec. This happened only when
5420 deliver_drop_privilege was set or when the Exim user was set to root. The
5421 effect of the bug was that timeouts during subsequent deliveries caused
5422 crashes instead of being properly handled. The handler is now left at its
5423 default (and expected) setting.
5425 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5426 message, again when deliver_drop_privilege is set or Exim is run as root.
5427 The bug described in (1) was not present in this case, but the tidying up
5428 of the other signals was missing. I have made the two cases consistent.
5430 3. The ignore_target_hosts setting on a manualroute router was being ignored
5431 for hosts that were looked up using the /MX notation.
5433 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5436 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5437 operated on the sender address. After changing the $sender_address to <>
5438 for the sender address verify, Exim was re-instated it as the original
5439 (before rewriting) address, but remembering that it had rewritten it, so it
5440 wasn't rewriting it again. This bug also had the effect of breaking the
5441 sender address verification caching when the sender address was rewritten.
5443 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5444 This has been changed so that if the ip literal address matches
5445 ignore_target_hosts, the router declines.
5447 7. Added expansion conditions match_domain, match_address, and match_local_
5448 part (NOT match_host).
5450 8. The placeholder for the Received: header didn't have a length field set.
5452 9. Added code to Exim itself and to exim_lock to test for a specific race
5453 condition that could lead to file corruption when using MBX delivery. The
5454 issue is with the lockfile that is created in /tmp. If this file is removed
5455 after a process has opened it but before that process has acquired a lock,
5456 there is the potential for a second process to recreate the file and also
5457 acquire a lock. This could lead to two Exim processes writing to the file
5458 at the same time. The added code performs the same test as UW imapd; it
5459 checks after acquiring the lock that its file descriptor still refers to
5460 the same named file.
5462 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5463 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5466 11. Added the smtp_active_hostname option. If used, this will typically be made
5467 to depend on the incoming interface address. Because $interface_address is
5468 not set up until the daemon has forked a reception process, error responses
5469 that can happen earlier (such as "too many connections") no longer contain
5472 12. If an expansion in a condition on a "warn" statement fails because a lookup
5473 defers, the "warn" statement is abandoned, and the next ACL statement is
5474 processed. Previously this caused the whole ACL to be aborted.
5476 13. Added the iplsearch lookup type.
5478 14. Added ident_timeout as a log selector.
5480 15. Added tls_certificate_verified as a log selector.
5482 16. Added a global option tls_require_ciphers (compare the smtp transport
5483 option of the same name). This controls incoming TLS connections.
5485 17. I finally figured out how to make tls_require_ciphers do a similar thing
5486 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5487 before starting the TLS session.
5489 18. Tabs are now shown as \t in -bP output.
5491 19. If the log selector return_path_on_delivery was set, Exim crashed when
5492 bouncing a message because it had too many Received: header lines.
5494 20. If two routers both had headers_remove settings, and the first one included
5495 a superfluous trailing colon, the final name in the first list and the
5496 first name in the second list were incorrectly joined into one item (with a
5497 colon in the middle).
5503 1. Added -C and -D options to the exinext utility, mainly to make it easier
5504 to include in the automated testing, but these could be helpful when
5505 multiple configurations are in use.
5507 2. The exinext utility was not formatting the output nicely when there was
5508 an alternate port involved in the retry record key, nor when there was a
5509 message id as well (for retries that were specific to a specific message
5510 and a specific host). It was also confused by IPv6 addresses, because of
5511 the additional colons they contain. I have fixed the IPv4 problem, and
5512 patched it up to do a reasonable job for IPv6.
5514 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5515 delivery, the log line now contains "pipelined" if PIPELINING was used.
5517 4. An SMTP transport process used to panic and die if the bind() call to set
5518 an explicit outgoing interface failed. This has been changed; it is now
5519 treated in the same way as a connect() failure.
5521 5. A reference to $sender_host_name in the part of a conditional expansion
5522 that was being skipped was still causing a DNS lookup. This no longer
5525 6. The def: expansion condition was not recognizing references to header lines
5526 that used bh_ and bheader_.
5528 7. Added the _cache feature to named lists.
5530 8. The code for checking quota_filecount in the appendfile transport was
5531 allowing one more file than it should have been.
5533 9. For compatibility with Sendmail, the command line option
5541 and sets the incoming protocol and host name (for trusted callers). The
5542 host name and its colon can be omitted when only the protocol is to be set.
5543 Note the Exim already has two private options, -pd and -ps, that refer to
5544 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5545 "s", but I don't think that's a major issue.
5547 10. A number of refactoring changes to the code, none of which should affect
5550 (a) The number of logging options was getting close to filling up the
5551 32-bit word that was used as a bit map. I have split them into two classes:
5552 those that are passed in the argument to log_write(), and those that are
5553 only ever tested independently outside of that function. These are now in
5554 separate 32-bit words, so there is plenty of room for expansion again.
5555 There is no change in the user interface or the logging behaviour.
5557 (b) When building, for example, log lines, the code previously used a
5558 macro that called string_cat() twice, in order to add two strings. This is
5559 not really sufficiently general. Furthermore, there was one instance where
5560 it was actually wrong because one of the argument was used twice, and in
5561 one call a function was used. (As it happened, calling the function twice
5562 did not affect the overall behaviour.) The macro has been replaced by a
5563 function that can join an arbitrary number of extra strings onto a growing
5566 (c) The code for expansion conditions now uses a table and a binary chop
5567 instead of a serial search (which was left over from when there were very
5568 few conditions). Also, it now recognizes conditions like "pam" even when
5569 the relevant support is not compiled in: a suitably worded error message is
5570 given if an attempt is made to use such a condition.
5572 11. Added ${time_interval:xxxxx}.
5574 12. A bug was causing one of the ddress fields not to be passed back correctly
5575 from remote delivery subprocesses. The field in question was not being
5576 subsequently used, so this caused to problems in practice.
5578 13. Added new log selectors queue_time and deliver_time.
5580 14. Might have fixed a bug in maildirsizefile handling that threw up
5581 "unexpected character" debug warnings, and recalculated the data
5582 unnecessarily. In any case, I expanded the warning message to give more
5585 15. Added the message "Restricted characters in address" to the statements in
5586 the default ACL that block characters like @ and % in local parts.
5588 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5589 Three changes have been made:
5591 (a) There was a serious bug; a negative response to MAIL caused the whole
5592 recipient domain to be cached as invalid, thereby blocking all messages
5593 to all local parts at the same domain, from all senders. This bug has
5594 been fixed. The domain is no longer cached after a negative response to
5595 MAIL if the sender used is not empty.
5597 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5600 (c) A new callout option, "use_sender" has been added for people who want
5601 the modified behaviour.
5607 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5610 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5611 indeed breaks things for older releases.
5613 3. Added additional logging to the case where there is a problem reading data
5614 from a filter that is running in a subprocess using a pipe, in order to
5615 try to track down a specific problem.
5617 4. Testing facility fudge: when running in the test harness and attempting
5618 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5619 getting "No route to host". Convert this to a timeout.
5621 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5624 6. Some OS don't have socklen_t but use size_t instead. This affects the
5625 fifth argument of getsockopt() amongst other things. This is now
5626 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5627 can be set for individual OS. I have set it for SunOS5, OSF1, and
5628 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5629 some earlier ones do not.
5631 7. Change 4.30/15 was not doing the test caselessly.
5633 8. The standard form for an IPv6 address literal was being rejected by address
5634 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5635 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5636 this, as well as the form without the "IPv6" on the front (but only when
5637 address literals are enabled, of course).
5639 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5641 10. Exim crashed if a message with an empty sender address specified by -f
5642 encountered a router with an errors_to setting. This could be provoked only
5643 by a command such as
5647 where an empty string was supplied; "<>" did not hit this bug.
5649 11. Installed PCRE release 4.5.
5651 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5652 remained set. It is now erased.
5654 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5655 times from message ids (which are base 36 rather than the normal 62).
5657 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5658 were being counted as actual protocol errors, and logged if the log
5659 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5660 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5661 and DATA following a set of rejected RCPTs do not count as protocol errors.
5662 In other words, Exim assumes they were pipelined, though this may not
5663 actually be the case. Of course, in all cases the client gets an
5664 appropriate error code.
5666 15. If a lookup fails in an ACL condition, a message about the failure may
5667 be available; it is used if testing the ACL cannot continue, because most
5668 such messages specify what the cause of the deferral is. However, some
5669 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5670 that caused an old message to be retained and used if a later statement
5671 caused a defer, replacing the real cause of the deferral.
5673 16. If an IP address had so many PTR records that the DNS lookup buffer
5674 was not large enough to hold them, Exim could crash while trying to process
5675 the truncated data. It now detects and logs this case.
5677 17. Further to 4.21/58, another change has been made: if (and only if) the
5678 first line of a message (the first header line) ends with CRLF, a bare LF
5679 in a subsequent header line has a space inserted after it, so as not to
5680 terminate the header.
5682 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5683 unnecessarily, used atoi() instead of strtol(), and didn't check the
5684 termination when getting file sizes from file names by regex.
5686 19. Completely re-implemented the support for maildirsize files, in the light
5687 of a number of problems with the previous contributed implementation
5688 (4.30/29). In particular:
5690 . If the quota is zero, the maildirsize file is maintained, but no quota is
5693 . If the maildir directory does not exist, it is created before any attempt
5694 to write a maildirsize file.
5696 . The quota value in the file is just a cache; if the quota is changed in
5697 the transport, the new value overrides.
5699 . A regular expression is available for excluding directories from the
5702 20. The autoreply transport checks the characters in options that define the
5703 message's headers; it allows continued headers, but it was checking with
5704 isspace() after an embedded newline instead of explicitly looking for a
5707 21. If all the "regular" hosts to which an address was routed had passed their
5708 expiry times, and had not reached their retry times, the address was
5709 bounced, even if fallback hosts were defined. Now Exim should go on to try
5712 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5713 equivalent code in the SMTP transport. Some hosts send humungous responses
5714 to HELO/EHLO, more than 1024 it seems.
5716 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5717 gives compiler warnings in some environments. I've now done it "properly",
5720 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5721 (because of problems with the built-in one) was declared to return uschar *
5722 instead of char *, causing compiler failure.
5724 25. Fixed a file descriptor leak when processing alias/forward files.
5726 26. Fixed a minor format string issue in dbfn.c.
5728 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5730 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5731 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5734 29. When a sender address is verified, it is cached, to save repeating the test
5735 when there is more than one recipient in a message. However, when the
5736 verification involves a callout, it is possible for different callout
5737 options to be set for different recipients. It is too complicated to keep
5738 track of this in the cache, so now Exim always runs a verification when a
5739 callout is required, relying on the callout cache for the optimization.
5740 The overhead is duplication of the address routing, but this should not be
5743 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5744 to be verified with callout=postmaster, and the main callout worked but the
5745 postmaster check failed, the verification correctly failed. However, if a
5746 subsequent RCPT command asked for sender verification *without* the
5747 postmaster check, incorrect caching caused this verification also to fail,
5750 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5751 it was not caching the DNS options (qualify_single, search_parents) that
5752 were used when the lookup failed. A subsequent lookup with different
5753 options therefore always gave the same answer, though there were cases
5754 where it should not have. (Example: a "domains = !$mx_any" option on a
5755 dnslookup router: the "domains" option is always processed without any
5756 widening, but the router might have qualify_single set.) Now Exim uses the
5757 cached value only when the same options are set.
5759 32. Added John Jetmore's "exipick" utility to the distribution.
5761 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5762 than a timeout (e.g. a certificate is required, and is not provided), an
5763 Exim server now closes the connection immediately. Previously it waited for
5764 the client to close - but if the client is SSL, it seems that they each
5765 wait for each other, leading to a delay before one of them times out.
5767 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5768 maintained 0.8.x compatibility because I don't think many are using it, and
5769 it is clearly obsolete.
5771 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5774 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5775 client certificate was expired. A simple patch fixes this, though I don't
5776 understand the full logic of why the verify callback is called multiple
5779 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5780 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5781 which causes problems with some clients (such as the Certicom SSL Plus
5782 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5783 disables the coutermeasure allowing Eudora to connect."
5785 38. Exim was not checking that a write() to a log file succeeded. This could
5786 lead to Bad Things if a log got too big, in particular if it hit a file
5787 size limit. Exim now panics and dies if it cannot write to a log file, just
5788 as it does if it cannot open a log file.
5790 39. Modified OS/Makefile-Linux so that it now contains
5792 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5794 The two -D definitions ensure that Exim is compiled with large file
5795 support, which makes it possible to handle log files that are bigger than
5798 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5799 instance) a domain was checked against a named list that involved a lookup,
5800 causing $domain_data to be set, then another domain was checked against the
5801 same list, then the first domain was re-checked, the value of $domain_data
5802 after the final check could be wrong. In particular, if the second check
5803 failed, it could be set empty. This bug probably also applied to
5806 41. The strip_trailing_dot option was not being applied to the address given
5807 with the -f command-line option.
5809 42. The code for reading a message's header from the spool was incrementing
5810 $received_count, but never initializing it. This meant that the value was
5811 incorrect (doubled) while delivering a message in the same process in which
5812 it was received. In the most common configuration of Exim, this never
5813 happens - a fresh exec is done - but it can happen when
5814 deliver_drop_privilege is set.
5816 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5817 it now includes up to 150 characters of the unexpected data in the log
5820 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5821 and building data strings. The size of both of these buffers was 10 000
5822 bytes - far larger than anybody would *ever* want, thought I. Needless to
5823 say, somebody hit the limit. I have increased the maximum line length to
5824 20 000 and the maximum data length of concatenated lines to 100 000. I have
5825 also fixed two bugs, because there was no checking on these buffers. Tsk,
5826 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5827 buffer is too small.
5829 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5830 lsearch lookups. Now it does.
5832 46. When parsing a route_list item in a manualroute router, a fixed-length
5833 buffer was used for the list of hosts. I made this 1024 bytes long,
5834 thinking that nobody would ever have a list of hosts that long. Wrong.
5835 Somebody had a whole pile of complicated expansion conditions, and the
5836 string was silently truncated, leading to an expansion error. It turns out
5837 that it is easier to change to an unlimited length (owing to other changes
5838 that have happened since this code was originally written) than to build
5839 structure for giving a limitation error. The length of the item that
5840 expands into the list of hosts is now unlimited.
5842 47. The lsearch lookup could not handle data where the length of text line was
5843 more than 4095 characters. Such lines were truncated, leading to shortened
5844 data being returned. It should now handle lines of any length.
5846 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5847 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5850 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5851 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5852 no longer happens in comment lines. A list of replacements is now placed
5853 at the head of all of the source files, except those whose only change is
5854 to replace PERL_COMMAND in the very first #! line.
5856 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5857 messages on the queue, with a bottom-up merge sort, using code contributed
5858 by Michael Haardt. This should make operations like -bp somewhat faster on
5859 large queues. It won't affect queue runners, except when queue_run_in_order
5862 51. Installed eximstats 1.31 in the distribution.
5864 52. Added support for SRV lookups to the dnslookup router.
5866 53. If an ACL referred to $message_body or $message_body_end, the value was not
5867 reset for any messages that followed in the same SMTP session.
5869 54. The store-handling optimization for building very long strings was not
5870 differentiating between the different store pools. I don't think this
5871 actually made any difference in practice, but I've tidied it.
5873 55. While running the routers to verify a sender address, $sender_address
5874 was still set to the sender address. This is wrong, because when routing to
5875 send a bounce to the sender, it would be empty. Therefore, I have changed
5876 it so that, while verifying a sender address, $sender_address is set to <>.
5877 (There is no change to what happens when verifying a recipient address.)
5879 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5880 target A or AAAA records (if not already returned) without resetting the
5881 qualify_single or search_parents options of the DNS resolver. These are
5882 inappropriate in this case because the targets of MX and SRV records must
5883 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5884 target that, when qualified, matched something in the local domain. These
5885 two options are now turned off when doing these lookups.
5887 57. It seems that at least some releases of Reiserfs (which does not have the
5888 concept of a fixed number of inodes) returns zero and not -1 for the
5889 number of available inodes. This interacted badly with check_spool_inodes,
5890 which assumed that -1 was the "no such thing" setting. What I have done is
5891 to check that the total number of inodes is greater than zero before doing
5892 the test of how many are available.
5894 58. When a "warn" ACL statement has a log_message modifier, the message is
5895 remembered, and not repeated. This is to avoid a lot of repetition when a
5896 message has many recipients that cause the same warning to be written.
5897 However, Exim was preserving the list of already written lines for an
5898 entire SMTP session, which doesn't seem right. The memory is now reset if a
5899 new message is started.
5901 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5902 debugging output unless log_rewrite was also set.
5904 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5905 of (int)(handle) when we know that handle contains (void *)(-1).
5907 61. The Exim daemon panic-logs an error return when it closes the incoming
5908 connection. However "connection reset by peer" seems to be common, and
5909 isn't really an error worthy of noting specially, so that particular error
5912 62. When Exim is trying to find all the local interfaces, it used to panic and
5913 die if the ioctl to get the interface flags failed. However, it seems that
5914 on at least one OS (Solaris 9) it is possible to have an interface that is
5915 included in the list of interfaces, but for which you get a failure error
5916 for this call. This happens when the interface is not "plumbed" into a
5917 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5918 failure of the "get flags" call assumes that the interface is down.
5920 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5921 makes life easier for people who are doing arithmetic on fields extracted
5922 from dates, where you often get leading zeros that should not be
5923 interpreted as octal.
5925 64. Added qualify_domain to the redirect router, to override the global
5928 65. If a pathologically long header line contained very many addresses (the
5929 report of this problem mentioned 10 000) and each of them was rewritten,
5930 Exim could use up a very large amount of memory. (It kept on making new
5931 copies of the header line as it rewrote, and never released the old ones.)
5932 At the expense of a bit more processing, the header rewriting function has
5933 been changed so that it no longer eats memory in this way.
5935 66. The generation of the Received: header has been moved from the time that a
5936 message starts to be received, to the time that it finishes. The timestamp
5937 in the Received: header should now be very close to that of the <= log
5938 line. There are two side-effects of this change:
5940 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5941 logged header lines no longer include the local Received: line, because
5942 it has not yet been created. The same applies to a copy of the message
5943 that is returned to a non-SMTP sender when a message is rejected.
5945 (b) When a filter file is tested using -bf, no additional Received: header
5946 is added to the test message. After some thought, I decided that this
5949 This change does not affect the value of $received_for. It is still set
5950 after address rewriting, but before local_scan() is called.
5952 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5954 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5955 gave an unhelpful panic error message, and a defer error. I have managed to
5956 change this behaviour so that it now rejects any supplied certificate,
5957 which seems right, as the list of acceptable certificates is empty.
5959 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5960 gave an unhelpful defer error. I have not managed to make this reject any
5961 supplied certificates, but the error message it gives is "no certificate
5962 supplied", which is not helpful.
5964 70. exigrep's output now also includes lines that are not associated with any
5965 message, but which match the given pattern. Implemented by a patch from
5966 Martin Sluka, which also tidied up the Perl a bit.
5968 71. Recipient callout verification, like sender verification, was using <> in
5969 the MAIL FROM command. This isn't really the right thing, since the actual
5970 sender may affect whether the remote host accepts the recipient or not. I
5971 have changed it to use the actual sender in the callout; this means that
5972 the cache record is now keyed on a recipient/sender pair, not just the
5973 recipient address. There doesn't seem to be a real danger of callout loops,
5974 since a callout by the remote host to check the sender would use <>.
5975 [SEE ABOVE: changed after hitting problems.]
5977 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5978 temporary errors. However, in the case of such a code being given after
5979 the end of a data transmission (i.e. after ".") Exim was failing to write
5980 a retry record for the message. (Yes, there was some broken host that was
5981 actually sending 8xx at this point.)
5983 73. An unknown lookup type in a host list could cause Exim to panic-die when
5984 the list was checked. (An example that provoked this was putting <; in the
5985 middle of a list instead of at the start.) If this happened during a DATA
5986 ACL check, a -D file could be left lying around. This kind of configuration
5987 error no longer causes Exim to die; instead it causes a defer error. The
5988 incident is still logged to the main and panic logs.
5990 74. Buglet left over from Exim 3 conversion. The message "too many messages
5991 in one connection" was written to the rejectlog but not the mainlog, except
5992 when address rewriting (yes!) was being logged.
5994 75. Added write_rejectlog option.
5996 76. When a system filter was run not as root (that is, when system_filter_user
5997 was set), the values of the $n variables were not being returned to the
5998 main process; thus, they were not subsequently available in the $sn
6001 77. Added +return_path_on_delivery log selector.
6003 78. A connection timeout was being treated differently from recipients deferred
6004 when testing hosts_max_try with a message that was older than the host's
6005 retry timeout. (The host should not be counted, thus allowing all hosts to
6006 be tried at least once before bouncing.) This may have been the cause of an
6007 occasionally reported bug whereby a message would remain on the queue
6008 longer than the retry timeout, but would be bounced if a delivery was
6009 forced. I say "may" because I never totally pinned down the problem;
6010 setting up timeout/retry tests is difficult. See also the next item.
6012 79. The ultimate address timeout was not being applied to errors that involved
6013 a combination of host plus message (for example, a timeout on a MAIL
6014 command). When an address resolved to a number of possible hosts, and they
6015 were not all tried for each delivery (e.g. because of hosts_max_try), a
6016 message could remain on the queue longer than the retry timeout.
6018 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6021 81. Fixed an obscure SMTP outgoing bug which required at least the following
6022 conditions: (a) there was another message waiting for the same server;
6023 (b) the server returned 5xx to all RCPT commands in the first message so
6024 that the message was not completed; (c) the server dropped the connection
6025 or gave a negative response to the RSET that Exim sends to abort the
6026 transaction. The observed case was a dropped connection after DATA that had
6027 been sent in pipelining mode. That is, the server had advertised PIPELINING
6028 but was not implementing it correctly. The effect of the bug was incorrect
6029 behaviour, such as trying another host, and this could lead to a crash.
6035 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6036 and daemon.c were passed as pointers to ints; they should have been
6037 pointers to socklen_t variables (which are typically unsigned ints).
6039 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6042 3. Fixed a really odd bug that affected only the testing scheme; patching a
6043 certain fixed string in the binary changed the value of another string that
6044 happened to be identical to the end of the original first string.
6046 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6047 name", it returns that address as the IP address. On some operating
6048 systems (e.g. Solaris), it also passes back the IP address string as the
6049 "host name". However, on others (e.g. Linux), it passes back an empty
6050 string. Exim wasn't checking for this, and was changing the host name to an
6051 empty string, assuming it had been canonicalized.
6053 5. Although rare, it is permitted to have more than one PTR record for a given
6054 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6055 all the names associated with an address, because they do in Solaris.
6056 However, it seems that they do not in Linux for data that comes from the
6057 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6058 I found this out when I moved to a new Linux workstation and tried to run
6059 the Exim test suite.
6061 To get round this problem I have changed the code so that it now does its
6062 own call to the DNS to look up PTR records when searching for a host name.
6063 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6064 addresses that are only in /etc/hosts are still found.
6066 This behaviour is, however, controlled by an option called host_lookup_
6067 order, which defaults to "bydns:byaddr". If people want to use the other
6068 order, or indeed, just use one or the other means of lookup, they can
6069 specify it in this variable.
6071 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6072 some operating systems, this comes back from gethostbyaddr() as an empty
6073 string, and this is what Exim used to test for. However, it seems that in
6074 other systems, "." is yielded. Exim now tests for this case too.
6076 7. The values of check_spool_space and check_log_space are now held internally
6077 as a number of kilobytes instead of an absolute number of bytes. If a
6078 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6079 kilobyte. This means that much larger values can be stored.
6081 8. Exim monitor: an attempt to get the action menu when not actually pointing
6082 at a message produces an empty menu entitled "No message selected". This
6083 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6084 no entries in it ("Shell widget menu has zero width and/or height"). So I
6085 have added a single, blank menu entry in this case.
6087 9. Added ${quote_local_part.
6089 10. MIME decoding is now applied to the contents of Subject: header lines when
6092 11. Now that a reference to $sender_host_address automatically causes a reverse
6093 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6094 host lookup before query-style lookups in lists that might use this
6095 variable. This has therefore been abolished, and the "net-" prefix is no
6096 longer necessary for query-style lookups.
6098 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6099 have been a typo for LFLAGS, so it has been changed.
6101 13. The install script calls Exim with "-C /dev/null" in order to find the
6102 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6103 to be output. However, since Exim outputs its version number before the
6104 error, it didn't break the script. It just looked ugly. I fixed this by
6105 always allowing "-C /dev/null" if the caller is root.
6107 14. Ignore overlarge ACL variable number when reading spool file - insurance
6108 against a later release with more variables having written the file.
6110 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6111 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6112 this, as well as the form without the "IPv6" on the front.
6114 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6115 OS/Makefile-Darwin file.
6117 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6118 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6120 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6121 variable. However, some OS use ints for inodes. I've added cast to long int
6122 to get rid of the compiler warning.
6124 19. I had forgotten to lock out "/../" in configuration file names when
6125 ALT_CONFIG_PREFIX was set.
6127 20. Routers used for verification do not need to specify transports. However,
6128 if such a router generated a host list, and callout was configured, Exim
6129 crashed, because it could not find a port number from the (non-existent)
6130 transport. It now assumes port 25 in this circumstance.
6132 21. Added the -t option to exigrep.
6134 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6135 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6136 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6137 them is set, LOOKUP_LSEARCH is forced.
6139 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6141 24. Added sender and host information to the "rejected by local_scan()" log
6142 line; previously there was no indication of these.
6144 25. Added .include_if_exists.
6146 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6147 Linux. It turns out that not all file systems support this. Apparently some
6148 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6149 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6150 not supported on the file descriptor, is now ignored when Exim is trying to
6151 sync a directory. This applies only to Linux.
6153 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6155 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6156 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6159 29. Added support for maildirsize files from supplied patch (modified a bit).
6161 30. The use of :fail: followed by an empty string could lead Exim to respond to
6162 sender verification failures with (e.g.):
6164 550 Verification failed for <xxx>
6165 550 Sender verify failed
6167 where the first response line was missing the '-' that indicates it is not
6168 the final line of the response.
6170 31. The loop for finding the name of the user that called Exim had a hardwired
6171 limit of 10; it now uses the value of finduser_retries, which is used for
6172 all other user lookups.
6174 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6177 33. Exim was neglecting to zero errno before one call of strtol() when
6178 expanding a string and expecting an integer value. On some systems this
6179 resulted in spurious "integer overflow" errors. Also, it was casting the
6180 result into an int without checking.
6182 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6183 did not work. The code looks as if it has *never* worked, though it appears
6184 to have been documented since at least release 1.62. I have made it work.
6186 35. The "timeout_DNS" error in retry rules, also documented since at least
6187 1.62, also never worked. As it isn't clear exactly what this means, and
6188 clearly it isn't a major issue, I have abolished the feature by treating it
6189 as "timeout", and writing a warning to the main and panic logs.
6191 36. The display of retry rules for -brt wasn't always showing the error code
6194 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6195 timeout_connect_A, timeout_connect_MX.
6197 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6198 to the empty sender.
6200 39. The daemon was not analysing the content of -oX till after it had closed
6201 stderr and disconnected from the controlling terminal. This meant that any
6202 syntax errors were only noted on the panic log, and the return code from
6203 the command was 0. By re-arranging the code a little, I've made the
6204 decoding happen first, so such errors now appear on stderr, and the return
6205 code is 1. However, the actual setting up of the sockets still happens in
6206 the disconnected process, so errors there are still only recorded on the
6209 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6210 connections (as happens on some IP stacks) was logged at start up time as
6211 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6212 it from "IPv6 and IPv4", which means that two separate sockets are being
6215 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6216 says whether AF_INET or AF_INET6 was passed as an argument.
6218 42. Exiwhat output was messed up when time zones were included in log
6221 43. Exiwhat now gives more information about the daemon's listening ports,
6222 and whether -tls-on-connect was used.
6224 44. The "port" option of the smtp transport is now expanded.
6226 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6227 silently ignored. Now an error message is written to the main and panic
6230 46. There's a new ACL modifier called "logwrite" which writes to a log file
6231 as soon as it is encountered.
6233 47. Added $local_user_uid and $local_user_gid at routing time.
6235 48. Exim crashed when trying to verify a sender address that was being
6238 49. Exim was recognizing only a space character after ".include". It now also
6239 recognizes a tab character.
6241 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6242 extracting the relevant information from the specification. The man page no
6243 longer contains scrambled data for the -d option, and I've added a section
6244 at the front about calling Exim under different names.
6246 51. Added "extra_headers" argument to the "mail" command in filter files.
6248 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6251 53. Installed eximstats 1.29.
6253 54. Added transport_filter_timeout as a generic transport option.
6255 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6256 Cc: header lines. This was required by RFC 822, but it not required by RFC
6259 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6260 incoming messages that did not have them. Now it does so only if the
6261 message originates locally, that is, if there is no associated remote host
6262 address. When Resent- header lines are present, this applies to the Resent-
6263 lines rather than the non-Resent- lines.
6265 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6266 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6268 58. Messages for configuration errors now include the name of the main
6269 configuration file - useful now that there may be more than one file in a
6270 list (.included file names were always shown).
6272 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6273 for those rare installations that do not start the daemon as root or run it
6274 setuid root. I've cut out the call to initgroups() if the daemon is not
6277 60. The Exim user and group can now be bound into the binary as text strings
6278 that are looked up at the start of Exim's processing.
6280 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6282 62. Added $mailstore_basename variable.
6284 63. Installed patch to sieve.c from Michael Haardt.
6286 64. When Exim failed to open the panic log after failing to open the main log,
6287 the original message it was trying to log was written to stderr and debug
6288 output, but if they were not available (the usual case in production), it
6289 was lost. Now it is written to syslog before the two lines that record the
6290 failures to open the logs.
6292 65. Users' Exim filters run in subprocesses under the user's uid. It is
6293 possible for a "deliver" command or an alias in a "personal" command to
6294 provoke an address rewrite. If logging of address rewriting is configured,
6295 this fails because the process is not running as root or exim. There may be
6296 a better way of dealing with this, but for the moment (because 4.30 needs
6297 to be released), I have disabled address rewrite logging when running a
6298 filter in a non-root, non-exim process.
6304 1. The buildconfig auxiliary program wasn't quoting the value set for
6305 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6306 not defined. This bug was masked in 4.22 by the effect that was fixed in
6309 2. Some messages that were rejected after a message id was allocated were
6310 shown as "incomplete" by exigrep. It no longer does this for messages that
6311 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6313 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6314 have allow_domain_literals set, the ID did not get logged in the <= line.
6315 Domain literals are now always recognized in Message-ID: header lines.
6317 4. The first argument for a ${extract expansion item is the key name or field
6318 number. Leading and trailing spaces in this item were not being ignored,
6319 causing some misleading effects.
6321 5. When deliver_drop_privilege was set, single queue runner processes started
6322 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6323 same command in the process it spins off) were not dropping privilege.
6325 6. When the daemon running as "exim" started a queue runner, it always
6326 re-executed Exim in the spun-off process. This is a waste of effort when
6327 deliver_drop_privilege is set. The new process now just calls the
6328 queue-runner function directly.
6334 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6337 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6338 ignored. Though the use of -oP was forcing the writing of a pid file, it
6339 was always written to the default place.
6341 3. If the message "no IP address found for host xxxx" is generated during
6342 incoming verification, it is now followed by identification of the incoming
6343 connection (so you can more easily find what provoked it).
6345 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6347 5. Added some features to "harden" Exim a bit more against certain attacks:
6349 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6350 be put in Local/Makefile. This is like the never_users runtime option,
6351 but it cannot be overridden. The default setting is "root".
6353 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6354 prefix string with which any file named in a -C command line option
6357 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6358 is retained for -C and -D only if the caller of Exim is root. Without
6359 it, the exim user may also use -C and -D and retain privilege.
6361 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6362 command line option is disabled.
6364 6. Macro names set by the -D option must start with an upper case letter, just
6365 like macro names defined in the configuration file.
6367 7. Added "dereference=" facility to LDAP.
6369 8. Two instances of the typo "uknown" in the source files are fixed.
6371 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6372 the Configure-Makefile script screwed up while processing it.
6374 10. Incorporated PCRE 4.4.
6376 11. The SMTP synchronization check was not operating right at the start of an
6377 SMTP session. For example, it could not catch a HELO sent before the client
6378 waited for the greeting. There is now a check for outstanding input at the
6379 point when the greeting is written. Because of the duplex, asynchronous
6380 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6381 way, but not yet received, when the check is performed.
6383 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6384 on TCP/IP sockets, because this apparently causes some broken clients to
6387 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6388 unchanged) from the Cygwin maintainer.
6390 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6391 was supported instead of testing for mbx. Effectively a typo.
6393 15. The spa authenticator server code was not checking that the input it
6394 received was valid base64.
6396 16. The debug output line for the "set" modifier in ACLs was not showing the
6397 name of the variable that was being set.
6399 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6401 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6402 lookup on its own. Something else was needed to trigger the lookup. For
6403 example, a match in host_lookup or the need for a host name in a host list.
6404 Now, if $sender_host_name is referenced and the host name has not yet been
6405 looked up, a lookup is performed. If the lookup fails, the variable remains
6406 empty, and $host_lookup_failed is set to "1".
6408 19. Added "eqi" as a case-independent comparison operator.
6410 20. The saslauthd authentication condition could segfault if neither service
6411 nor realm was specified.
6413 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6414 error message that was logged was misleading, and incoming SMTP
6415 connections were dropped. The message is now more accurate, and temporary
6416 errors are given to SMTP connections.
6418 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6419 (except RSET) with a 421 error, until QUIT is received. However, it was
6420 failing to send a response to QUIT.
6422 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6423 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6424 ACL is now run after the helo_try_verify_hosts code.
6426 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6427 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6428 case-independent, so other case variants are also recognized). Apparently
6429 some systems use these upper case variants.
6431 25. If more than two messages were waiting for the same host, and a transport
6432 filter was specified for the transport, Exim sent two messages over the
6433 same TCP/IP connection, and then failed with "socket operation on non-
6434 socket" when it tried to send the third.
6436 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6438 27. The extern definition of crypt16() in expand.c was not being excluded when
6439 the OS had its own crypt16() function.
6441 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6442 as a preferred synonym for return_size_limit, both as an option and as an
6445 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6447 30. Changed the default configuration ACL to relax the local part checking rule
6448 for addresses that are not in any local domains. For these addresses,
6449 slashes and pipe symbols are allowed within local parts, but the sequence
6450 /../ is explicitly forbidden.
6452 31. SPA server authentication was not clearing the challenge buffer before
6455 32. log_message in a "warn" ACL statement was writing to the reject log as
6456 well as to the main log, which contradicts the documentation and doesn't
6457 seem right (because no rejection is happening). So I have stopped it.
6459 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6460 However, I am unable to do any testing of this.
6462 34. Fixed an infelicity in the appendfile transport. When checking directories
6463 for a mailbox, to see if any needed to be created, it was accidentally
6464 using path names with one or more superfluous leading slashes; tracing
6465 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6467 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6468 contents are added to the log line that is written for every discarded
6469 recipient. (Previously a log_message setting was ignored.)
6471 36. The ${quote: operator now quotes the string if it is empty.
6473 37. The install script runs exim in order to find its version number. If for
6474 some reason other than non-existence or emptiness, which it checks, it
6475 could not run './exim', it was installing it with an empty version number,
6476 i.e. as "exim-". This error state is now caught, and the installation is
6479 38. An argument was missing from the function that creates an error message
6480 when Exim fails to connect to the socket for saslauthd authentication.
6481 This could cause Exim to crash, or give a corrupted message.
6483 39. Added isip, isip4, and isip6 to ${if conditions.
6485 40. The ACL variables $acl_xx are now saved with the message, and can be
6486 accessed later in routers, transports, and filters.
6488 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6489 strings in the file are not string-expanded.
6491 42. If a MAIL command specified a SIZE value that was too large to fit into an
6492 int variable, the check against message_size_limit failed. Such values are
6493 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6494 day this will have to be increased, but I don't think I want to be around
6495 when emails are that large.
6502 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6503 iconv() is not standard in FreeBSD.
6505 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6506 IPv6 enabled. The observed symptom was a segmentation fault on return from
6507 the function os_common_find_running_interfaces() in src/os.c.
6509 3. In the check_special_case() function in daemon.c I had used "errno" as an
6510 argument name, which causes warnings on some systems. This was basically a
6511 typo, since it was named "eno" in the comments!
6513 4. The code that waits for the clock to tick (at a resolution of some fraction
6514 of a second) so as to ensure message-id uniqueness was always waiting for
6515 at least one whole tick, when it could have waited for less. [This is
6516 almost certainly not relevant at current processor speeds, where it is
6517 unlikely to ever wait at all. But we try to future-proof.]
6519 5. The function that sleeps for a time interval that includes fractions of a
6520 second contained a race. It did not block SIGALRM between setting the
6521 timer, and suspending (a couple of lines later). If the interval was short
6522 and the sigsuspend() was delayed until after it had expired, the suspension
6523 never ended. On busy systems this could lead to processes getting stuck for
6526 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6527 process, before it forks any delivery processes. The open lookup caching
6528 mechanism meant that the open file or database connection was passed into
6529 the delivery process. The problem was that delivery processes always tidy
6530 up cached lookup data. This could cause a problem for the next delivery
6531 process started by the queue runner, because the external queue runner
6532 process does not know about the closure. So the next delivery process
6533 still has data in the lookup cache. In the case of a file lookup, there was
6534 no problem because closing a file descriptor in a subprocess doesn't affect
6535 the parent. However, if the lookup was caching a connection to a database,
6536 the connection was closed, and the second delivery process was likely to
6537 see errors such as "PGSQL: query failed: server closed the connection
6538 unexpectedly". The problem has been fixed by closing all cached lookups
6539 in a queue runner before running a delivery process.
6541 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6542 seem to have the "const" qualifier which it has on other OS. I've
6545 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6546 *of the same type* (client or server) with the same public name that an
6547 error should be diagnosed.
6549 9. When Exim looked up a host name for an IP address, but failed to find the
6550 original IP address when looking up the host name (a safety check), it
6551 output the message "<ip address> does not match any IP for NULL", which was
6552 confusing, to say the least. The bug was that the host name should have
6553 appeared instead of "NULL".
6555 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6556 user that is built into the binary, and the -C or -D options is used, root
6557 privilege is dropped before the configuration file is read. In addition,
6558 logging is switched to stderr instead of the normal log files. If the
6559 configuration then re-defines the Exim user, the unprivileged environment
6560 is probably not what is expected, so Exim logs a panic warning message (but
6563 However, if deliver_drop_privilege is set, the unprivileged state may well
6564 be exactly what is intended, so the warning has been cut out in that case,
6565 and Exim is allowed to try to write to its normal log files.
6571 1. smtp_return_error_details was not giving details for temporary sender
6572 or receiver verification errors.
6574 2. Diagnose a configuration error if two authenticators have the same public
6577 3. Exim used not to create the message log file for a message until the first
6578 delivery attempt. This could be confusing when incoming messages were held
6579 for policy or load reasons. The message log file is now created at the time
6580 the message is received, and an initial "Received" line is written to it.
6582 4. The automatically generated man page for command line options had a minor
6583 bug that caused no ill effects; however, a more serious problem was that
6584 the procedure for building the man page automatically didn't always
6585 operate. Consequently, release 4.20 contains an out-of-date version. This
6586 shouldn't happen again.
6588 5. When building Exim with embedded Perl support, the script that builds the
6589 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6590 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6592 6. The freeze_tell option was not being used for messages that were frozen on
6593 arrival, either by an ACL or by local_scan().
6595 7. Added the smtp_incomplete_transaction log selector.
6597 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6598 was accepting AUTH without a new EHLO.
6600 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6601 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6604 10. Logging of TCP/IP connections (when configured) now happens in the main
6605 daemon process instead of the child process, so that the TCP/IP connection
6606 count is more accurate (but it can never be perfect).
6608 11. The use of "drop" in a nested ACL was not being handled correctly in the
6609 outer ACL. Now, if condition failure induced by the nested "drop" causes
6610 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6611 or "require"), the connection is dropped.
6613 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6614 that yield "discard" can now be used with an "accept" or a "discard" verb,
6615 but an error is generated for any others (because I can't see a useful way
6616 to define what should happen).
6618 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6619 are now processed in the same way as lines in the Exim configuration file.
6620 In particular, continuation lines are supported.
6622 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6624 15. Added -ti meaning -t -i.
6626 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6627 domains, and warn by logging if others are found.
6629 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6630 returned from GIFCONF when Exim is trying to find the list of interfaces on
6631 a host. The code in os.c has been modified to copy each ifreq to an aligned
6632 structure in all cases.
6634 Also, in some cases, the returned ifreq's were being copied to a 'struct
6635 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6636 means the last couple of bytes of an IPv6 address could be chopped if the
6637 ifreq contained only a normal sockaddr (14 bytes storage).
6639 18. Named domain lists were not supported in the hosts_treat_as_local option.
6640 An entry such as +xxxx was not recognized, and was treated as a literal
6643 19. Ensure that header lines added by a DATA ACL are included in the reject log
6644 if the ACL subsequently rejects the message.
6646 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6647 MD5 (which is deprecated).
6649 21. When testing a filter file using -bf, Exim was writing a message when it
6650 took the sender from a "From " line in the message, but it was not doing so
6651 when it took $return_path from a Return-Path: header line. It now does.
6653 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6654 with a valid header line field name (a series of printing characters
6655 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6657 23. Changed "disc" in the source to "disk" to conform to the documentation and
6658 the book and for uniformity.
6660 24. Ignore Sendmail's -Ooption=value command line item.
6662 25. When execve() failed while trying to run a command in a pipe transport,
6663 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6664 could be confused with a return value of 69 from the command itself. This
6665 has been changed to 127, the value the shell returns if it is asked to run
6666 a non-existent command. The wording for the related log line suggests a
6667 non-existent command as the problem.
6669 26. If received_header_text expands to an empty string, do not add a Received:
6670 header line to the message. (Well, it adds a token one on the spool, but
6671 marks it "old" so that it doesn't get used or transmitted.)
6673 27. Installed eximstats 1.28 (addition of -nt option).
6675 28. There was no check for failure on the call to getsockname() in the daemon
6676 code. This can fail if there is a shortage of resources on the system, with
6677 ENOMEM, for example. A temporary error is now given on failure.
6679 29. Contrary to the C standard, it seems that in some environments, the
6680 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6681 program. Exim now does this explicitly; it affects the formatting of
6682 timestamps using strftime().
6684 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6685 complaints. I've now initialized all the variables, to avoid this.
6687 32. Header lines added by a system filter were not being "seen" during
6688 transport-time rewrites.
6690 33. The info_callback() function passed to OpenSSL is set up with type void
6691 (*)(SSL *, int, int), as described somewhere. However, when calling the
6692 function (actually a macro) that sets it up, the type void(*)() is
6693 expected. I've put in a cast to prevent warnings from picky compilers.
6695 34. If a DNS black list lookup found a CNAME record, but there were no A
6696 records associated with the domain it pointed at, Exim crashed.
6698 35. If a DNS black list lookup returned more than one A record, Exim ignored
6699 all but the first. It now scans all returned addresses if a particular IP
6700 value is being sought. In this situation, the contents of the
6701 $dnslist_value variable are a list of all the addresses, separated by a
6704 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6705 to accept a host name and all its aliases if the forward lookup for any of
6706 them yielded the IP address of the incoming connection. Now it accepts only
6707 those names whose forward lookup yields the correct IP address. Any other
6708 names are discarded. This closes a loophole whereby a rogue DNS
6709 administrator could create reverse DNS records to break through a
6710 wildcarded host restriction in an ACL.
6712 37. If a user filter or a system filter that ran in a subprocess used any of
6713 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6714 the wrong values were passed to the pipe command ($thisaddress had the
6715 value of $0, $0 had the value of $1, etc). This bug was introduced by
6716 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6718 38. Improved the line breaking for long SMTP error messages from ACLs.
6719 Previously, if there was no break point between 40 and 75 characters, Exim
6720 left the rest of the message alone. Two changes have been made: (a) I've
6721 reduced the minimum length to 35 characters; (b) if it can't find a break
6722 point between 35 and 75 characters, it looks ahead and uses the first one
6723 that it finds. This may give the occasional overlong line, but at least the
6724 remaining text gets split now.
6726 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6727 file descriptors might be low, and that setting 1000 would always raise it.
6728 It turns out that in some environments, the limit is already over 1000 and
6729 that lowering it causes trouble. So now Exim takes care not to decrease it.
6731 40. When delivering a message, the value of $return_path is set to $sender_
6732 address at the start of routing (routers may change the value). By an
6733 oversight, this default was not being set up when an address was tested by
6734 -bt or -bv, which affected the outcome if any router or filter referred to
6737 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6738 log only when "message" or "log_message" are set. However, if one of the
6739 conditions was an address verification, or a call to a nested ACL, the
6740 messages generated by the underlying test were being passed through. This
6741 no longer happens. The underlying message is available in $acl_verify_
6742 message for both "message" and "log_message" expansions, so it can be
6743 passed through if needed.
6745 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6746 new expansion $bh_ to give the encoded byte string without charset
6747 translation. Translation happens only if iconv() is available; HAVE_ICONV
6748 indicates this at build time. HEADERS_CHARSET gives the charset to
6749 translate to; headers_charset can change it in the configuration, and
6750 "headers charset" can change it in an individual filter file.
6752 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6753 that creates RFC 2047 encoded "words" labels them as that charset instead
6754 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6755 expansion operator; (ii) when Exim creates a From: line for a local
6756 message; (iii) when a header line is rewritten to include a "phrase" part.
6758 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6759 buggy, causing it to skip the first lines of messages whose message ID
6760 ended in 'D'. This would not have bitten before Exim release 4.14, because
6761 message IDs were unlikely to end in 'D' before then. The effect was to have
6762 incorrect size information for certain domains.
6764 45. #include "config.h" was missing at the start of the crypt16.c module. This
6765 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6768 46. If there was a timeout during a "random" callout check, Exim treated it as
6769 a failure of the random address, and carried on sending RSET and the real
6770 address. If the delay was just some slowness somewhere, the response to the
6771 original RCPT would be taken as a response to RSET and so on, causing
6772 mayhem of various kinds.
6774 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6775 when I implemented it. It didn't allow for the fact that some option values
6776 may legitimately be negative (e.g. size_addition), and it didn't even do
6777 the right test for positive values.
6779 48. Domain names in DNS records are case-independent. Exim always looks them up
6780 in lower case. Some resolvers return domain names in exactly the case they
6781 appear in the zone file, that is, they may contain uppercase letters. Not
6782 all resolvers do this - some return always lower case. Exim was treating a
6783 change of case by a resolver as a change of domain, similar to a widening
6784 of a domain abbreviation. This triggered its re-routing code and so it was
6785 trying to route what was effectively the same domain again. This normally
6786 caused routing to fail (because the router wouldn't handle the domain
6787 twice). Now Exim checks for this case specially, and just changes the
6788 casing of the domain that it ultimately uses when it transmits the message
6791 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6794 50. If a filter generated a file delivery with a non-absolute name (possible if
6795 no home directory exists for the router), the forbid_file option was not
6798 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6799 the existing equality matching.
6801 52. Exim was using ints instead of ino_t variables in some places where it was
6802 dealing with inode numbers.
6804 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6805 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6806 called TMPDIR, and if it finds it is different, it changes its value.
6808 54. The smtp_printf() function is now made available to local_scan() so
6809 additional output lines can be written before returning. There is also an
6810 smtp_fflush() function to enable the detection of a dropped connection.
6811 The variables smtp_input and smtp_batched_input are exported to
6814 55. Changed the default runtime configuration: the message "Unknown user"
6815 has been removed from the ACL, and instead placed on the localuser router,
6816 using the cannot_route_message feature. This means that any verification
6817 failures that generate their own messages won't get overridden. Similarly,
6818 the "Unrouteable address" message that was in the ACL for unverifiable
6819 relay addresses has also been removed.
6821 56. Added hosts_avoid_esmtp to the smtp transport.
6823 57. The exicyclog script was not checking for the esoteric option
6824 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6825 will work only if exicyclog is run under the appropriate euid.
6827 58. Following a discussion on the list, the rules by which Exim recognises line
6828 endings on incoming messages have been changed. The -dropcr and drop_cr
6829 options are now no-ops, retained only for backwards compatibility. The
6830 following line terminators are recognized: LF CRLF CR. However, special
6831 processing applies to CR:
6833 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6834 nor a local message in the state where . is a terminator.
6836 (ii) If a bare CR is encountered in a header line, an extra space is added
6837 after the line terminator so as not to end the header. The reasoning
6838 behind this is that bare CRs in header lines are most likely either
6839 to be mistakes, or people trying to play silly games.
6841 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6842 was being incorrectly given as 18 bytes larger than it should have been.
6843 This is a VOB (very old bug).
6845 60. This may never have affected anything current, but just in case it has:
6846 When the local host is found other than at the start of a list of hosts,
6847 the local host, those with the same MX, and any that follow, are discarded.
6848 When the list in question was part of a longer list of hosts, the following
6849 hosts (not currently being processed) were also being discarded. This no
6850 longer happens. I'm not sure if this situation could ever has previously
6853 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6856 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6857 header. This is something that is recommended in a new Internet Draft, and
6858 is something that is documented as being done by Sendmail. There are two
6859 possible values. For messages generated by the autoreply transport, Exim
6862 Auto-Submitted: auto-replied
6864 whereas for all other generated messages (e.g. bounces) it adds
6866 Auto-Submitted: auto-generated
6868 63. The "personal" condition in filters now includes a test for the
6869 Auto-Submitted: header. If it contains the string "auto-" the message it
6870 not considered personal.
6872 64. Added rcpt_include_affixes as a generic transport option.
6874 65. Added queue_only_override (default true).
6876 66. Added the syslog_duplication option.
6878 67. If what should have been the first header line of a message consisted of
6879 a space followed by a colon, Exim was mis-interpreting it as a header line.
6880 It isn't of course - it is syntactically invalid and should therefore be
6881 treated as the start of the message body. The misbehaviour could have
6882 caused a number of strange effects, including loss of data in subsequent
6883 header lines, and spool format errors.
6885 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6886 client host had authenticated. This control can now be exercised by an ACL
6887 for more flexibility.
6889 69. By default, callouts do not happen when testing with -bh. There is now a
6890 variant, -bhc, which does actually run the callout code, including
6891 consulting and updating the callout cache.
6893 70. Added support for saslauthd authentication, courtesy of Alexander
6896 71. If statvfs() failed on the spool or log directories while checking their
6897 size for availability, Exim confusingly gave the error "space shortage".
6898 Furthermore, in debugging mode it crashed with a floating point exception.
6899 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6900 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6901 serious problem, Exim now writes to the main and panic logs when this
6902 happens, with details of the failure. It then refuses to accept the
6903 incoming message, giving the message "spool directory problem" or "log
6904 directory problem" with a 421 code for SMTP messages.
6906 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6907 0, 1, and 2 exist, because some OS complain for execs without them (see
6908 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6909 descriptors. However, the code omitted to check that the open succeeded,
6910 causing mysterious errors if for some reason the permissions on /dev/null
6911 got screwed. Now Exim writes a message to the main and panic logs, and
6912 bombs out if it can't open /dev/null.
6914 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6915 interact so that it is all more flexible. It is supposed to remain
6916 backwards compatible. Also added extra_local_interfaces.
6918 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6919 to bomb out with an assertion failure - to the client this appears as a
6920 connection drop. This problem occurs in the part of the code that was taken
6921 from the Samba project. Fortunately, the assertion is in a very simple
6922 function, so I have fixed this by reproducing the function inline in the
6923 one place where it is called, and arranging for authentication to fail
6924 instead of killing the process with assert().
6926 75. The SPA client code was not working when the server requested OEM rather
6927 than Unicode encoding.
6929 76. Added code to make require_files with a specific uid setting more usable in
6930 the case where statting the file as root fails - usually a non-root-mounted
6931 NFS file system. When this happens and the failure is EACCES, Exim now
6932 forks a subprocess and does the per-uid checking as the relevant uid.
6934 77. Added process_log_path.
6936 78. If log_file_path was not explicitly set, a setting of check_log_space or
6937 check_log_inodes was ignored.
6939 79. If a space check for the spool or log partitions fails, the incident is now
6940 logged. Of course, in the latter case the data may get lost...
6942 80. Added the %p formatting code to string_format() so that it can be used to
6943 print addresses in debug_print(). Adjusted all the address printing in the
6944 debugging in store.c to use %p rather than %d.
6946 81. There was a concern that a line of code in smtp_in.c could overflow a
6947 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6948 initially expressed, the concern was not well-founded, because trailing
6949 spaces are removed early. However, if the trailing spaces were followed by
6950 a NULL, they did not get removed, so the overflow was possible. Two fixes
6953 (a) I re-wrote the offending code in a cleaner fashion.
6954 (b) If an incoming SMTP command contains a NULL character, it is rejected
6957 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6958 runs initgroups(), so that if the Exim user is in any additional groups,
6959 they will be used during message reception.
6965 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob