1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
19 JH/01 Rework error string handling in TLS interface so that the caller in
20 more cases is responsible for logging. This permits library-sourced
21 string to be attached to addresses during delivery, and collapses
22 pairs of long lines into single ones.
24 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
25 during configuration. Wildcards are allowed and expanded.
27 JH/02 Rework error string handling in DKIM to pass more info back to callers.
28 This permits better logging.
30 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
31 do not close it down and have the child transport start it up again on
32 the passed-on TCP connection. Instead, proxy the child (and any
33 subsequent ones) for TLS via a unix-domain socket channel. Logging is
34 affected: the continued delivery log lines do not have any DNSSEC, TLS
35 Certificate or OCSP information. TLS cipher information is still logged.
37 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
38 identical IP addresses on different listening ports. Will also affect
41 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
42 add noisy ifdef guards to special-case this sillyness.
43 Patch from Bernd Kuhls.
45 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
46 than 255 are no longer allowed.
48 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
49 Disable the session-cache too, which might reduce our load. Since we
50 currrectly use a new context for every connection, both as server and
51 client, there is no benefit for these.
52 GnuTLS appears to not support tickets server-side by default (we don't
53 call gnutls_session_ticket_enable_server()) but client side is enabled
54 by default on recent versions (3.1.3 +) unless the PFS priority string
57 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
58 <https://reproducible-builds.org/specs/source-date-epoch/>.
60 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
61 the check for any unsuccessful recipients did not notice the limit, and
62 erroneously found still-pending ones.
64 JH/08 Pipeline CHUNKING command and data together, on kernels that support
65 MSG_MORE. Only in-clear (not on TLS connections).
67 JH/09 Avoid using a temporary file during transport using dkim. Unless a
68 transport-filter is involved we can buffer the headers in memory for
69 creating the signature, and read the spool data file once for the
70 signature and again for transmission.
72 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
73 4.77 as the kernel support then wasn't solid, having issues in 64bit
74 mode. Now, it's been long enough. Add support for FreeBSD also.
76 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
77 case where the routing stage had gathered several addresses to send to
78 a host before calling the transport for the first, we previously failed
79 to close down TLS in the old transport process before passing the TCP
80 connection to the new process. The new one sent a STARTTLS command
81 which naturally failed, giving a failed delivery and bloating the retry
82 database. Investigation and fix prototype from Wolfgang Breyha.
84 JH/12 Fix check on SMTP command input synchronisation. Previously there were
85 false-negatives in the check that the sender had not preempted a response
86 or prompt from Exim (running as a server), due to that code's lack of
87 awareness of the SMTP input buffering.
89 PP/04 Add commandline_checks_require_admin option.
90 Exim drops privileges sanely, various checks such as -be aren't a
91 security problem, as long as you trust local users with access to their
92 own account. When invoked by services which pass untrusted data to
93 Exim, this might be an issue. Set this option in main configuration
94 AND make fixes to the calling application, such as using `--` to stop
97 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
98 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
99 responses to those, into a single TLS record each way (this usually means
100 a single packet). As a side issue, smtp_enforce_sync now works on TLS
103 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
104 affects you only if you're dancing at the edge of the param size limits.
105 If you are, and this message makes sense to you, then: raise the
106 configured limit or use OpenSSL 1.1. Nothing we can do for older
109 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
110 cmdline element to get the documented default one. Previously it was
113 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
114 are unsuable for non-message-specific reasons - eg. connection timeout,
115 banner-time rejection.
117 JH/16 Fix logging of delivery remote port, when specified by router, under
120 PP/06 Repair manualroute's ability to take options in any order, even if one
121 is the name of a transport.
124 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
126 JH/17 Change the list-building routines interface to use the expanding-string
127 triplet model, for better allocation and copying behaviour.
129 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
130 Previously it was constructed the first time a possibly-matching string
131 was met in the configuration file input during startup; now it is done
134 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
135 compatible one, to avoid the (poorly documented) possibility of a config
136 file in the working directory redirecting the DB files, possibly correpting
137 some existing file. CVE-2017-10140 assigned for BDB.
139 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
140 cache-hot. Previously, although the result was properly cached, the
141 initial verify call returned a defer.
143 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
144 the main verify for receipient in uncached-mode.
146 JH/22 Retire historical build files to an "unsupported" subdir. These are
147 defined as "ones for which we have no current evidence of testing".
149 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
150 if present. Previously it was ignored.
152 JH/24 Start using specified-initialisers in C structure init coding. This is
153 a C99 feature (it's 2017, so now considered safe).
155 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
156 if was a fixed-sized field and bitmask ops via macros; it is now more
159 PP/07 GitHub PR 56: Apply MariaDB build fix.
160 Patch provided by Jaroslav Škarvada.
162 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
163 during Coverity cleanups [4.87 JH/47]
164 Diagnosis and fix provided by Michael Fischer v. Mollard.
166 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
167 the right size to place the terminating semicolon on its own folded
168 line, the header hash was calculated to an incorrect value thanks to
169 the (relaxed) space the fold became.
171 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
172 and confused the parent.
174 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
175 which could crash as a result. This could lead to undeliverable messages.
177 JH/28 Logging: "next input sent too soon" now shows where input was truncated
180 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
181 matters on fast-turnover and PID-randomising systems, which were getting
182 out-of-order delivery.
184 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
185 a possibly-overlapping copy. The symptom was that "Remote host closed
186 connection in response to HELO" was logged instead of the actual 4xx
189 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
190 Previously only that bufferd was discarded, resulting in SYMTP command
193 JH/32 DKIM: when a message has multiple signatures matching an identity given
194 in dkim_verify_signers, run the dkim acl once for each. Previously only
195 one run was done. Bug 2189.
197 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
198 "panic the current process" to "deliberately defer". The panic log is
199 still written with the problem list name; the mail and reject logs now
200 get a temp-reject line for the message that was being handled, saying
201 something like "domains check lookup or other defer". The SMTP 451
202 message is still "Temporary local problem".
204 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
205 A crafted sequence of BDAT commands could result in in-use memory beeing
206 freed. CVE-2017-16943.
208 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
209 from SMTP input. Previously it was always done; now only done for DATA
210 and not BDAT commands. CVE-2017-16944.
216 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
217 than -2003 did; needs libidn2 in addition to libidn.
219 JH/02 The path option on a pipe transport is now expanded before use.
221 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
222 Patch provided by "Björn", documentation fix added too.
224 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
225 missing a wire-to-host endian conversion.
227 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
228 close after a BDAT command line could be taken as a following command,
229 giving a synch failure. Fix by only checking for synch immediately
230 before acknowledging the chunk.
232 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
233 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
234 macro. Patches provided by Josh Soref.
236 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
237 Previously we did not; the RFC seems ambiguous and VRFY is not listed
238 by IANA as a service extension. However, John Klensin suggests that we
241 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
242 the dkim code may be unix-mode line endings rather than smtp wire-format
243 CRLF, so prepend a CR to any bare LF.
245 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
246 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
248 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
249 the first were themselves being wrongly included in the feed into dkim
250 processing; with most chunk sizes in use this resulted in an incorrect
251 body hash calculated value.
253 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
254 DKIM signature block, for verification. Although advised against by
255 standards it is specifically not ruled illegal.
257 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
259 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
260 missing a body hash (the bh= tag).
262 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
263 It seems that HAProxy sends the Proxy Protocol information in clear and
264 only then does a TLS startup, so do the same.
266 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
267 TCP connections (such as for Spamd) unless the daemon successfully set
268 Fast Open mode on its listening sockets. This fixes breakage seen on
269 too-old kernels or those not configured for Fast Open, at the cost of
270 requiring both directions being enabled for TFO, and TFO never being used
271 by non-daemon-related Exim processes.
273 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
274 endings, at least on the first header line. Try to canonify any that get
275 past that check, despite the cost.
277 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
278 now limited to an arbitrary five deep, while parsing addresses with the
279 strip_excess_angle_brackets option enabled.
281 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
282 instead leave the unprompted TLS handshake in socket buffer for the
283 TLS library to consume.
285 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
287 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
289 JH/16 Drop variables when they go out of scope. Memory management drops a whole
290 region in one operation, for speed, and this leaves assigned pointers
291 dangling. Add checks run only under the testsuite which checks all
292 variables at a store-reset and panics on a dangling pointer; add code
293 explicitly nulling out all the variables discovered. Fixes one known
294 bug: a transport crash, where a dangling pointer for $sending_ip_address
295 originally assigned in a verify callout, is re-used.
297 PP/06 Drop '.' from @INC in various Perl scripts.
299 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
301 PP/08 Reduce a number of compilation warnings under clang; building with
302 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
303 should be warning-free.
305 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
307 HS/01 Fix portability problems introduced by PP/08 for platforms where
308 realloc(NULL) is not equivalent to malloc() [SunOS et al].
310 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
311 chunk. This allows us to accept broken chunked messages. We need a more
312 general solution here.
314 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
315 already-broken messages in the queue.
317 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
319 JH/19 Fix reference counting bug in routing-generated-address tracking.
325 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
326 supports it and a size is available (ie. the sending peer gave us one).
328 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
329 years of being deprecated). The replacements are the ACLs
330 acl_smtp_mime and acl_not_smtp_mime.
332 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
333 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
334 or even an in-clear connection were permitted. Now, if the host lookup
335 was dnssec and dane was requested then the host is only used if the
336 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
337 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
338 if one fails this test.
339 This means that a poorly-configured remote DNS will make it incommunicado;
340 but it protects against a DNS-interception attack on it.
342 JH/04 Bug 1810: make continued-use of an open smtp transport connection
343 non-noisy when a race steals the message being considered.
345 JH/05 If main configuration option tls_certificate is unset, generate a
346 self-signed certificate for inbound TLS connections.
348 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
349 in rewrites and routers.
351 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
352 and logged a warning sing 4.83; now they are a configuration file error.
354 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
355 (lacking @domain). Apply the same qualification processing as RCPT.
357 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
359 JH/10 Support ${sha256:} applied to a string (as well as the previous
362 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
363 a cutthrough deliver is pending, as we always want to make a connection.
364 This also avoids re-routing the message when later placing the cutthrough
365 connection after a verify cache hit.
366 Do not update it with the verify result either.
368 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
369 when routing results in more than one destination address.
371 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
372 signing (which inhibits the cutthrough capability). Previously only
373 the presence of an option was tested; now an expansion evaluating as
374 empty is permissible (obviously it should depend only on data available
375 when the cutthrough connection is made).
377 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
378 the relevant preceding SMTP command did not note the pipelining mode.
380 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
381 Previously they were not counted.
383 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
384 as one having no matching records. Previously we deferred the message
385 that needed the lookup.
387 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
388 distinguished as "(=".
390 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
391 for missing MX records. Previously it only worked for missing A records.
393 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
395 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
396 after the data-go-ahead and data-ack. Patch from Jason Betts.
398 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
399 even for a "none" policy. Patch from Tony Meyer.
401 JH/22 Fix continued use of a connection for further deliveries. If a port was
402 specified by a router, it must also match for the delivery to be
405 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
406 When one of the recipients of a message was unsuitable for the connection
407 (has no matching addresses), we lost track of needing to mark it
408 deferred. As a result mail would be lost.
410 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
412 JH/25 Decoding ACL controls is now done using a binary search; the source code
413 takes up less space and should be simpler to maintain. Merge the ACL
414 condition decode tables also, with similar effect.
416 JH/26 Fix problem with one_time used on a redirect router which returned the
417 parent address unchanged. A retry would see the parent address marked as
418 delivered, so not attempt the (identical) child. As a result mail would
421 JH/27 Fix a possible security hole, wherein a process operating with the Exim
422 UID can gain a root shell. Credit to http://www.halfdog.net/ for
423 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
426 JH/28 Enable {spool,log} filesystem space and inode checks as default.
427 Main config options check_{log,spool}_{inodes,space} are now
428 100 inodes, 10MB unless set otherwise in the configuration.
430 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
431 Previously it only applied to the main-section connection policy
434 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
436 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
437 by me. Added RFC7919 DH primes as an alternative.
439 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
440 are not in the system include path.
442 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
443 GnuTLS, when a session startup failed (eg because the client disconnected)
444 Exim did stdio operations after fclose. This was exposed by a recent
445 change which nulled out the file handle after the fclose.
447 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
448 signed directly by the cert-signing cert, rather than an intermediate
449 OCSP-signing cert. This is the model used by LetsEncrypt.
451 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
453 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
454 an incoming connection.
456 HS/02 Bug 1802: Do not half-close the connection after sending a request
459 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
460 fallback to "prime256v1".
462 JH/34 SECURITY: Use proper copy of DATA command in error message.
463 Could leak key material. Remotely exploitable. CVE-2016-9963.
469 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
470 and 3.4.4 - once the server is enabled to respond to an OCSP request
471 it does even when not requested, resulting in a stapling non-aware
472 client dropping the TLS connection.
474 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
475 support variable-length bit vectors. No functional change.
477 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
478 The I= interface field on outgoing lines is now after the H= remote
479 host field, same as incoming lines. There is a separate
480 outgoing_interface log selector which allows you to disable the
483 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
484 If not running log_selector +smtp_connection the mainlog would be held
485 open indefinitely after a "too many connections" event, including to a
486 deleted file after a log rotate. Leave the per net connection logging
487 leaving it open for efficiency as that will be quickly detected by the
488 check on the next write.
490 HS/01 Bug 1671: Fix post transport crash.
491 Processing the wait-<transport> messages could crash the delivery
492 process if the message IDs didn't exist for some reason. When
493 using 'split_spool_directory=yes' the construction of the spool
494 file name failed already, exposing the same netto behaviour.
496 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
497 mime_regex ACL conditions.
499 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
500 to DSN fail messages (bounces): remote IP, remote greeting, remote response
501 to HELO, local diagnostic string.
503 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
504 log line to debug. Even when configured with a tls authenticator many
505 client connections are expected to not authenticate in this way, so
506 an authenticate fail is not an error.
508 HS/02 Add the Exim version string to the process info. This way exiwhat
509 gives some more detail about the running daemon.
511 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
512 matter for fast-change records such as DNSBLs.
514 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
515 retry record, even if constant. There may be multiple transports with
516 different interface settings and the retry behaviour needs to be kept
519 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
521 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
523 JH/10 Bug 840: fix log_defer_output option of pipe transport
525 JH/11 Bug 830: use same host for all RCPTS of a message, even under
526 hosts_randomize. This matters a lot when combined with mua_wrapper.
528 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
529 ${quote_pgsql:<string>} operator.
531 JH/13 Bug 1708: avoid misaligned access in cached lookup.
533 JH/14 Change header file name for freeradius-client. Relevant if compiling
534 with Radius support; from the Gentoo tree and checked under Fedora.
536 JH/15 Bug 1712: Introduce $prdr_requested flag variable
538 JH/16 Bug 1714: Permit an empty string as expansion result for transport
539 option transport_filter, meaning no filtering.
541 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
543 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
544 defaults to "*" (all hosts). The variable is now available when not built
545 with TLS, default unset, mainly to enable keeping the testsuite sane.
546 If a server certificate is not supplied (via tls_certificate) an error is
547 logged, and clients will find TLS connections fail on startup. Presumably
548 they will retry in-clear.
549 Packagers of Exim are strongly encouraged to create a server certificate
550 at installation time.
552 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
553 with the $config_file variable.
555 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
556 in transport context, after the attempt, and per-recipient. The latter type
557 is per host attempted. The event data is the error message, and the errno
558 information encodes the lookup type (A vs. MX) used for the (first) host,
559 and the trailing two digits of the smtp 4xx response.
561 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
562 to write to mainlog (or rejectlog, paniclog) in the window between file
563 creation and permissions/ownership being changed. Particularly affects
564 installations where exicyclog is run as root, rather than exim user;
565 result is that the running daemon panics and dies.
567 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
569 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
570 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
571 "pri" and "weight". Note that the previous implicit priority given by the
572 list order is no longer honoured.
574 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
577 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
578 by defining SUPPORT_SOCKS.
580 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
581 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
582 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
583 variables are renamed to proxy_{local,external}_{address,port}.
585 JH/27 Move Internationalisation support from Experimental to mainline, enabled
586 for a build by defining SUPPORT_I18N
588 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
589 of the query string, and make ${quote_redis:} do that quoting.
591 JH/29 Move Events support from Experimental to mainline, enabled by default
592 and removable for a build by defining DISABLE_EVENT.
594 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
596 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
597 cached by the daemon.
599 JH/32 Move Redis support from Experimental to mainline, enabled for a build
600 by defining LOOKUP_REDIS. The libhiredis library is required.
602 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
603 keys are given for lookup.
605 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
606 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
607 only supported when built with TLS support. The PolarSSL SHA routines
608 are still used when the TLS library is too old for convenient support.
610 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
611 openssl_options), for security. OpenSSL forces this from version 1.1.0
612 server-side so match that on older versions.
614 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
615 allocation for $value could be released as the expansion processing
616 concluded, but leaving the global pointer active for it.
618 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
619 and to use the domains and local_parts ACL conditions.
621 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
622 incorrectly not doubled on cutthrough transmission, hence seen as a
623 body-termination at the receiving system - resulting in truncated mails.
624 Commonly the sender saw a TCP-level error, and retransmitted the message
625 via the normal store-and-forward channel. This could result in duplicates
626 received - but deduplicating mailstores were liable to retain only the
627 initial truncated version.
629 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
631 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
633 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
634 we're in there, support oversigning also; bug 1309.
636 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
638 HS/04 Add support for keep_environment and add_environment options.
640 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
641 either intentional arithmetic overflow during PRNG, or testing config-
644 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
645 delivery resulted in actual delivery. Cancel cutthrough before DATA
648 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
649 rejects a recipient: pass the reject to the originator.
651 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
652 Many were false-positives and ignorable, but it's worth fixing the
655 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
656 for the new environment-manipulation done at startup. Move the routines
657 from being local to tls.c to being global via the os.c file.
659 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
660 an extract embedded as result-arg for a map, the first arg for extract
661 is unavailable so we cannot tell if this is a numbered or keyed
662 extraction. Accept either.
668 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
671 JH/02 The smtp transport option "multi_domain" is now expanded.
673 JH/03 The smtp transport now requests PRDR by default, if the server offers
676 JH/04 Certificate name checking on server certificates, when exim is a client,
677 is now done by default. The transport option tls_verify_cert_hostnames
678 can be used to disable this per-host. The build option
679 EXPERIMENTAL_CERTNAMES is withdrawn.
681 JH/05 The value of the tls_verify_certificates smtp transport and main options
682 default to the word "system" to access the system default CA bundle.
683 For GnuTLS, only version 3.0.20 or later.
685 JH/06 Verification of the server certificate for a TLS connection is now tried
686 (but not required) by default. The verification status is now logged by
687 default, for both outbound TLS and client-certificate supplying inbound
690 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
693 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
694 Status Notification (bounce) messages are now MIME format per RFC 3464.
695 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
696 under the control of the dsn_advertise_hosts option, and routers may
697 have a dsn_lasthop option.
699 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
700 default, modifiable by a malware= option. The list separator for
701 the options can now be changed in the usual way. Bug 68.
703 JH/10 The smtp_receive_timeout main option is now expanded before use.
705 JH/11 The incoming_interface log option now also enables logging of the
706 local interface on delivery outgoing connections.
708 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
709 if the interface and destination host and port all match.
711 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
714 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
715 Patch from Andrew Lewis.
717 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
718 now supports optional time-restrictions, weighting, and priority
719 modifiers per server. Patch originally by <rommer@active.by>.
721 JH/16 The spamd_address main option now supports a mixed list of local
722 and remote servers. Remote servers can be IPv6 addresses, and
723 specify a port-range.
725 JH/17 Bug 68: The spamd_address main option now supports an optional
726 timeout value per server.
728 JH/18 Bug 1581: Router and transport options headers_add/remove can
729 now have the list separator specified.
731 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
734 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
737 JH/21 Support for the A6 type of dns record is withdrawn.
739 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
740 rather than the verbs used.
742 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
743 from 255 to 1024 chars.
745 JH/24 Verification callouts now attempt to use TLS by default.
747 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
748 are generic router options now. The defaults didn't change.
750 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
751 Original patch from Alexander Shikoff, worked over by JH.
753 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
754 files if ZCAT_COMMAND is not executable.
756 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
758 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
760 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
761 Normally benign, it bites when the pair was led to by a CNAME;
762 modern usage is to not canonicalize the domain to a CNAME target
763 (and we were inconsistent anyway for A-only vs AAAA+A).
765 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
767 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
768 when evaluating $sender_host_dnssec.
770 JH/31 Check the HELO verification lookup for DNSSEC, adding new
771 $sender_helo_dnssec variable.
773 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
775 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
777 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
779 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
780 documented as working, but never had. Support all but $spam_report.
782 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
783 added for tls authenticator.
785 HS/03 Add perl_taintmode main config option
791 TL/01 When running the test suite, the README says that variables such as
792 no_msglog_check are global and can be placed anywhere in a specific
793 test's script, however it was observed that placement needed to be near
794 the beginning for it to behave that way. Changed the runtest perl
795 script to read through the entire script once to detect and set these
796 variables, reset to the beginning of the script, and then run through
797 the script parsing/test process like normal.
799 TL/02 The BSD's have an arc4random API. One of the functions to induce
800 adding randomness was arc4random_stir(), but it has been removed in
801 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
802 function when detected.
804 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
805 cause callback expansion.
807 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
808 syntax errors in an expansion can be treated as a string instead of
809 logging or causing an error, due to the internal use of bool_lax
810 instead of bool when processing it.
812 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
813 server certificates when making smtp deliveries.
815 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
817 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
819 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
821 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
822 Merged patch from Sebastian Wiedenroth.
824 JH/05 Fix results-pipe from transport process. Several recipients, combined
825 with certificate use, exposed issues where response data items split
826 over buffer boundaries were not parsed properly. This eventually
827 resulted in duplicates being sent. This issue only became common enough
828 to notice due to the introduction of connection certificate information,
829 the item size being so much larger. Found and fixed by Wolfgang Breyha.
831 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
832 size buffer was used, resulting in syntax errors when an expansion
835 JH/07 Add support for directories of certificates when compiled with a GnuTLS
836 version 3.3.6 or later.
838 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
839 is EXPERIMENTAL_EVENT, the main-configuration and transport options
840 both become "event_action", the variables become $event_name, $event_data
841 and $event_defer_errno. There is a new variable $verify_mode, usable in
842 routers, transports and related events. The tls:cert event is now also
843 raised for inbound connections, if the main configuration event_action
846 TL/06 In test suite, disable OCSP for old versions of openssl which contained
847 early OCSP support, but no stapling (appears to be less than 1.0.0).
849 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
850 server certificate names available under the smtp transport option
851 "tls_verify_cert_hostname" now do not permit multi-component wildcard
854 JH/10 Time-related extraction expansions from certificates now use the main
855 option "timezone" setting for output formatting, and are consistent
856 between OpenSSL and GnuTLS compilations. Bug 1541.
858 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
859 encoded parameter in the incoming message. Bug 1558.
861 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
862 include certificate info, eximon was claiming there were spoolfile
865 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
867 JH/14 Log delivery-related information more consistently, using the sequence
868 "H=<name> [<ip>]" wherever possible.
870 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
871 are problematic for Debian distribution, omit them from the release
874 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
876 JH/16 Fix string representation of time values on 64bit time_t architectures.
879 JH/17 Fix a null-indirection in certextract expansions when a nondefault
880 output list separator was used.
885 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
886 checkers that were complaining about end of non-void function with no
889 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
890 This was a regression introduced in 4.83 by another bugfix.
892 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
894 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
895 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
901 TF/01 Correctly close the server side of TLS when forking for delivery.
903 When a message was received over SMTP with TLS, Exim failed to clear up
904 the incoming connection properly after forking off the child process to
905 deliver the message. In some situations the subsequent outgoing
906 delivery connection happened to have the same fd number as the incoming
907 connection previously had. Exim would try to use TLS and fail, logging
908 a "Bad file descriptor" error.
910 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
911 utilities have not been installed.
913 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
914 temporary space as the ACL may create new global variables.
916 TL/01 LDAP support uses per connection or global context settings, depending
917 upon the detected version of the libraries at build time.
919 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
920 to extract and use the src ip:port in logging and expansions as if it
921 were a direct connection from the outside internet. PPv2 support was
922 updated based on HAProxy spec change in May 2014.
924 JH/02 Add ${listextract {number}{list}{success}{fail}}.
926 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
927 Properly escape header and check for NULL return.
929 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
932 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
934 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
935 characters in header names, implemented as a verify condition.
936 Contributed by Michael Fischer v. Mollard.
938 TL/05 Rename SPF condition results err_perm and err_temp to standardized
939 results permerror and temperror. Previous values are deprecated but
940 still accepted. In a future release, err_perm and err_temp will be
941 completely removed, which will be a backward incompatibility if the
942 ACL tests for either of these two old results. Patch contributed by
943 user bes-internal on the mailing list.
945 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
947 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
948 selectors, in both main and reject logs.
950 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
953 JH/07 Add malware type "sock" for talking to simple daemon.
955 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
957 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
958 routers/transports under cutthrough routing.
960 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
961 numbers. Touch up "bool" conditional to keep the same definition.
963 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
965 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
967 JH/12 Expand items in router/transport headers_add or headers_remove lists
968 individually rather than the list as a whole. Bug 1452.
970 Required for reasonable handling of multiple headers_ options when
971 they may be empty; requires that headers_remove items with embedded
972 colons must have them doubled (or the list-separator changed).
974 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
975 view the policy declared in the DMARC record. Currently, $dmarc_status
976 is a combined value of both the record presence and the result of the
979 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
981 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
982 dnslookup router and the smtp transport (applying to the forward
985 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
986 of ldap servers used for a specific lookup. Patch provided by Heiko
989 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
990 New variable $lookup_dnssec_authenticated for observability.
992 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
993 Patch submitted by Lars Timman.
995 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
997 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
998 Requires trusted mode and valid format message id, aborts otherwise.
999 Patch contributed by Heiko Schlichting.
1001 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1002 certextract with support for various fields. Bug 1358.
1004 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1005 is requested by default, modifiable by smtp transport option
1008 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1009 operate on certificate variables to give certificate fingerprints
1010 Also new ${sha256:cert_variable}.
1012 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1014 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1015 Christian Aistleitner.
1017 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1019 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1020 file. Patch from Wolfgang Breyha.
1022 JH/25 Expand the coverage of the delivery $host and $host_address to
1023 client authenticators run in verify callout. Bug 1476.
1025 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1026 align with daemon_smtp_ports. Bug 72.
1028 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1029 support and error reporting did not work properly.
1031 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1032 and is readable. Patch from Andrew Colin Kissa.
1034 TL/14 Enhance documentation of ${run expansion and how it parses the
1035 commandline after expansion, particularly in the case when an
1036 unquoted variable expansion results in an empty value.
1038 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1040 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1041 by renaming away from it. Addresses GH issue 15, reported by
1044 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1046 TL/15 SECURITY: prevent double expansion in math comparison functions
1047 (can expand unsanitized data). Not remotely exploitable.
1054 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1056 PP/02 Make -n do something, by making it not do something.
1057 When combined with -bP, the name of an option is not output.
1059 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1062 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1063 $sender_host_name and config options to manage this, and basic check
1066 PP/05 DSCP support for outbound connections and control modifier for inbound.
1068 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1069 (Only plugin which currently uses this is kerberos4, which nobody should
1070 be using, but we should make it available and other future plugins might
1071 conceivably use it, even though it would break NAT; stuff *should* be
1072 using channel bindings instead).
1074 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1075 name; added for Sendmail compatibility; requires admin caller.
1076 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1077 just ignore it); requires trusted caller.
1078 Also parse but ignore: -Ac -Am -X<logfile>
1081 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1083 TL/02 Add +smtp_confirmation as a default logging option.
1085 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1086 Patch by Magnus Holmgren from 2007-02-20.
1088 TL/04 Bugzilla 1281 - Spec typo.
1089 Bugzilla 1283 - Spec typo.
1090 Bugzilla 1290 - Spec grammar fixes.
1092 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1094 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1096 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1097 mailing list by by Dmitry Isaikin.
1099 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1101 JH/02 Support "G" suffix to numbers in ${if comparisons.
1103 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1105 NM/01 Bugzilla 1197 - Spec typo
1106 Bugzilla 1196 - Spec examples corrections
1108 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1110 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1111 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1114 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1115 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1117 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1118 "acl {{name}{arg}...}", and optional args on acl condition
1121 JH/05 Permit multiple router/transport headers_add/remove lines.
1123 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1125 JH/07 Avoid using a waiting database for a single-message-only transport.
1126 Performance patch from Paul Fisher. Bugzilla 1262.
1128 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1131 JH/09 Add $headers_added variable, with content from use of ACL modifier
1132 add_header (but not yet added to the message). Bugzilla 199.
1134 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1135 Pulled from Bugzilla 817 by Wolfgang Breyha.
1137 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1139 (nb: this is the same fix as in Exim 4.80.1)
1141 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1144 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1147 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1149 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1150 advertise SMTP AUTH mechanism to us, instead of a generic
1151 protocol violation error. Also, make Exim more robust to bad
1152 data from the Dovecot auth socket.
1154 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1156 When a queue runner is handling a message, Exim first routes the
1157 recipient addresses, during which it prunes them based on the retry
1158 hints database. After that it attempts to deliver the message to
1159 any remaining recipients. It then updates the hints database using
1162 So if a recipient address works intermittently, it can get repeatedly
1163 deferred at routing time. The retry hints record remains fresh so the
1164 address never reaches the final cutoff time.
1166 This is a fairly common occurrence when a user is bumping up against
1167 their storage quota. Exim had some logic in its local delivery code
1168 to deal with this. However it did not apply to per-recipient defers
1169 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1171 This change adds a proper retry rule check during routing so that the
1172 final cutoff time is checked against the message's age. We only do
1173 this check if there is an address retry record and there is not a
1174 domain retry record; this implies that previous attempts to handle
1175 the address had the retry_use_local_parts option turned on. We use
1176 this as an approximation for the destination being like a local
1177 delivery, as in LMTP.
1179 I suspect this new check makes the old local delivery cutoff check
1180 redundant, but I have not verified this so I left the code in place.
1182 TF/02 Correct gecos expansion when From: is a prefix of the username.
1184 Test 0254 submits a message to Exim with the header
1188 When I ran the test suite under the user fanf2, Exim expanded
1189 the header to contain my full name, whereas it should have added
1190 a Resent-Sender: header. It erroneously treats any prefix of the
1191 username as equal to the username.
1193 This change corrects that bug.
1195 GF/01 DCC debug and logging tidyup
1196 Error conditions log to paniclog rather than rejectlog.
1197 Debug lines prefixed by "DCC: " to remove any ambiguity.
1199 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1201 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1202 Bug spotted by Jeremy Harris; was flawed since initial commit.
1203 Would have resulted in OCSP responses post-SNI triggering an Exim
1204 NULL dereference and crash.
1206 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1208 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1209 Bug detection, analysis and fix by Samuel Thibault.
1210 Bugzilla 1331, Debian bug #698092.
1212 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1214 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1215 Server implementation by Todd Lyons, client by JH.
1216 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1217 config variable "prdr_enable" controls whether the server
1218 advertises the facility. If the client requests PRDR a new
1219 acl_data_smtp_prdr ACL is called once for each recipient, after
1220 the body content is received and before the acl_smtp_data ACL.
1221 The client is controlled by both of: a hosts_try_prdr option
1222 on the smtp transport, and the server advertisement.
1223 Default client logging of deliveries and rejections involving
1224 PRDR are flagged with the string "PRDR".
1226 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1227 fclose(). Diagnosis by Todd Lyons.
1229 PP/17 Update configure.default to handle IPv6 localhost better.
1230 Patch by Alain Williams (plus minor tweaks).
1233 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1234 This is now consistent with GnuTLS, and is now documented: the
1235 previous undocumented portable approach to treating the option as
1236 unset was to force an expansion failure. That still works, and
1237 an empty string is now equivalent.
1239 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1240 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1241 not performing validation itself.
1243 PP/20 Added force_command boolean option to pipe transport.
1244 Patch from Nick Koston, of cPanel Inc.
1246 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1249 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1251 PP/21 Fix eximon continuous updating with timestamped log-files.
1252 Broken in a format-string cleanup in 4.80, missed when I repaired the
1253 other false fix of the same issue.
1254 Report and fix from Heiko Schlichting.
1257 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1258 Report from Prashanth Katuri.
1260 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1261 It's SecureTransport, so affects any MacOS clients which use the
1262 system-integrated TLS libraries, including email clients.
1264 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1265 using a MIME ACL for non-SMTP local injection.
1266 Report and assistance in diagnosis by Warren Baker.
1268 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1270 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1272 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1273 last id that failed so it may be referenced in subsequent ACL's.
1275 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1278 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1279 ldap library initialization, allowing self-signed CA's to be
1280 used. Also properly sets require_cert option later in code by
1281 using NULL (global ldap config) instead of ldap handle (per
1282 session). Bug diagnosis and testing by alxgomz.
1284 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1285 the src/util/ subdirectory.
1287 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1288 renamed to Transport Post Delivery Action by Jeremy Harris, as
1291 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1292 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1293 redis_servers = needs to be configured which will be used by the redis
1294 lookup. Patch from Warren Baker, of The Packet Hub.
1296 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1298 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1299 hostname or reverse DNS when processing a host list. Used suggestions
1300 from multiple comments on this bug.
1302 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1304 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1305 Missed a few lines, added it to make the runtest require no keyboard
1308 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1309 contains upper case chars. Make router use caseful_local_part.
1311 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1312 support when GnuTLS has been built with p11-kit.
1318 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1320 This, or similar/improved, will also be change PP/11 of 4.82.
1326 PP/01 Handle short writes when writing local log-files.
1327 In practice, only affects FreeBSD (8 onwards).
1328 Bugzilla 1053, with thanks to Dmitry Isaikin.
1330 NM/01 Bugzilla 949 - Documentation tweak
1332 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1335 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1337 PP/02 Implemented gsasl authenticator.
1339 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1341 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1342 `pkg-config foo` for cflags/libs.
1344 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1345 with rest of GSASL and with heimdal_gssapi.
1347 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1348 `pkg-config foo` for cflags/libs for the TLS implementation.
1350 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1351 properties get this fed in as external SSF. A number of robustness
1352 and debugging improvements to the cyrus_sasl authenticator.
1354 PP/08 cyrus_sasl server now expands the server_realm option.
1356 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1357 Patch by Jeremy Harris.
1359 PP/10 Added dbmjz lookup type.
1361 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1363 PP/12 MAIL args handles TAB as well as SP, for better interop with
1364 non-compliant senders.
1365 Analysis and variant patch by Todd Lyons.
1367 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1368 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1369 Patch from Dirk Mueller <dmueller@suse.com>
1371 PP/13 tls_peerdn now print-escaped for spool files.
1372 Observed some $tls_peerdn in wild which contained \n, which resulted
1373 in spool file corruption.
1375 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1376 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1377 or write after TLS renegotiation, which otherwise led to messages
1380 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1381 as a tracking header (ie: a signed header comes before the signature).
1382 Patch from Wolfgang Breyha.
1384 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1385 comma-sep list; embedded commas doubled.
1387 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1389 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1391 Report and patch from Dmitry Banschikov.
1393 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1394 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1395 failures. We appear to now support TLS1.1+ with Exim.
1397 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1398 lets Exim select keys and certificates based upon TLS SNI from client.
1399 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1400 before an outbound SMTP session. New log_selector, +tls_sni.
1402 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1403 NULL dereference. Report and patch from Alun Jones.
1405 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1406 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1407 Not seeing resolver debug output on NetBSD, but suspect this is a
1408 resolver implementation change.
1410 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1411 Left warnings. Added "eximon gdb" invocation mode.
1413 PP/21 Defaulting "accept_8bitmime" to true, not false.
1415 PP/22 Added -bw for inetd wait mode support.
1417 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1418 locate the relevant includes and libraries. Made this the default.
1420 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1421 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1423 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1424 This may cause build issues on older platforms.
1426 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1427 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1428 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1429 Added SNI support via GnuTLS too.
1430 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1432 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1434 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1435 Applied second patch from Janne, implementing suggestion to default
1436 multiple-strings-in-record handling to match SPF spec.
1438 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1440 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1441 read-only, out of scope).
1442 Patch from Wolfgang Breyha, report from Stuart Northfield.
1444 PP/29 Fix three issues highlighted by clang analyser static analysis.
1445 Only crash-plausible issue would require the Cambridge-specific
1446 iplookup router and a misconfiguration.
1447 Report from Marcin Mirosław.
1449 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1451 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1452 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1453 As part of this, removing so much warning spew let me fix some minor
1454 real issues in debug logging.
1456 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1457 assignment on my part. Fixed.
1459 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1460 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1461 Janne Snabb (who went above and beyond: thank you).
1463 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1464 string otherwise requires a connection and a bunch more work and it's
1465 relatively easy to get wrong. Should also expose TLS library linkage
1468 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1469 64-bit ${eval} (JH/03).
1471 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1472 GNU libc to support some of the 64-bit stuff, should not lead to
1473 conflicts. Defined before os.h is pulled in, so if a given platform
1474 needs to override this, it can.
1476 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1477 protection layer was required, which is not implemented.
1478 Bugzilla 1254, patch from Wolfgang Breyha.
1480 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1481 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1482 tls_dhparam take prime identifiers. Also unbreak combination of
1483 OpenSSL+DH_params+TLSSNI.
1485 PP/39 Disable SSLv2 by default in OpenSSL support.
1491 PP/01 Solaris build fix for Oracle's LDAP libraries.
1492 Bugzilla 1109, patch from Stephen Usher.
1494 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1496 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1499 TF/02 Fix a couple more cases where we did not log the error message
1500 when unlink() failed. See also change 4.74-TF/03.
1502 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1503 lock up or crash if it happened to be inside a call to libc when it
1504 got a SIGUSR1 from exiwhat.
1506 The SIGUSR1 handler appends the current process status to the process
1507 log which is later printed by exiwhat. It used to use the general
1508 purpose logging code to do this, but several functions it calls are
1509 not safe for signals.
1511 The new output code in the SIGUSR1 handler is specific to the process
1512 log, and simple enough that it's easy to inspect for signal safety.
1513 Removing some special cases also simplifies the general logging code.
1514 Removing the spurious timestamps from the process log simplifies
1517 TF/04 Improved ratelimit ACL condition.
1519 The /noupdate option has been deprecated in favour of /readonly which
1520 has clearer semantics. The /leaky, /strict, and /readonly update modes
1521 are mutually exclusive. The update mode is no longer included in the
1522 database key; it just determines when the database is updated. (This
1523 means that when you upgrade Exim will forget old rate measurements.)
1525 Exim now checks that the per_* options are used with an update mode that
1526 makes sense for the current ACL. For example, when Exim is processing a
1527 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1528 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1529 must specify per_mail/readonly. If you omit the update mode it defaults to
1530 /leaky where that makes sense (as before) or /readonly where required.
1532 The /noupdate option is now undocumented but still supported for
1533 backwards compatibility. It is equivalent to /readonly except that in
1534 ACLs where /readonly is required you may specify /leaky/noupdate or
1535 /strict/noupdate which are treated the same as /readonly.
1537 A useful new feature is the /count= option. This is a generalization
1538 of the per_byte option, so that you can measure the throughput of other
1539 aggregate values. For example, the per_byte option is now equivalent
1540 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1542 The per_rcpt option has been generalized using the /count= mechanism
1543 (though it's more complicated than the per_byte equivalence). When it is
1544 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1545 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1546 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1547 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1548 Note that using per_rcpt with a non-readonly update mode in more than
1549 one ACL will cause the recipients to be double-counted. (The per_mail
1550 and per_byte options don't have this problem.)
1552 The handling of very low rates has changed slightly. If the computed rate
1553 is less than the event's count (usually one) then this event is the first
1554 after a long gap. In this case the rate is set to the same as this event's
1555 count, so that the first message of a spam run is counted properly.
1557 The major new feature is a mechanism for counting the rate of unique
1558 events. The new per_addr option counts the number of different
1559 recipients that someone has sent messages to in the last time period. It
1560 behaves like per_rcpt if all the recipient addresses are different, but
1561 duplicate recipient addresses do not increase the measured rate. Like
1562 the /count= option this is a general mechanism, so the per_addr option
1563 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1564 example, measure the rate that a client uses different sender addresses
1565 with the options per_mail/unique=$sender_address. There are further
1566 details in the main documentation.
1568 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1570 TF/06 Removed a few PCRE remnants.
1572 TF/07 Automatically extract Exim's version number from tags in the git
1573 repository when doing development or release builds.
1575 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1576 Bugzilla 879. Patch from Paul Fisher.
1578 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1579 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1582 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1584 PP/05 Variable $av_failed, true if the AV scanner deferred.
1585 Bugzilla 1078. Patch from John Horne.
1587 PP/06 Stop make process more reliably on build failure.
1588 Bugzilla 1087. Patch from Heiko Schlittermann.
1590 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1591 Bugzilla 1089. Patch from Heiko Schlittermann.
1593 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1594 Bugzilla 1131. Patch from Holger Weiß.
1596 PP/09 Handle IPv6 addresses with SPF.
1597 Bugzilla 860. Patch from Wolfgang Breyha.
1599 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1601 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1604 PP/11 match_* no longer expand right-hand-side by default.
1605 New compile-time build option, EXPAND_LISTMATCH_RHS.
1606 New expansion conditions, "inlist", "inlisti".
1608 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1610 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1612 PP/14 fix log_write() format string regression from TF/03.
1613 Bugzilla 1152. Patch from Dmitry Isaikin.
1619 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1621 PP/02 Harmonised TLS library version reporting; only show if debugging.
1622 Layout now matches that introduced for other libraries in 4.74 PP/03.
1624 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1626 PP/04 New "dns_use_edns0" global option.
1628 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1631 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1632 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1634 TK/01 Updated PolarSSL code to 0.14.2.
1635 Bugzilla 1097. Patch from Andreas Metzler.
1637 PP/07 Catch divide-by-zero in ${eval:...}.
1638 Fixes bugzilla 1102.
1640 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1643 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1644 format-string attack -- SECURITY: remote arbitrary code execution.
1646 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1647 time unintentionally subject to list matching rules, letting the header
1648 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1649 arbitrary string expansion). This allowed for information disclosure.
1651 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1652 INT_MIN/-1 -- value coerced to INT_MAX.
1658 NM/01 Workaround for PCRE version dependency in version reporting
1661 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1662 This fixes portability to compilers other than gcc, notably
1663 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1665 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1666 makefiles for portability to HP-UX and POSIX correctness.
1668 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1669 Also via indented variable definition in the Makefile.
1670 (Debugging by Oliver Heesakkers).
1672 PP/02 Restore caching of spamd results with expanded spamd_address.
1673 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1675 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1676 Improves build reliability. Fix from: Frank Elsner
1678 NM/02 Fix wide character breakage in the rfc2047 coding
1679 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1681 NM/03 Allow underscore in dnslist lookups
1682 Fixes bug 1026. Patch from Graeme Fowler
1684 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1685 Code patches from Adam Ciarcinski of NetBSD.
1687 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1690 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1691 is logged, to avoid truncation. Patch from John Horne.
1693 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1694 Patch from Jakob Hirsch.
1696 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1697 SQL string expansion failure details.
1698 Patch from Andrey Oktyabrski.
1700 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1701 Patch from Simon Arlott.
1703 PP/09 New lookups functionality failed to compile on old gcc which rejects
1704 extern declarations in function scope.
1705 Patch from Oliver Fleischmann
1707 PP/10 Use sig_atomic_t for flags set from signal handlers.
1708 Check getgroups() return and improve debugging.
1709 Fixed developed for diagnosis in bug 927 (which turned out to be
1712 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1713 Patch from Mark Zealey.
1715 PP/12 Bugzilla 1056: Improved spamd server selection.
1716 Patch from Mark Zealey.
1718 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1719 Based on patch from Heiko Schlittermann.
1721 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1722 Patch from Uwe Doering, sign-off by Michael Haardt.
1724 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1725 variable declaration deep within a block. Bug and patch from
1728 PP/15 lookups-Makefile IRIX compatibility coercion.
1730 PP/16 Make DISABLE_DKIM build knob functional.
1732 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1733 Patch by Simon Arlott
1735 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1736 variable argument macros. Our copy now differs from upstream.
1742 TF/01 Failure to get a lock on a hints database can have serious
1743 consequences so log it to the panic log.
1745 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1746 controlled using the smtp_confirmation log selector.
1748 TF/03 Include the error message when we fail to unlink a spool file.
1750 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1751 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1752 for maintaining out-of-tree patches for some time.
1754 PP/01 Bugzilla 139: Documentation and portability issues.
1755 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1756 Handle per-OS dynamic-module compilation flags.
1758 PP/02 Let /dev/null have normal permissions.
1759 The 4.73 fixes were a little too stringent and complained about the
1760 permissions on /dev/null. Exempt it from some checks.
1761 Reported by Andreas M. Kirchwitz.
1763 PP/03 Report version information for many libraries, including
1764 Exim version information for dynamically loaded libraries. Created
1765 version.h, now support a version extension string for distributors
1766 who patch heavily. Dynamic module ABI change.
1768 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1769 privilege escalation vulnerability whereby the Exim run-time user
1770 can cause root to append content of the attacker's choosing to
1773 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1776 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1777 If dropping privileges for untrusted macros, we disabled normal logging
1778 on the basis that it would fail; for the Exim run-time user, this is not
1779 the case, and it resulted in successful deliveries going unlogged.
1780 Fixed. Reported by Andreas Metzler.
1786 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1787 only prepend for the Resent-* case. Fixes regression introduced in
1788 Exim 4.70 by NM/22 for Bugzilla 607.
1790 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1791 increasing numbers of administrators be bitten by this.
1793 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1795 PP/03 Bugzilla 994: added openssl_options main configuration option.
1797 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1799 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1801 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1803 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1804 without a peer certificate, leading to a segfault because of an
1805 assumption that peers always have certificates. Be a little more
1806 paranoid. Problem reported by Martin Tscholak.
1808 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1809 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1810 NB: ClamAV planning to remove STREAM in "middle of 2010".
1811 CL also introduces -bmalware, various -d+acl logging additions and
1812 more caution in buffer sizes.
1814 PP/09 Implemented reverse_ip expansion operator.
1816 PP/10 Bugzilla 937: provide a "debug" ACL control.
1818 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1820 PP/12 Bugzilla 973: Implement --version.
1822 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1824 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1826 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1828 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1829 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1830 ignore trailing whitespace.
1832 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1834 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1837 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1838 Notification from Dr Andrew Aitchison.
1840 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1841 ExtendedDetectionInfo response format.
1842 Notification from John Horne.
1844 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1847 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1848 XSL and documented dependency on system catalogs, with examples of how
1851 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1854 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1855 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1856 configuration file which is writeable by the Exim user or group.
1858 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1859 of configuration files to cover files specified with the -C option if
1860 they are going to be used with root privileges, not just the default
1863 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1864 option (effectively making it always true).
1866 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1867 files to be used while preserving root privileges.
1869 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1870 that rogue child processes cannot use them.
1872 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1873 run-time user, instead of root.
1875 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1876 Exim run-time user without dropping privileges.
1878 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1879 result string, instead of calling string_vformat() twice with the same
1882 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1883 for other users. Others should always drop root privileges if they use
1884 -C on the command line, even for a whitelisted configure file.
1886 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1888 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1894 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1895 $data_path, and $header_path variables; fixed documentation bugs and
1898 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1899 exipick to access non-standard spools, including the "frozen" queue
1902 NM/01 Bugzilla 965: Support mysql stored procedures.
1903 Patch from Alain Williams
1905 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1907 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1908 Patch from Andreas Metzler
1910 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1911 Patch from Kirill Miazine
1913 NM/05 Bugzilla 671: Added umask to procmail example.
1915 JJ/03 installed exipick 20100323.0, fixing doc bug
1917 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1918 directory. Notification and patch from Dan Rosenberg.
1920 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1922 TK/02 Improve log output when DKIM signing operation fails.
1924 MH/01 Treat the transport option dkim_domain as a colon separated
1925 list, not as a single string, and sign the message with each element,
1926 omitting multiple occurences of the same signer.
1928 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1929 Bugzilla 985, 986. Patch by Simon Arlott
1931 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1932 Patch by Simon Arlott
1934 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1935 MBX locking. Notification from Dan Rosenberg.
1941 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1943 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1945 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1947 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1949 NM/04 Bugzilla 915: Flag broken perl installation during build.
1955 TK/01 Added patch by Johannes Berg that expands the main option
1956 "spamd_address" if it starts with a dollar sign.
1958 TK/02 Write list of recipients to X-Envelope-Sender header when building
1959 the mbox-format spool file for content scanning (suggested by Jakob
1962 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1963 (http://www.dcc-servers.net/) support via dccifd. Activated by
1964 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1966 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1967 by Mark Daniel Reidel <mr@df.eu>.
1969 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1970 When building exim an external PCRE library is now needed -
1971 PCRE is a system library on the majority of modern systems.
1972 See entry on PCRE_LIBS in EDITME file.
1974 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1975 conversation. Added nologin parameter to request.
1976 Patch contributed by Kirill Miazine.
1978 TF/01 Do not log submission mode rewrites if they do not change the address.
1980 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1982 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1983 log files in place. Contributed by Roberto Lima.
1985 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1987 TF/03 Bugzilla 615: When checking the local_parts router precondition
1988 after a local_part_suffix or local_part_prefix option, Exim now
1989 does not use the address's named list lookup cache, since this
1990 contains cached lookups for the whole local part.
1992 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1993 Robert Millan. Documentation is in experimental-spec.txt.
1995 TF/04 Bugzilla 668: Fix parallel build (make -j).
1997 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1999 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2000 Patch provided by Jan Srzednicki.
2002 TF/05 Leading white space used to be stripped from $spam_report which
2003 wrecked the formatting. Now it is preserved.
2005 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2006 that they are available at delivery time.
2008 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2010 TF/08 TLS error reporting now respects the incoming_interface and
2011 incoming_port log selectors.
2013 TF/09 Produce a more useful error message if an SMTP transport's hosts
2014 setting expands to an empty string.
2016 NM/06 Bugzilla 744: EXPN did not work under TLS.
2017 Patch provided by Phil Pennock.
2019 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2020 Patch provided by Richard Godbee.
2022 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2023 acl_smtp_notquit, added index entry.
2025 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2026 Patch provided by Eugene Bujak.
2028 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2029 accept(). Patch provided by Maxim Dounin.
2031 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2032 Patch provided by Phil Pennock.
2034 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2036 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2037 Patch provided by Brad "anomie" Jorsch.
2039 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2040 Patch provided by Dean Brooks.
2042 TK/05 Add native DKIM support (does not depend on external libraries).
2044 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2045 Patch provided by Graeme Fowler.
2047 NM/16 Bugzilla 851: Documentation example syntax fix.
2049 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2051 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2054 NM/19 Bugzilla 745: TLS version reporting.
2055 Patch provided by Phil Pennock.
2057 NM/20 Bugzilla 167: bool: condition support.
2058 Patch provided by Phil Pennock.
2060 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2061 clients. Patch provided by Phil Pennock.
2063 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2064 Patch provided by Brad "anomie" Jorsch.
2066 NM/23 Bugzilla 687: Fix misparses in eximstats.
2067 Patch provided by Heiko Schlittermann.
2069 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2070 Patch provided by Heiko Schlittermann.
2072 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2073 plus update to original patch.
2075 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2077 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2078 Patch provided by David Brownlee.
2080 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2082 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2084 NM/30 Bugzilla 888: TLS documentation bugfixes.
2086 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2088 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2089 Unlike the original bugzilla I have changed all shell scripts in src tree.
2091 NM/33 Bugzilla 898: Transport filter timeout fix.
2092 Patch by Todd Rinaldo.
2094 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2095 Patch by Serge Demonchaux.
2097 NM/35 Bugzilla 39: Base64 decode bug fixes.
2098 Patch by Jakob Hirsch.
2100 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2102 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2104 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2106 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2112 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2113 ALT-N's libdkim that I have put here:
2114 http://duncanthrax.net/exim-experimental/
2116 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2117 were called 'true' and it seems that C99 defines that as a reserved
2118 keyword to be used with 'bool' variable types. That means you could
2119 not include C99-style headers which use bools without triggering
2120 build errors in sieve.c.
2122 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2123 as mailq or other aliases. Changed the --help handling significantly
2124 to do whats expected. exim_usage() emits usage/help information.
2126 SC/01 Added the -bylocaldomain option to eximstats.
2128 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2130 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2132 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2138 PH/01 Another patch from the Sieve maintainer.
2140 PH/02 When an IPv6 address is converted to a string for single-key lookup
2141 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2142 dots are used instead of colons so that keys in lsearch files need not
2143 contain colons. This was done some time before quoting was made available
2144 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2145 (notated using the quote facility) so as to distinguish them from IPv4
2146 keys. This meant that lookups for IP addresses in host lists did not work
2147 for iplsearch lookups.
2149 This has been fixed by arranging for IPv6 addresses to be expressed with
2150 colons if the lookup type is iplsearch. This is not incompatible, because
2151 previously such lookups could never work.
2153 The situation is now rather anomalous, since one *can* have colons in
2154 ordinary lsearch keys. However, making the change in all cases is
2155 incompatible and would probably break a number of configurations.
2157 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2160 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2161 conversion specification without a maximum field width, thereby enabling
2162 a rogue spamd server to cause a buffer overflow. While nobody in their
2163 right mind would setup Exim to query an untrusted spamd server, an
2164 attacker that gains access to a server running spamd could potentially
2165 exploit this vulnerability to run arbitrary code as the Exim user.
2167 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2168 $primary_hostname instead of what libspf2 thinks the hosts name is.
2170 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2171 a directory entry by the name of the lookup key. Previously, if a
2172 symlink pointed to a non-existing file or a file in a directory that
2173 Exim lacked permissions to read, a lookup for a key matching that
2174 symlink would fail. Now it is enough that a matching directory entry
2175 exists, symlink or not. (Bugzilla 503.)
2177 PH/03 The body_linecount and body_zerocount variables are now exported in the
2180 PH/04 Added the $dnslist_matched variable.
2182 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2183 This means they are set thereafter only if the connection becomes
2186 PH/06 Added the client_condition to authenticators so that some can be skipped
2187 by clients under certain conditions.
2189 PH/07 The error message for a badly-placed control=no_multiline_responses left
2190 "_responses" off the end of the name.
2192 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2194 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2195 (without spaces) instead of just copying the configuration text.
2197 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2199 PH/11 Added $max_received_linelength.
2201 PH/12 Added +ignore_defer and +include_defer to host lists.
2203 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2204 way in which PCRE > 7.0 is built.
2206 PH/14 Implemented queue_only_load_latch.
2208 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2209 MAIL command. The effect was to mangle the value on 64-bit systems.
2211 PH/16 Another patch from the Sieve maintainer.
2213 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2215 PH/18 If a system quota error occurred while trying to create the file for
2216 a maildir delivery, the message "Mailbox is full" was not appended to the
2217 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2218 only to a quota excession during the actual writing of the file.
2220 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2221 characters?) which causes problems in log lines. The DN values are now
2222 passed through string_printing() before being added to log lines.
2224 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2225 and InterBase are left for another time.)
2227 PH/21 Added message_body_newlines option.
2229 PH/22 Guard against possible overflow in moan_check_errorcopy().
2231 PH/23 POSIX allows open() to be a macro; guard against that.
2233 PH/24 If the recipient of an error message contained an @ in the local part
2234 (suitably quoted, of course), incorrect values were put in $domain and
2235 $local_part during the evaluation of errors_copy.
2241 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2242 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2245 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2246 issue a MAIL command.
2248 PH/02 In an ACL statement such as
2250 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2252 if a client was not listed at all, or was listed with a value other than
2253 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2254 the condition was not true (as it should be), so access was not denied.
2255 The bug was that the ! inversion was incorrectly passed on to the second
2256 item. This has been fixed.
2258 PH/03 Added additional dnslists conditions == and =& which are different from
2259 = and & when the dns lookup returns more than one IP address.
2261 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2262 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2264 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2265 FSYNC, which compiles an option called disable_fsync that allows for
2266 bypassing fsync(). The documentation is heavily laced with warnings.
2268 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2270 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2271 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2272 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2273 including adding "make clean"; (3) Added -fPIC when compiling the test
2274 dynamically loaded module, to get rid of a warning.
2276 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2277 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2278 = 0s. The bug is otherwise harmless.
2280 PH/07 There was a bug in the dovecot authenticator such that the value of
2281 $auth1 could be overwritten, and so not correctly preserved, after a
2282 successful authentication. This usually meant that the value preserved by
2283 the server_setid option was incorrect.
2285 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2287 PH/09 Installed PCRE release 7.0.
2289 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2290 run for batched SMTP input. It is now run at the start of every message
2291 in the batch. While fixing this I discovered that the process information
2292 (output by running exiwhat) was not always getting set for -bs and -bS
2293 input. This is fixed, and it now also says "batched" for BSMTP.
2295 PH/11 Added control=no_pipelining.
2297 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2298 patch, slightly modified), and move the expansion of helo_data till after
2299 the connection is made in the smtp transport (so it can use these
2302 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2304 PH/14 Added log_selector = +pid.
2306 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2308 PH/16 Add ${if forany and ${if forall.
2310 PH/17 Added dsn_from option to vary the From: line in DSNs.
2312 PH/18 Flush SMTP output before performing a callout, unless control =
2313 no_callout_flush is set.
2315 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2316 was true (the default) a successful delivery failed to delete the retry
2317 item, thus causing premature timeout of the address. The bug is now
2320 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2322 PH/21 Long custom messages for fakedefer and fakereject are now split up
2323 into multiline responses in the same way that messages for "deny" and
2324 other ACL rejections are.
2326 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2327 with slight modification.
2329 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2330 draft, changing the syntax and factoring some duplicate code".
2332 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2333 for deliveries of the second and subsequent messages over the same SMTP
2336 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2337 ${reduce, with only minor "tidies".
2339 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2341 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2342 expansion side effects.
2344 PH/27 When a message times out after an over-quota error from an Exim-imposed
2345 quota, the bounce message says "mailbox is full". This message was not
2346 being given when it was a system quota that was exceeded. It now should
2349 MH/03 Made $recipients available in local_scan(). local_scan() already has
2350 better access to the recipient list through recipients_list[], but
2351 $recipients can be useful in postmaster-provided expansion strings.
2353 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2354 in the case of a MAIL command with additional options following the
2355 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2356 were accidentally chopped off.
2358 PH/29 SMTP synchronization checks are implemented when a command is read -
2359 there is a check that no more input is waiting when there shouldn't be
2360 any. However, for some commands, a delay in an ACL can mean that it is
2361 some time before the response is written. In this time, more input might
2362 arrive, invalidly. So now there are extra checks after an ACL has run for
2363 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2364 pipelining has not been advertised.
2366 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2368 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2369 returned in the SMTP response when smtp_return_error_details was false.
2370 This has been fixed.
2372 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2373 instead of the C I/O that was originally supplied, because problems were
2374 reported on Solaris.
2376 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2377 Exim which did not show up earlier: it was assuming that a call to
2378 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2379 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2380 was a macro that became an assignment, so it seemed to work. This has
2381 changed to a proper function call with a void return, hence the compile
2382 error. Exim's code has been fixed.
2384 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2387 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2389 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2391 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2392 because it turns out to be too big or there is a timeout) there is no
2393 "Completed" line in the log. When some messages of this type were
2394 selected by exigrep, they were listed as "not completed". Others were
2395 picked up by some special patterns. I have improved the selection
2396 criteria to be more general.
2398 PH/38 The host_find_failed option in the manualroute router can now be set
2399 to "ignore", to completely ignore a host whose IP address cannot be
2400 found. If all hosts are ignored, the behaviour is controlled by the new
2401 host_all_ignored option.
2403 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2404 homing or because of multiple MX records with /mx) generated more than
2405 one IP address, and the following item turned out to be the local host,
2406 all the secondary addresses of the first item were incorrectly removed
2407 from the list, along with the local host and any following hosts (which
2408 is what is supposed to happen).
2410 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2411 whoever called Exim into the -H file. In the case of the daemon it was
2412 behaving confusingly. When first started, it used values for whoever
2413 started the daemon, but after a SIGHUP it used the Exim user (because it
2414 calls itself on a restart). I have changed the code so that it now always
2417 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2418 message are rejected with the same error (e.g. no authentication or bad
2419 sender address), and a DATA command is nevertheless sent (as can happen
2420 with PIPELINING or a stupid MUA), the error message that was given to the
2421 RCPT commands is included in the rejection of the DATA command. This is
2422 intended to be helpful for MUAs that show only the final error to their
2425 PH/42 Another patch from the Sieve maintainer.
2427 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2428 Eximstats - Fixed some broken HTML links and added missing column headers
2430 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2431 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2433 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2435 PH/43 Yet another patch from the Sieve maintainer.
2437 PH/44 I found a way to check for a TCP/IP connection going away before sending
2438 the response to the final '.' that terminates a message, but only in the
2439 case where the client has not sent further data following the '.'
2440 (unfortunately, this is allowed). However, in many cases there won't be
2441 any further data because there won't be any more messages to send. A call
2442 to select() can be used: if it shows that the input is "ready", there is
2443 either input waiting, or the socket has been closed. An attempt to read
2444 the next input character can distinguish the two cases. Previously, Exim
2445 would have sent an OK response which the client would never have see.
2446 This could lead to message repetition. This fix should cure that, at
2447 least in a lot of common cases.
2449 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2450 advertised in response to EHLO.
2456 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2457 fixed by 4.65/MH/01 (is this a record?) are fixed:
2459 (i) An empty string was always treated as zero by the numeric comparison
2460 operators. This behaviour has been restored.
2462 (ii) It is documented that the numeric comparison operators always treat
2463 their arguments as decimal numbers. This was broken in that numbers
2464 starting with 0 were being interpreted as octal.
2466 While fixing these problems I realized that there was another issue that
2467 hadn't been noticed. Values of message_size_limit (both the global option
2468 and the transport option) were treated as octal if they started with 0.
2469 The documentation was vague. These values are now always treated as
2470 decimal, and I will make that clear in the documentation.
2476 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2477 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2480 MH/01 Don't check that the operands of numeric comparison operators are
2481 integers when their expansion is in "skipping" mode (fixes bug
2482 introduced by 4.64-PH/07).
2484 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2485 child addresses, Exim now panics and dies. Previously, because the count
2486 is held in a short int, deliveries were likely to be lost. As such a
2487 large number of recipients for a single message is ridiculous
2488 (performance will be very, very poor), I have chosen to impose a limit
2489 rather than extend the field.
2495 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2496 leftover -K file (the existence of which was triggered by #402).
2497 While we were at it, introduced process PID as part of the -K
2498 filename. This should rule out race conditions when creating
2501 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2502 processing considerably. Previous code took too long for large mails,
2503 triggering a timeout which in turn triggers #401.
2505 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2506 in the DK code in transports.c. sendfile() is not really portable,
2507 hence the _LINUX specificness.
2509 TF/01 In the add_headers option to the mail command in an Exim filter,
2510 there was a bug that Exim would claim a syntax error in any
2511 header after the first one which had an odd number of characters
2514 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2515 callout verification, Exim cached a "reject" for the entire domain. This
2516 is correct for most verifications, but it is not correct for a recipient
2517 verification with use_sender or use_postmaster set, because in that case
2518 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2519 case of MAIL FROM:<> rejection from other early rejections (e.g.
2520 rejection of HELO). When verifying a recipient using a non-null MAIL
2521 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2522 Whatever the result of the callout, the value of the domain cache is
2523 left unchanged (for any other kind of callout, getting as far as trying
2524 RCPT means that the domain itself is ok).
2526 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2529 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2530 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2531 session, the daemon ignores ECONNECT errors and logs others; it now
2532 ignores EPIPE as well.
2534 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2535 (quoted-printable decoding).
2537 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2538 later the small subsequent patch to fix an introduced bug.
2540 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2542 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2544 PH/08 An error is now given if message_size_limit is specified negative.
2546 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2547 to be given (somewhat) arbitrary names.
2549 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2552 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2553 miscellaneous code fixes
2555 PH/10 Added the log_reject_target ACL modifier to specify where to log
2558 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2559 hostname. This is wrong, because it relates to the incoming message (and
2560 probably the interface on which it is arriving) and not to the outgoing
2561 callout (which could be using a different interface). This has been
2562 changed to use the value of the helo_data option from the smtp transport
2563 instead - this is what is used when a message is actually being sent. If
2564 there is no remote transport (possible with a router that sets up host
2565 addresses), $smtp_active_hostname is used.
2567 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2568 tweaks were necessary in order to get it to work (see also 21 below):
2569 (a) The code assumed that strncpy() returns a negative number on buffer
2570 overflow, which isn't the case. Replaced with Exim's string_format()
2572 (b) There were several signed/unsigned issues. I just did the minimum
2573 hacking in of casts. There is scope for a larger refactoring.
2574 (c) The code used strcasecmp() which is not a standard C function.
2575 Replaced with Exim's strcmpic() function.
2576 (d) The code set only $1; it now sets $auth1 as well.
2577 (e) A simple test gave the error "authentication client didn't specify
2578 service in request". It would seem that Dovecot has changed its
2579 interface. Fortunately there's a specification; I followed it and
2580 changed what the client sends and it appears to be working now.
2582 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2585 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2586 address A is aliased to B and C, where B exists and C does not. Without
2587 -v the output is "A verified" because verification stops after a
2588 successful redirection if more than one address is generated. However,
2589 with -v the child addresses are also verified. Exim was outputting "A
2590 failed to verify" and then showing the successful verification for C,
2591 with its parentage. It now outputs "B failed to verify", showing B's
2592 parentage before showing the successful verification of C.
2594 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2595 look up a TXT record in a specific list after matching in a combined
2598 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2599 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2600 they consult the DNS. I had assumed they would set it the way they
2601 wanted; and indeed my experiments on Linux seem to show that in some
2602 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2603 To be on the safe side, however, I have now made the interface to
2604 host_find_byname() similar to host_find_bydns(), with an argument
2605 containing the DNS resolver options. The host_find_byname() function now
2606 sets these options at its start, just as host_find_bydns() does. The smtp
2607 transport options dns_qualify_single and dns_search_parents are passed to
2608 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2609 of host_find_byname() use the default settings of RES_DEFNAMES
2610 (qualify_single) but not RES_DNSRCH (search_parents).
2612 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2613 spool_read_header() do less string testing, by means of a preliminary
2614 switch on the second character of optional "-foo" lines. (This is
2615 overdue, caused by the large number of possibilities that now exist.
2616 Originally there were few.) While I was there, I also converted the
2617 str(n)cmp tests so they don't re-test the leading "-" and the first
2618 character, in the hope this might squeeze out yet more improvement.
2620 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2621 flag allowing group syntax was set by the header_syntax check but not
2622 turned off, possible causing trouble later; (2) The flag was not being
2623 set at all for the header_verify test, causing "group"-style headers to
2624 be rejected. I have now set it in this case, and also caused header_
2625 verify to ignore an empty address taken from a group. While doing this, I
2626 came across some other cases where the code for allowing group syntax
2627 while scanning a header line wasn't quite right (mostly, not resetting
2628 the flag correctly in the right place). These bugs could have caused
2629 trouble for malformed header lines. I hope it is now all correct.
2631 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2632 with the "reply" argument non-NULL. The code, however (which originally
2633 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2634 but it didn't always do it. This confused somebody who was copying the
2635 code for some other use. I have removed all the tests.
2637 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2638 feature that was used to support insecure browsers during the U.S. crypto
2639 embargo. It requires special client support, and Exim is probably the
2640 only MTA that supported it -- and would never use it because real RSA is
2641 always available. This code has been removed, because it had the bad
2642 effect of slowing Exim down by computing (never used) parameters for the
2643 RSA_EXPORT functionality.
2645 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2646 authenticator to fail if there's a tab character in the incoming data
2647 (there should never be unless someone is messing about, as it's supposed
2648 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2649 if the connection is using TLS or if the remote IP is the same as the
2650 local IP, and the "valid-client-cert option" if a client certificate has
2653 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2654 authenticators. This can be used for authorization after authentication
2655 succeeds. (In the case of plaintext, it servers for both authentication
2658 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2659 if any retry times were supplied.
2661 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2662 connection, where there is no client IP address to check. In this
2663 situation, the verify now always succeeds.
2665 PH/25 Applied John Jetmore's -Mset patch.
2667 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2669 PH/27 In a string expansion for a processed (not raw) header when multiple
2670 headers of the same name were present, leading whitespace was being
2671 removed from all of them, but trailing whitespace was being removed only
2672 from the last one. Now trailing whitespace is removed from each header
2673 before concatenation. Completely empty headers in a concatenation (as
2674 before) are ignored.
2676 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2677 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2679 PH/29 [Removed. This was a change that I later backed out, and forgot to
2680 correct the ChangeLog entry (that I had efficiently created) before
2681 committing the later change.]
2683 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2684 address errors (4xx response to RCPT) over the same connection as other
2685 messages routed to the same hosts. Such deliveries are always "forced",
2686 so retry times are not inspected. This resulted in far too many retries
2687 for the affected addresses. The effect occurred only when there were more
2688 hosts than the hosts_max_try setting in the smtp transport when it had
2689 the 4xx errors. Those hosts that it had tried were not added to the list
2690 of hosts for which the message was waiting, so if all were tried, there
2691 was no problem. Two fixes have been applied:
2693 (i) If there are any address or message errors in an SMTP delivery, none
2694 of the hosts (tried or untried) are now added to the list of hosts
2695 for which the message is waiting, so the message should not be a
2696 candidate for sending over the same connection that was used for a
2697 successful delivery of some other message. This seems entirely
2698 reasonable: after all the message is NOT "waiting for some host".
2699 This is so "obvious" that I'm not sure why it wasn't done
2700 previously. Hope I haven't missed anything, but it can't do any
2701 harm, as the worst effect is to miss an optimization.
2703 (ii) If, despite (i), such a delivery is accidentally attempted, the
2704 routing retry time is respected, so at least it doesn't keep
2705 hammering the server.
2707 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2708 in ${readsocket because some servers need this prod.
2710 PH/32 Added some extra debug output when updating a wait-xxx database.
2712 PH/33 The hint "could be header name not terminated by colon", which has been
2713 given for certain expansion errors for a long time, was not being given
2714 for the ${if def:h_colon_omitted{... case.
2716 PH/34 The spec says: "With one important exception, whenever a domain list is
2717 being scanned, $domain contains the subject domain." There was at least
2718 one case where this was not true.
2720 PH/35 The error "getsockname() failed: connection reset by peer" was being
2721 written to the panic log as well as the main log, but it isn't really
2722 panic-worthy as it just means the connection died rather early on. I have
2723 removed the panic log writing for the ECONNRESET error when getsockname()
2726 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2727 runs only) independently of the message's sender address. This meant
2728 that, if the 4xx error was in fact related to the sender, a different
2729 message to the same recipient with a different sender could confuse
2730 things. In particular, this can happen when sending to a greylisting
2731 server, but other circumstances could also provoke similar problems.
2732 I have changed the default so that the retry time for these errors is now
2733 based a combination of the sender and recipient addresses. This change
2734 can be overridden by setting address_retry_include_sender=false in the
2737 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2738 remote server are returned as part of bounce messages. This was not
2739 happening for LMTP over a pipe (the lmtp transport), but now it is the
2740 same for both kinds of LMTP.
2742 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2743 in header lines that were in fact CNAMEs. This is no longer the case.
2745 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2746 and queue runs started by the daemon processed all messages. This has
2747 been fixed so that -R and -S can now usefully be given with -q<time>.
2749 PH/40 Import PCRE release 6.7 (fixes some bugs).
2751 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2753 PH/42 Give an error if -q is specified more than once.
2755 PH/43 Renamed the variables $interface_address and $interface_port as
2756 $received_ip_address and $received_port, to make it clear that these
2757 values apply to message reception, and not to the outgoing interface when
2758 a message is delivered. (The old names remain recognized, of course.)
2760 PH/44 There was no timeout on the connect() call when using a Unix domain
2761 socket in the ${readsocket expansion. There now is.
2763 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2764 be meaningful with "accept".
2766 SC/01 Eximstats V1.43
2767 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2769 SC/02 Eximstats V1.44
2770 Use a glob alias rather than an array ref in the generated
2771 parser. This improves both readability and performance.
2773 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2774 Collect SpamAssassin and rejection statistics.
2775 Don't display local sender or destination tables unless
2776 there is data to show.
2777 Added average volumes into the top table text output.
2779 SC/04 Eximstats V1.46
2780 Collect data on the number of addresses (recipients)
2781 as well as the number of messages.
2783 SC/05 Eximstats V1.47
2784 Added 'Message too big' to the list of mail rejection
2785 reasons (thanks to Marco Gaiarin).
2787 SC/06 Eximstats V1.48
2788 Mainlog lines which have GMT offsets and are too short to
2789 have a flag are now skipped.
2791 SC/07 Eximstats V1.49 (Alain Williams)
2792 Added the -emptyok flag.
2794 SC/08 Eximstats V1.50
2795 Fixes for obtaining the IP address from reject messages.
2797 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2798 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2799 whitespace changes from 4.64-PH/27
2801 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2804 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2805 are found, allow negative numbers in numeric criteria)
2807 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2809 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2810 to match changes made in 4.64-PH/43
2812 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2814 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2815 "domain" is actually an IP address, and force a failure. This locks out
2816 those revolvers/nameservers that support "A-for-A" lookups, in
2817 contravention of the specifications.
2819 PH/48 When a host name was looked up from an IP address, and the subsequent
2820 forward lookup of the name timed out, the host name was left in
2821 $sender_host_name, contrary to the specification.
2823 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2824 restricted to single-key lookups, Exim was not diagnosing an error if
2825 * or *@ was used with a query-style lookup.
2827 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2829 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2830 long ago, but noone interested enough thought of it. Let's just say that
2831 the "1.1" means that there are some new functions that weren't there at
2832 some point in the past.
2834 PH/51 Error processing for expansion failure of helo_data from an smtp
2835 transport during callout processing was broken.
2837 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2838 tested/used via the -bh/-bhc/-bs options.
2840 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2841 bug, fixed in subsequent PCRE releases).
2843 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2844 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2846 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2852 SC/01 Use a glob alias rather than an array ref in eximstats generated
2853 parser. This improves both readability and performance.
2855 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2856 Don't display local sender or destination tables in eximstats unless
2857 there is data to show.
2858 Added average volumes into the eximstats top table text output.
2860 SC/03 Collect data on the number of addresses (recipients) as well
2861 as the number of messages in eximstats.
2863 TF/01 Correct an error in the documentation for the redirect router. Exim
2864 does (usually) call initgroups() when daemonizing.
2866 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2867 with consistent privilege compared to when running as a daemon.
2869 TF/03 Note in the spec that $authenticated_id is not set for local
2870 submissions from trusted users.
2872 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2873 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2875 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2876 by adding some example configuration directives to the default
2877 configuration file. A little bit of work is required to uncomment the
2878 directives and define how usernames and passwords are checked, but
2879 there is now a framework to start from.
2881 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2882 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2883 without this. I don't know how relevant this is to other LDAP libraries.
2885 PH/02 Add the verb name to the "unknown ACL verb" error.
2887 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2889 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2891 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2892 directory not expanded when it should be if an expanded home directory
2893 was set for the address (which is overridden by the transport).
2895 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2898 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2899 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2900 because it is too late at that time, and has no effect.
2902 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2903 security issue with \' (bugzilla #107). I could not use the
2904 PQescapeStringConn() function, because it needs a PGconn value as one of
2907 PH/08 When testing addresses using -bt, indicate those final addresses that
2908 are duplicates that would not cause an additional delivery. At least one
2909 person was confused, thinking that -bt output corresponded to deliveries.
2910 (Suppressing duplicates isn't a good idea as you lose the information
2911 about possibly different redirections that led to the duplicates.)
2913 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2914 systems where poll() doesn't work, in particular OS X.
2916 PH/10 Added more information to debugging output for retry time not reached.
2918 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2919 operations in malware.c.
2921 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2924 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2925 syslog_duplication set false, log lines that would normally be written
2926 both the the main log and to the reject log were not written to syslog at
2929 PH/14 In the default configuration, change the use of "message" in ACL warn
2930 statements to "add_header".
2932 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2933 not followed by a command (e.g. "seen endif").
2935 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2936 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2939 PH/17 Added extra conditions to the default value of delay_warning_condition
2943 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2944 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2945 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2948 The Auto-Submitted: and various List- headers are standardised, whereas I
2949 don't think Precedence: ever was.
2951 PH/18 Refactored debugging code in route_finduser() to show more information,
2952 in particular, the error code if getpwnam() issues one.
2954 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2955 This is apparently needed in addition to the PH/07 change above to avoid
2956 any possible encoding problems.
2958 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2959 but not after initializing Perl.
2961 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2962 output them only if debugging. By default they are written stderr,
2963 apparently, which is not desirable.
2965 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2968 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2971 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2973 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2974 authenticated or an ident call has been made. Suppress the default
2975 values for $authenticated_id and $authenticated_sender (but permit -oMai
2976 and -oMas) when testing with -bh.
2978 PH/24 Re-jigged the order of the tests in the default configuration so that the
2979 tests for valid domains and recipients precede the DNS black list and CSA
2980 tests, on the grounds that those ones are more expensive.
2982 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2983 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2984 This bug exists in every version of Exim that I still have, right back to
2987 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2988 However, an attempt to turn on case-sensitivity in a regex key by
2989 including (?-i) didn't work because the subject string was already
2990 lowercased, and the effects were non-intuitive. It turns out that a
2991 one-line patch can be used to allow (?-i) to work as expected.
2997 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2998 other effects) broke the use of negated acl sub-conditions.
3000 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3003 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3004 "Deny" causes Exim to reject the incoming connection with a 554 error.
3005 Unfortunately, if there is a major crisis, such as a disk failure,
3006 tcp-wrappers gives "deny", whereas what one would like would be some
3007 kind of temporary error. A kludge has been added to help with this.
3008 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3009 554 error is used if errno is still zero or contains ENOENT (which occurs
3010 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3013 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3015 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3016 errors. Otherwise a message that provokes a temporary error (when other
3017 messages do not) can cause a whole host to time out.
3019 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3020 addresses were routed directly to files or pipes from a redirect router.
3021 File deliveries just didn't batch; pipe deliveries might have suffered
3024 PH/06 A failure to get a lock for a hints database would erroneously always say
3025 "Failed to get write lock", even when it was really a read lock.
3027 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3028 of 0600. This has been changed to use the value of the lockfile_mode
3029 option (which defaults to 0600).
3031 PH/08 Applied small patch from the Sieve maintainer.
3033 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3034 folder from quota calculations, a direct delivery into this folder messed
3035 up the contents of the maildirsize file. This was because the regex was
3036 used only to exclude .Trash (or whatever) when the size of the mailbox
3037 was calculated. There was no check that a delivery was happening into an
3038 excluded directory. This bug has been fixed by ignoring all quota
3039 processing for deliveries into excluded directories.
3041 PH/10 Added the maildirfolder_create_regex option to appendfile.
3047 PH/01 The code for finding all the local interface addresses on a FreeBSD
3048 system running IPv6 was broken. This may well have applied to all BSD
3049 systems, as well as to others that have similar system calls. The broken
3050 code found IPv4 interfaces correctly, but gave incorrect values for the
3051 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3052 that it would not match correctly against @[] and not recognize the IPv6
3055 PH/02 The ipliteral router was not recognizing addresses of the form user@
3056 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3058 PH/03 Added disable_ipv6.
3060 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3061 decoded form, because it is most often used to construct To: headers
3062 lines in autoreplies, and the decoded form may well be syntactically
3063 invalid. However, $reply_address has leading white space removed, and all
3064 newlines turned into spaces so that the autoreply transport does not
3067 PH/05 If group was specified without a user on a router, and no group or user
3068 was specified on a transport, the group from the router was ignored.
3070 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3071 for visible compile-time settings that can be used to change these
3072 numbers, for those that want even more. Backwards compatibility with old
3073 spool files has been maintained. However, going back to a previous Exim
3074 release will lost any variables that are in spool files.
3076 PH/07 Two small changes when running in the test harness: increase delay when
3077 passing a TCP/IP connection to a new process, in case the original
3078 process has to generate a bounce, and remove special handling of
3079 127.0.0.2 (sic), which is no longer necessary.
3081 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3082 be the same on different OS.
3084 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3087 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3088 whether --show-vars was specified or not
3090 JJ/02 exipick: Added support for new ACL variable spool format introduced
3093 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3094 syntactically invalid From: or Reply-to: line, and a filter used this to
3095 generate an autoreply, and therefore failed to obtain an address for the
3096 autoreply, Exim could try to deliver to a non-existent relative file
3097 name, causing unrelated and misleading errors. What now happens is that
3098 it logs this as a hard delivery error, but does not attempt to create a
3101 PH/11 The exinext utility has a -C option for testing purposes, but although
3102 the given file was scanned by exinext itself; it wasn't being passed on
3103 when Exim was called.
3105 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3106 an end-of-file indication when reading a command response.
3108 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3109 compiled. In many other places in Exim, IPv6 addresses are always
3110 recognized, so I have changed this. It also means that IPv4 domain
3111 literals of the form [IPV4:n.n.n.n] are now always recognized.
3113 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3114 used if the router is not running as root, for example, when verifying at
3115 ACL time, or when using -bh. The debugging output from this situation was
3116 non-existent - all you got was a failure to exec. I have made two
3119 (a) Failures to set uid/gid, the current directory, or a process leader
3120 in a subprocess such as that created by queryprogram now generate
3121 suitable debugging output when -d is set.
3123 (b) The queryprogram router detects when it is not running as root,
3124 outputs suitable debugging information if -d is set, and then runs
3125 the subprocess without attempting to change uid/gid.
3127 PH/15 Minor change to Makefile for building test_host (undocumented testing
3130 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3131 additional section of a DNS packet that returns MX or SRV records.
3132 Instead, it always explicitly searches for A/AAAA records. This avoids
3133 major problems that occur when a DNS server includes only records of one
3134 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3135 fixed another bug: if SRV records were looked up and the corresponding
3136 address records were *not* found in the additional section, the port
3137 values from the SRV records were lost.
3139 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3140 using the correct key (the original address) when searching the retry
3141 rules in order to find which one to use for generating the retry hint.
3143 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3144 adding the default one. Similarly, if it contains a Reply-To: header, the
3145 errors_reply_to option, if set, is not used.
3147 PH/19 When calculating a retry time, Exim used to measure the "time since
3148 failure" by looking at the "first failed" field in the retry record. Now
3149 it does not use this if it is later than than the arrival time of the
3150 message. Instead it uses the arrival time. This makes for better
3151 behaviour in cases where some deliveries succeed, thus re-setting the
3152 "first failed" field. An example is a quota failure for a huge message
3153 when small messages continue to be delivered. Without this change, the
3154 "time since failure" will always be short, possible causing more frequent
3155 delivery attempts for the huge message than are intended.
3156 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3158 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3159 $1, $2, $3) because the numerical variables can be reset during some
3160 expansion items (e.g. "match"), thereby losing the authentication data.
3162 PH/21 Make -bV show the size of off_t variables so that the test suite can
3163 decide whether to run tests for quotas > 2G.
3165 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3166 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3167 filecount value is greater than 2G or if a quota value is greater than 2G
3168 on a system where the size of off_t is not greater than 4, a panic error
3171 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3172 never match. The debug and -bh output now contains an explicit error
3173 message indicating a malformed IPv4 address or mask.
3175 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3176 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3177 PH/23 above applies.
3179 PH/25 Do not write to syslog when running in the test harness. The only
3180 occasion when this arises is a failure to open the main or panic logs
3181 (for which there is an explicit test).
3183 PH/26 Added the /no_tell option to "control=freeze".
3185 PH/27 If a host name lookup failed very early in a connection, for example, if
3186 the IP address matched host_lookup and the reverse lookup yielded a name
3187 that did not have a forward lookup, an error message of the form "no IP
3188 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3189 could be logged. Now it outputs the IP address instead of "NULL".
3191 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3192 allows the sender and the authenticated sender to be set when
3193 submitting a message from within Exim. Since child_open_exim() is
3194 documented for local_scan(), the new function should be too.
3196 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3197 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3198 results in an empty string is now treated as unset.
3200 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3202 PH/31 Added sender_verify_fail logging option.
3204 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3205 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3206 tidied the source and removed it altogether.
3208 PH/33 When a queue run was abandoned because the load average was too high, a
3209 log line was always written; now it is written only if the queue_run log
3210 selector is set. In addition, the log line for abandonment now contains
3211 information about the queue run such as the pid. This is always present
3212 in "start" and "stop" lines but was omitted from the "abandon" line.
3214 PH/34 Omit spaces between a header name and the colon in the error message that
3215 is given when verify = headers_syntax fails (if there are lots of them,
3216 the message gets confusing).
3218 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3219 names, as there are now some PTR records that contain slashes. This check
3220 is only to protect against broken name servers that fall over on strange
3221 characters, so the fact that it applies to all lookups doesn't matter.
3223 PH/36 Now that the new test suite is complete, we can remove some of the
3224 special code in Exim that was needed for the old test suite. For example,
3225 sorting DNS records because real resolvers return them in an arbitrary
3226 order. The new test suite's fake resolver always returns records in the
3229 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3230 bounces) except when queue_only is set, to avoid logging races between
3231 the different processes.
3233 PH/38 Panic-die if .include specifies a non-absolute path.
3235 PH/39 A tweak to the "H" retry rule from its user.
3237 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3238 a label. They prevented compilation on older perls.
3240 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3241 a warning to be raised on newish perls.
3243 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3244 on queue. Changes to match documented behaviour of showing count of
3245 messages matching specified criteria.
3247 PH/40 Changed the default ident timeout from 30s to 5s.
3249 PH/41 Added support for the use of login_cap features, on those BSD systems
3250 that have them, for controlling the resources used by pipe deliveries.
3252 PH/42 The content-scanning code uses fopen() to create files in which to put
3253 message data. Previously it was not paying any attention to the mode of
3254 the files. Exim runs with umask(0) because the rest of the code creates
3255 files with open(), and sets the required mode explicitly. Thus, these
3256 files were ending up world-writeable. This was not a big issue, because,
3257 being within the spool directory, they were not world-accessible. I have
3258 created a function called modefopen, which takes an additional mode
3259 argument. It sets umask(777), creates the file, chmods it to the required
3260 mode, then resets the umask. All the relevant calls to fopen() in the
3261 content scanning code have been changed to use this function.
3263 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3264 to 24 hours. This avoids potential overflow problems when processing G
3265 and H retry rules. I suspect nobody ever tinkers with this value.
3267 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3269 PH/45 When the plaintext authenticator is running as a client, the server's
3270 challenges are checked to ensure they are valid base64 strings. By
3271 default, the authentication attempt is cancelled if an invalid string is
3272 received. Setting client_ignore_invalid_base64 true ignores these errors.
3273 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3274 they are received. Thus, the responses can be made to depend on the
3275 challenges. If an invalid string is ignored, an empty string is placed in
3278 PH/46 Messages that are created by the autoreply transport now contains a
3279 References: header, in accordance with RFCs 2822 and 3834.
3281 PH/47 Added authenticated_sender_force to the smtp transport.
3283 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3285 PH/49 Installed latest patch from the Sieve maintainer.
3287 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3288 was also set, the appendfile transport was unnecessarily scanning a
3289 directory of message files (e.g. for maildir delivery) to find the count
3290 of files (along with the size), even though it did not need this
3291 information. It now does the scan only if it needs to find either the
3292 size of the count of files.
3294 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3296 PH/52 Two bugs concerned with error handling when the smtp transport is
3299 (i) Exim was not creating retry information for temporary errors given
3300 for individual recipients after the DATA command when the smtp transport
3301 was used in LMTP mode. This meant that they could be retried too
3302 frequently, and not timed out correctly.
3304 (ii) Exim was setting the flag that allows error details to be returned
3305 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3306 recipients that were returned after the DATA command.
3308 PH/53 This is related to PH/52, but is more general: for any failing address,
3309 when detailed error information was permitted to be returned to the
3310 sender, but the error was temporary, then after the final timeout, only
3311 "retry timeout exceeded" was returned. Now it returns the full error as
3312 well as "retry timeout exceeded".
3314 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3315 do this, and (what is worse) MTAs that accept it.
3317 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3318 will now be deprecated.
3320 PH/56 New os.c-cygwin from the Cygwin maintainer.
3322 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3323 formats (previously only available in exim formats via -bpr, -bpru,
3324 and -bpra. Now also available in native and exiqgrep formats)
3326 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3327 with very large, slow to parse queues
3329 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3331 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3333 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3334 responses to authentication challenges, though it was showing the
3335 challenges; (ii) I've removed the CR characters from the debug output for
3338 PH/58 Allow for the insertion of a newline as well as a space when a string
3339 is turned into more than one encoded-word during RFC 2047 encoding. The
3340 Sieve code now uses this.
3342 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3343 data_4xx, lost_connection, tls_required.
3345 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3346 message was being sent as an SMTP response.
3348 PH/61 Add -l and -k options to exicyclog.
3350 PH/62 When verifying, if an address was redirected to one new address, so that
3351 verification continued, and the new address failed or deferred after
3352 having set something in $address_data, the value of $address_data was not
3353 passed back to the ACL. This was different to the case when no
3354 redirection occurred. The value is now passed back in both cases.
3356 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3357 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3358 use login_cap.h, so on its own it isn't the distinguishing feature. The
3359 new name refers directly to the setclassresources() function.
3361 PH/65 Added configuration files for NetBSD3.
3363 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3365 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3366 is preferred over IPv4.
3368 PH/68 The bounce_return_message and bounce_return_body options were not being
3369 honoured for bounces generated during the reception of non-SMTP messages.
3370 In particular, this applied to messages rejected by the ACL. This bug has
3371 been fixed. However, if bounce_return_message is true and bounce_return_
3372 body is false, the headers that are returned for a non-SMTP message
3373 include only those that have been read before the error was detected.
3374 (In the case of an ACL rejection, they have all been read.)
3376 PH/69 The HTML version of the specification is now built in a directory called
3377 spec_html instead of spec.html, because the latter looks like a path with
3378 a MIME-type, and this confuses some software.
3380 PH/70 Catch two compiler warnings in sieve.c.
3382 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3383 function verify_get_ident() calls ip_connect() to connect a socket, but
3384 if the "connect()" function timed out, ip_connect() used to close the
3385 socket. However, verify_get_ident() also closes the socket later, and in
3386 between Exim writes to the log, which may get opened at this point. When
3387 the socket was closed in ip_connect(), the log could get the same file
3388 descriptor number as the socket. This naturally causes chaos. The fix is
3389 not to close the socket in ip_connect(); the socket should be closed by
3390 the function that creates it. There was only one place in the code where
3391 this was missing, in the iplookup router, which I don't think anybody now
3392 uses, but I've fixed it anyway.
3394 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3395 well as to direct DNS lookups. Otherwise the handling of names in host
3396 lists is inconsistent and therefore confusing.
3402 PH/01 Two changes to the default runtime configuration:
3404 (1) Move the checks for relay_from_hosts and authenticated clients from
3405 after to before the (commented out) DNS black list checks.
3407 (2) Add control=submission to the relay_from_hosts and authenticated
3408 clients checks, on the grounds that messages accepted by these
3409 statements are most likely to be submissions.
3411 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3413 (1) Generate an error if the third argument for the ${prvs expansion is
3416 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3419 (3) Reset the variables that are obtained from the first argument of
3420 ${prvscheck and used in the second argument before leaving the code,
3421 because their memory is reclaimed, so using them afterwards may do
3424 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3425 one (it's much easier than Tom thought :-).
3427 (5) Because of (4), we can now allow for the use of $prvscheck_result
3428 inside the third argument.
3430 PH/03 For some reason, the default setting of PATH when running a command from
3431 a pipe transport was just "/usr/bin". I have changed it to
3434 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3435 anything to be listed in the output from -bV.
3437 PH/05 When a filter generated an autoreply, the entire To: header line was
3438 quoted in the delivery log line, like this:
3440 => >A.N.Other <ano@some.domain> <original@ddress> ...
3442 This has been changed so that it extracts the operative address. There
3443 may be more than one such address. If so, they are comma-separated, like
3446 => >ano@some.domain,ona@other.domain <original@ddress> ...
3448 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3449 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3450 not being looked up in the rDNS to get a host name, Exim was showing the
3451 IP address twice in Received: lines, even though the IP addresses were
3452 identical. For example:
3454 Received: from [1.2.3.4] (helo=[1.2.3.4])
3456 However, if the real host name was known, it was omitting the HELO data
3457 if it matched the actual IP address. This has been tidied up so that it
3458 doesn't show the same IP address twice.
3460 PH/07 When both +timestamp and +memory debugging was on, the value given by
3461 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3462 called by the debug printing, thereby overwriting the timestamp buffer.
3463 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3465 PH/08 When the original message was included in an autoreply transport, it
3466 always said "this is a copy of the message, including all the headers",
3467 even if body_only or headers_only was set. It now gives an appropriate
3470 PH/09 Applied a patch from the Sieve maintainer which:
3472 o fixes some comments
3473 o adds the (disabled) notify extension core
3474 o adds some debug output for the result of if/elsif tests
3475 o points to the current vacation draft in the documentation
3476 and documents the missing references header update
3480 o fixes a bug in processing the envelope test (when testing
3481 multiple envelope elements, the last element determined the
3484 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3485 Electronic Mail") by including:
3487 Auto-submitted: auto-generated
3489 in the messages that it generates (bounce messages and others, such as
3490 warnings). In the case of bounce messages for non-SMTP messages, there was
3491 also a typo: it was using "Auto_submitted" (underscore instead of
3492 hyphen). Since every message generated by Exim is necessarily in response
3493 to another message, thes have all been changed to:
3495 Auto-Submitted: auto-replied
3497 in accordance with these statements in the RFC:
3499 The auto-replied keyword:
3501 - SHOULD be used on messages sent in direct response to another
3502 message by an automatic process,
3504 - MUST NOT be used on manually-generated messages,
3506 - MAY be used on Delivery Status Notifications (DSNs) and Message
3507 Disposition Notifications (MDNs),
3509 - MUST NOT be used on messages generated by automatic or periodic
3510 processes, except for messages which are automatic responses to
3513 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3514 to the default Received: header definition.
3516 PH/12 Added log selector acl_warn_skipped (default on).
3518 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3519 variables because (a) they are in the wrong storage pool and (b) even if
3520 they were copied, it wouldn't work properly because of the caching.
3522 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3523 checking when decoding. Apparently there are clients that generate
3524 overlong encoded strings. Why am I not surprised?
3526 PH/15 If the first argument of "${if match_address" was not empty, but did not
3527 contain an "@" character, Exim crashed. Now it writes a panic log message
3528 and treats the condition as false.
3530 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3532 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3533 extension "envelope-auth". The code is finished and in agreement with
3534 other implementations, but there is no documentation so far and in fact,
3535 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3536 not changing the active code.
3538 Print executed "if" and "elsif" statements when debugging is used. This
3539 helps a great deal to understand what a filter does.
3541 Document more things not specified clearly in RFC3028. I had all this
3542 sorted out, when out of a sudden new issues came to my mind. Oops."
3544 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3547 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3548 canonical form (as documented). However, after a host name lookup from
3549 the IP address, check_host() was doing a simple string comparison with
3550 addresses acquired from the DNS when checking that the found name did
3551 have the original IP as one of its addresses. Since any found IPv6
3552 addresses are likely to be in abbreviated form, the comparison could
3553 fail. Luckily, there already exists a function for doing the comparison
3554 by converting both addresses to binary, so now that is used instead of
3555 the text comparison.
3557 PH/20 There was another similar case to PH/19, when a complete host name was
3558 given in a host list; looking up its IP address could give an abbreviated
3559 form, whereas the current host's name might or might not be abbreviated.
3560 The same fix has been applied.
3566 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3567 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3570 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3571 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3573 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3575 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3576 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3577 Also, for "Auto-Submitted", treat anything other than "no" as
3578 non-personal, in accordance with RFC 3834. (Previously it treated
3579 anything starting "auto-" as non-personal.)
3581 TF/01 The control=submission/name=... option had a problem with syntax
3582 errors if the name included a slash character. The /name= option
3583 now slurps the rest of the string, so it can include any characters
3584 but it must come last in the list of options (after /sender_retain
3587 PH/05 Some modifications to the interface to the fake nameserver for the new
3595 TK/01 Added the "success_on_redirect" address verification option. See
3596 NewStuff for rationale and an example.
3598 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3600 PH/02 Patch to exigrep to allow it to work on syslog lines.
3602 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3603 fread() to skip over the body file's header line, because in Cygwin the
3604 header line is locked and is inaccessible.
3606 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3607 co-exist for some time) to make it clear that it is the Exim ID that is
3608 referenced, not the Message-ID: header line.
3610 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3611 string_format() function, because snprintf() does not exist on all
3614 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3615 ${stat: expansion item.
3617 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3618 protocol synchronization error", to keep the pedants happy.
3620 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3621 well as for IRIX systems, when gcc is being used. See the host.c source
3624 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3626 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3629 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3630 transport and to the smtp transport in LMTP mode.
3632 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3634 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3635 run a filter in a subprocess. This could lead to confusion in subsequent
3636 lookups in the parent process. There should also be a search_tidyup() at
3637 the end of the subprocess.
3639 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3640 only if the host matched helo_try_verify_hosts, which caused the
3641 verification to occur when the EHLO/HELO command was issued. The ACL just
3642 tested the remembered result. Now, if a previous verification attempt has
3643 not happened, "verify = helo" does it there and then.
3645 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3647 TK/03 Fix log output including CR from clamd.
3649 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3650 exist provoked a memory error which could cause a segfault.
3652 PH/15 Installed PCRE 6.2
3654 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3656 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3657 of the problem. Specifically, suggested +O2 rather than +O1 for the
3660 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3662 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3663 record, the port was not correctly logged when the outgoing_port log
3664 selector was set (it logged the transort's default port).
3666 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3667 fallback_hosts, and "hosts" in the smtp transport.
3669 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3670 host errors such as "Connection refused".
3672 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3673 authentication with radiusclient 0.4.9:
3675 - Error returned from rc_read_config was caught wrongly
3676 - Username/password not passed on to radius server due to wrong length.
3678 The presumption is that some radiusclient API changes for 4.51/PH/17
3679 were not taken care of correctly. The code is still untested by me (my
3680 Linux distribution still has 0.3.2 of radiusclient), but it was
3681 contributed by a Radius user.
3683 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3684 expanding the "port" option of the smtp transport.
3686 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3687 while reading a MIME header. Thanks to Tom Hughes for a patch.
3689 PH/24 Include config.h inside local_scan.h so that configuration settings are
3692 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3693 that in an ACL for RCPT (for example), you can examine exactly what was
3696 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3697 commands, but it was not correctly comparing the address with the actual
3698 client host address. Thus, it would show the EHLO address in Received:
3699 header lines when this was not necessary.
3701 PH/27 Added the % operator to ${eval:}.
3703 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3704 it should be ignoring failures (because with -C, for example, it has lost
3705 privilege). It wasn't ignoring creation failures other than "already
3708 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3711 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3712 because an input line was too long, either on its own, or by virtue of
3713 too many continuations, the temporary file was not being removed, and the
3714 return code was incorrect.
3716 PH/31 Missing "BOOL" in function definition in filtertest.c.
3718 PH/32 Applied Sieve patches from the maintainer.
3720 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3722 PH/33 Added "verify = not_blind".
3724 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3725 Local/Makefile (with some defaults set). These are used in built scripts
3726 such as exicyclog, but they have never been used in the exim_install
3727 script (though there are many overriding facilities there). I have
3728 arranged that the exim_install script now takes note of these two
3731 PH/35 Installed configuration files for Dragonfly.
3733 PH/36 When a locally submitted message by a trusted user did not contain a
3734 From: header, and the sender address was obtained from -f or from an SMTP
3735 MAIL command, and the trusted user did not use -F to supply a sender
3736 name, $originator_name was incorrectly used when constructing a From:
3737 header. Furthermore, $originator_name was used for submission mode
3738 messages from external hosts without From: headers in a similar way,
3739 which is clearly wrong.
3741 PH/37 Added control=suppress_local_fixups.
3743 PH/38 When log_selector = +received_sender was set, and the addition of the
3744 sender made the log line's construction buffer exactly full, or one byte
3745 less than full, an overflow happened when the terminating "\n" was
3748 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3749 when the result of a list match is failure because a DNS lookup failed.
3751 PH/40 RM_COMMAND is now used in the building process.
3753 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3754 the "build-* directories that it finds.
3756 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3757 address in a domain literal was a prefix of an interface address.
3759 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3760 when verifying a sender address, unless rewrite_headers is false.
3762 PH/44 Wrote a long comment about why errors_to addresses are verified as
3763 recipients, not senders.
3765 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3766 the ratelimit ACL was added.
3768 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3770 PH/46 Added extra information about PostgreSQL errors to the error string.
3772 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3773 avoiding the need to install special zones in a real server. This is
3774 backwards compatible; if it can't find the fake resolver, it drops back.
3775 Thus, both old and new test suites can be run.
3777 TF/02 Added util/ratelimit.pl
3779 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3780 clock is set back in time.
3782 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3783 Candler <B.Candler@pobox.com>.
3785 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3786 OK for addresses that are the result of redirections.
3788 PH/48 A number of further additions for the benefit of the new test suite,
3789 including a fake gethostbyname() that interfaces to the fake DNS resolver
3792 TF/06 The fix for widen_domains has also been applied to qualify_single and
3793 search_parents which are the other dnslookup options that can cause
3796 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3799 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3801 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3802 service" if it encountered an MX record with an empty target hostname.
3803 The message is now "an MX or SRV record indicated no SMTP service".
3805 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3806 if the DNS of the sending site is misconfigured. This is quite a
3807 common situation. This change restores the behaviour of treating a
3808 helo verification defer as a failure.
3810 PH/51 If self=fail was set on a router, the bounce message did not include the
3811 actual error message.
3817 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3819 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3820 command itself ended in error, the underlying message about the transport
3821 filter timeout was being overwritten with the pipe command error. Now the
3822 underlying error message should be appended to the second error message.
3824 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3826 PH/02 Reduce the amount of output that "make" produces by default. Full output
3827 can still be requested.
3829 PH/03 The warning log line about a condition test deferring for a "warn" verb
3830 was being output only once per connection, rather than after each
3831 occurrence (because it was using the same function as for successful
3832 "warn" verbs). This seems wrong, so I have changed it.
3834 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3835 it should not have, which might have caused a crash in the right
3836 circumstances, but probably never did.
3838 PH/04 Installed a modified version of Tony Finch's patch to make submission
3839 mode fix the return path as well as the Sender: header line, and to
3840 add a /name= option so that you can make the user's friendly name appear
3843 TF/03 Added the control = fakedefer ACL modifier.
3845 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3846 Mark Lowes for thorough testing.
3848 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3850 TK/03 Merged latest SRS patch from Miles Wilton.
3852 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3853 with the definition in sysexits.h (which is #included earlier).
3854 Fortunately, Exim does not actually use EX_OK. The code used to try to
3855 preserve the sysexits.h value, by assuming that macro definitions were
3856 scanned for macro replacements. I have been disabused of this notion,
3857 so now the code just undefines EX_OK before #including unistd.h.
3859 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3860 in the smtp transport. When a block could not be written in a single
3861 write() function, the timeout was being re-applied to each part-write.
3862 This seems wrong - if the receiver was accepting one byte at a time it
3863 would take for ever. The timeout is now adjusted when this happens. It
3864 doesn't have to be particularly precise.
3866 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3867 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3869 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3870 without a domain if the check to <postmaster@domain> fails.
3872 SC/01 Eximstats: added -xls and the ability to specify output files
3873 (patch written by Frank Heydlauf).
3875 SC/02 Eximstats: use FileHandles for outputting results.
3877 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3879 SC/04 Eximstats: fixed display of large numbers with -nvr option
3881 SC/05 Eximstats: fixed merging of reports with empty tables.
3883 SC/06 Eximstats: added the -include_original_destination flag
3885 SC/07 Eximstats: removed tabs and trailing whitespace.
3887 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3889 TK/06 MBOX spool code: Add real "From " MBOX separator line
3890 so the .eml file is really in mbox format (even though
3891 most programs do not really care). Patch from Alex Miller.
3893 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3894 The latter is generated from $received_to and is only set if the
3895 message has one envelope recipient. SA can use these headers,
3896 obviously out-of-the-box. Patch from Alex Miller.
3898 PH/08 The ${def test on a variable was returning false if the variable's
3899 value was "0", contrary to what the specification has always said!
3900 The result should be true unless the variable is empty.
3902 PH/09 The syntax error of a character other than { following "${if
3903 def:variable_name" (after optional whitespace) was not being diagnosed.
3904 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3905 accidental colon was present, for example, could give incorrect results.
3907 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3908 result is used (not including appendfile, where other changes are about
3911 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3912 This involved changing a lot of size variables from int to off_t. It
3913 should work with maildirs and everything.
3915 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3916 spamd dying while we are connected to it.
3918 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3921 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3922 function for BDB 4.3.
3924 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3926 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3927 the sources for PCRE have been re-arranged and more files are now
3930 PH/15 The code I had for printing potentially long long variables in PH/11
3931 above was not the best (it lost precision). The length of off_t variables
3932 is now inspected at build time, and an appropriate printing format (%ld
3933 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3934 to be "long long int" or "long int". This is needed for the internal
3935 formatting function string_vformat().
3937 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3938 the configuration file to be ":syslog", then the script "guesses" where
3939 the logs files are, rather than using the compiled in default. In our
3940 case the guess is not the same as the compiled default, so the script
3941 suddenly stopped working when I started to use syslog. The patch checks
3942 to see if log_file_path is "". If so, it attempts to read it from exim
3943 with no configuration file to get the compiled in version, before it
3944 falls back to the previous guessing code."
3946 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3947 implementing BATV in an Exim configuration. See NewStuff for the gory
3950 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3951 Makefile that are specific to HP-UX.
3953 PH/18 If the "use_postmaster" option was set for a recipient callout together
3954 with the "random" option, the postmaster address was used as the MAIL
3955 FROM address for the random test, but not for the subsequent recipient
3956 test. It is now used for both.
3958 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3959 patch removes a few documentation additions to RFC 3028, because the
3960 latest draft now contains them. It adds the new en;ascii-case comparator
3961 and a new error check for 8bit text in MIME parts. Comparator and
3962 require names are now matched exactly. I enabled the subaddress
3963 extension, but it is not well tested yet (read: it works for me)."
3965 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3966 rework some of the code of TK/09 above to avoid the hardwired use of
3967 "%lld" and "long long". Replaced the call to snprintf() with a call to
3970 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3971 records point to non-existent hosts", "retry timeout exceeded", and
3972 "retry time not reached for any host after a long failure period".
3974 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3975 experimental DomainKeys support:
3977 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3978 (2) On an error such as an illegally used "control", the wrong name for
3979 the control was given.
3981 These problems did NOT occur unless DomainKeys support was compiled.
3983 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3985 PH/24 Added ${if match_ip condition.
3987 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3988 databases so that it will be absolutely obvious if a crash occurs in the
3989 DB library. This is a regular occurrence (often caused by mis-matched
3992 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3993 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3994 picked up on a user's system that detects such things. There doesn't seem
3995 to be a gcc warning option for this - only an attribute that has to be
3996 put on the function's prototype. It seems that in Fedora Core 4 they have
3997 set this on a number of new functions. No doubt there will be more in due
4000 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4001 specify a transport. However, if an address that was verified by such a
4002 router was the subject of a callout, Exim crashed because it tried to
4003 read the rcpt_include_affixes from the non-existent transport. Now it
4004 just assumes that the setting of that option is false. This bug was
4005 introduced by 4.51/PH/31.
4007 PH/28 Changed -d+all to exclude +memory, because that information is very
4008 rarely of interest, but it makes the output a lot bigger. People tend to
4009 do -d+all out of habit.
4011 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4012 code in os-type was giving problems when libc.so lives in lib64, like on
4015 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4016 aren't the modern standard, and it seems that some systems' include files
4017 don't always have them. Exim was already checking for some of the newer
4018 ones like T_AAAA, and defining it itself. I've added checks for all the
4019 record types that Exim uses.
4021 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4022 not automatically generating a new one, as it is supposed to. This
4023 prevented TLS from working. If the file did exist, but contained invalid
4024 data, a new version was generated, as expected. It was only the case of a
4025 non-existent file that was broken.
4027 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4028 with a change in libdomainkeys > 0.64.
4030 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4031 from DNS. If the selector record carries the flag, it now has
4032 precedence over the domain-wide flag.
4034 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4036 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4037 the use of a port name that isn't defined in /etc/services) occurred, the
4038 message was deferred as in a normal delivery, and thus remained on the
4039 spool, instead of being failed because of the mua_wrapper setting. This
4040 is now fixed, and I tidied up some of the mua_wrapper messages at the
4043 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4044 the messages in a hash of arrays rather than using individual hashes.
4045 This is a bit cleaner and results in dramatic memory savings, albeit
4046 at a slight CPU cost.
4048 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4049 as requested by Marc Sherman.
4051 SC/10 Eximstats: added histograms for user specified patterns as requested
4054 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4056 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4057 fopen() in the content-scanning modules that did not already have it.
4063 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4064 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4066 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4068 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4070 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4071 to test Sieve filters that use "vacation".
4073 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4074 that changes the way the GnuTLS parameters are stored in the cache file.
4075 The new format can be generated externally. For backward compatibility,
4076 if the data in the cache doesn't make sense, Exim assumes it has read an
4077 old-format file, and it generates new data and writes a new file. This
4078 means that you can't go back to an older release without removing the
4081 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4082 work if there are any delivery delays because "one_time" forces the
4083 parent to be marked "delivered", so its unseen clone is never tried
4084 again. For this reason, Exim now forbids the simultaneous setting of
4087 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4088 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4089 entry if you want to know the details. The fix, however, neglected to
4090 consider the case when local delivery batching is involved. The test for
4091 "previously delivered" was not happening when checking to see if an
4092 address could be batched with a previous (undelivered) one; under
4093 certain circumstances this could lead to multiple deliveries to the same
4096 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4097 in its include files, and this causes problems building Exim.
4099 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4100 header_syntax) but Exim was just ignoring anything given after a slash.
4101 In particular, this caused confusion with an attempt to use "verify =
4102 reverse_host_lookup/defer_ok". An error is now given when options are
4103 supplied for verify items that do not have them. (Maybe reverse_host_
4104 lookup should have a defer_ok option, but that's a different point.)
4106 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4107 defined by RFC 821) to 2048, because there were problems with some AUTH
4108 commands, and RFC 1869 says the size should be increased for extended
4109 SMTP commands that take arguments.
4111 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4114 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4115 "unknown" error; now it says that the functionality isn't in the binary.
4117 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4118 an address' error message when a string expansion fails (syntax or
4119 whatever). Otherwise the password may appear in the log. Following change
4120 PH/42 below, there is no longer a chance of it appearing in a bounce
4123 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4125 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4126 of its addresses was ever tried. (Bugzilla bug #2.)
4128 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4129 the result incorrectly in the debug output. (It correctly added a newline
4130 to what was transported.)
4132 TF/01 Added $received_time.
4134 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4135 commented out examples of how to interface to a virus scanner and to
4136 SpamAssassin. Also added commented examples of av_scanner and
4137 spamd_address settings.
4139 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4140 and controls are allowed in which ACLs. There were a couple of minor
4141 errors. Some of the entries in the conditions table (which is a table of
4142 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4143 negation of where the condition IS allowed.
4145 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4147 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4148 header file does not have a version number, so I've had to invent a new
4149 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4150 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4151 radiusclient), but it was contributed by a Radius user.
4153 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4154 files or directories, for OpenSSL.
4156 PH/19 When an Exim process that is running as root has to create an Exim log
4157 file, it does so in a subprocess that runs as exim:exim so as to get the
4158 ownership right at creation (otherwise, other Exim processes might see
4159 the file with the wrong ownership). There was no test for failure of this
4160 fork() call, which would lead to the process getting stuck as it waited
4161 for a non-existent subprocess. Forks do occasionally fail when resources
4162 run out. I reviewed all the other calls to fork(); they all seem to check
4165 PH/20 When checking for unexpected SMTP input at connect time (before writing
4166 the banner), Exim was not dealing correctly with a non-positive return
4167 from the read() function. If the client had disconnected by this time,
4168 the result was a log entry for a synchronization error with an empty
4169 string after "input=" when read() returned zero. If read() returned -1
4170 (an event I could not check), uninitialized data bytes were printed.
4171 There were reports of junk text (parts of files, etc) appearing after
4174 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4176 PH/22 Added support for macro redefinition, and (re)definition in between
4177 driver and ACL definitions.
4179 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4180 forgetting to use the resulting value; it was using the unexpanded value.
4182 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4183 hadn't been configured. The fix is from Juergen Kreileder, who
4184 understands it better than I do:
4186 "Here's what I see happening with three configured cyrus_sasl
4187 authenticators configured (plain, login, cram-md5):
4189 On startup auth_cyrus_sasl_init() gets called for each of these.
4190 This means three calls to sasl_listmech() without a specified mech_list.
4191 => SASL tests which mechs of all available mechs actually work
4192 => three warnings about OTP not working
4193 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4195 With the patch, sasl_listmech() also gets called three times. But now
4196 SASL's mech_list option is set to the server_mech specified in the the
4197 authenticator. Or in other words, the answer from sasl_listmech()
4198 gets limited to just the mech you're testing for (which is different
4200 => the return list contains just 'plain' or 'login', 'cram-md5' or
4201 nothing depending on the value of ob->server_mech.
4203 I've just tested the patch: Authentication still works fine,
4204 unavailable mechs specified in the exim configuration are still
4205 caught, and the auth.log warnings about OTP are gone."
4207 PH/25 When debugging is enabled, the contents of the command line are added
4208 to the debugging output, even when log_selector=+arguments is not
4211 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4212 answer is "GNU", and only if the return is "GNU/something" is the answer
4215 PH/27 $acl_verify_message is now set immediately after the failure of a
4216 verification in an ACL, and so is available in subsequent modifiers. In
4217 particular, the message can be preserved by coding like this:
4219 warn !verify = sender
4220 set acl_m0 = $acl_verify_message
4222 Previously, $acl_verify_message was set only while expanding "message"
4223 and "log_message" when a very denied access.
4225 PH/28 Modified OS/os.c-Linux with
4227 -#ifndef OS_LOAD_AVERAGE
4228 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4230 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4231 nomenclature these days.)
4233 PH/29 Installed patch from the Sieve maintainer that adds the options
4234 sieve_useraddress and sieve_subaddress to the redirect router.
4236 PH/30 In these circumstances:
4237 . Two addresses routed to the same list of hosts;
4238 . First host does not offer TLS;
4239 . First host accepts first address;
4240 . First host gives temporary error to second address;
4241 . Second host offers TLS and a TLS session is established;
4242 . Second host accepts second address.
4243 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4244 and peerdn, if requested) that were in fact used only for the second
4247 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4248 attention to any local part prefix or suffix that was matched by the
4249 router that accepted the address. It now behaves in the same way as it
4250 does for delivery: the affixes are removed from the local part unless
4251 rcpt_include_affixes is set on the transport.
4253 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4254 timeout during the DATA phase of an incoming message.
4256 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4257 applied a patch sanctioned by the Sieve maintainer.
4259 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4260 the uid or gid is negative. A case of a negative gid caused this to be
4261 noticed. The fix allows for either to be negative.
4263 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4264 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4267 PH/36 The change PH/12 above was broken. Fixed it.
4269 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4270 the grounds that routing the same address twice would always produce the
4271 same answer. This might have been true once, but it is certainly no
4272 longer true now. Routing a child address may depend on the previous
4273 routing that produced that child. Some complicated redirection strategies
4274 went wrong when messages had multiple recipients, and made Exim's
4275 behaviour dependent on the order in which the addresses were given.
4277 I have moved the duplicate checking until after the routing is complete.
4278 Exim scans the addresses that are assigned to local and remote
4279 transports, and removes any duplicates. This means that more work will be
4280 done, as duplicates will always all be routed, but duplicates are
4281 presumably rare, so I don't expect this is of any significance.
4283 For deliveries to pipes, files, and autoreplies, the duplicate checking
4284 still happens during the routing process, since they are not going to be
4287 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4288 It corrects a timeout issue with spamd. This is Ian's comment: "The
4289 background is that sometimes spamd either never reads data from a
4290 connection it has accepted, or it never writes response data. The exiscan
4291 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4292 blindly assumes that writes won't block so it may never time out."
4294 PH/39 Allow G after quota size as well as K and M.
4296 PH/40 The value set for $authenticated_id in an authenticator may not contain
4297 binary zeroes or newlines because the value is written to log lines and
4298 to spool files. There was no check on this. Now the value is run through
4299 the string_printing() function so that such characters are converted to
4300 printable escape sequences.
4302 PH/41 $message_linecount is a new variable that contains the total number of
4303 lines in the message. Compare $body_linecount, which is the count for the
4306 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4307 bounce and delay warning messages, except in certain special cases, which
4310 (a) An SMTP error message from a remote host;
4311 (b) A message specified in a :fail: redirection;
4312 (c) A message specified in a "fail" command in a system filter;
4313 (d) A message specified in a FAIL return from the queryprogram router;
4314 (e) A message specified by the cannot_route_message router option.
4316 In these cases only, Exim does include the error details in bounce and
4317 warning messages. There are also a few cases where bland messages such
4318 as "unrouteable address" or "local delivery error" are given.
4320 PH/43 $value is now also set for the "else" part of a ${run expansion.
4322 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4323 being worked on, but at least Exim now implements the latest version to
4326 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4327 process to complete was treated as a delivery failure, a timeout while
4328 writing the message to the pipe was logged, but erroneously treated as a
4329 successful delivery. Such timeouts include transport filter timeouts. For
4330 consistency with the overall process timeout, these timeouts are now
4331 treated as errors, giving rise to delivery failures by default. However,
4332 there is now a new Boolean option for the pipe transport called
4333 timeout_defer, which, if set TRUE, converts the failures into defers for
4334 both kinds of timeout. A transport filter timeout is now identified in
4337 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4338 systems where "make" and "gmake" are different, calling "gmake" at top
4339 level broke things. I've arranged for the value of $(MAKE) to be passed
4340 from the Makefile to this script so that it can call the same version of
4344 A note about Exim versions 4.44 and 4.50
4345 ----------------------------------------
4347 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4348 changes of various kinds. As a consequence, a big documentation update was
4349 needed. This delayed the release for rather longer than seemed good, especially
4350 in the light of a couple of (minor) security issues. Therefore, the changes
4351 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4352 release. So 4.44 and 4.50 are in effect two different branches that both start
4355 I have left the 4.50 change log unchanged; it contains all the changes since
4356 4.43. The change log for 4.44 is below; many of its items are identical to
4357 those for 4.50. This seems to be the most sensible way to preserve the
4358 historical information.
4364 1. Minor wording change to the doc/README.SIEVE file.
4366 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4367 computation of the current number of files was incorrect.
4369 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4370 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4373 4. Give more explanation in the error message when the command for a transport
4374 filter fails to execute.
4376 5. There are several places where Exim runs a non-Exim command in a
4377 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4378 was being done only for the command run by the queryprogram router. It is
4379 now done for all such subprocesses. The other cases are: ${run, transport
4380 filters, and the commands run by the lmtp and pipe transports.
4382 6. Added CONFIGURE_GROUP build-time option.
4384 7. Some older OS have a limit of 256 on the maximum number of file
4385 descriptors. Exim was using setrlimit() to set 1000 as a large value
4386 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4387 systems. I've change it so that if it can't get 1000, it tries for 256.
4389 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4390 was an oversight, and furthermore, ever since the addition of extra
4391 controls (e.g. 4.43/32), the checks on when to allow different forms of
4392 "control" were broken. There should now be diagnostics for all cases when a
4393 control that does not make sense is encountered.
4395 9. Added the /retain_sender option to "control=submission".
4397 10. $recipients is now available in the predata ACL (oversight).
4399 11. Tidy the search cache before the fork to do a delivery from a message
4400 received from the command line. Otherwise the child will trigger a lookup
4401 failure and thereby defer the delivery if it tries to use (for example) a
4402 cached ldap connection that the parent has called unbind on.
4404 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4405 of $address_data from the recipient verification was clobbered by the
4406 sender verification.
4408 13. The value of address_data from a sender verification is now available in
4409 $sender_address_data in subsequent conditions in the ACL statement.
4411 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4413 15. Added a new option "connect=<time>" to callout options, to set a different
4416 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4417 was its contents. (It was OK if the option was not defined at all.)
4419 17. A "Completed" log line is now written for messages that are removed from
4420 the spool by the -Mrm option.
4422 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4423 information about exactly what failed.
4425 19. Added -dd to debug only the daemon process.
4427 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4428 handles timeouts, both on the server side and network timeouts. Renamed the
4429 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4431 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4432 It is now set to "smtps".
4434 22. $host_address is now set to the target address during the checking of
4435 ignore_target_hosts.
4437 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4438 being passed; this would have caused $sender_host_name to have been used if
4439 matching the list had actually called for a host name (not very likely,
4440 since this list is usually IP addresses). A host name is now passed as
4443 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4444 code that specifies a non-restarting handler (typically sigaction() in
4445 modern systems) in an attempt to fix a rare and obscure crash bug.
4447 25. Narrowed the window for a race in the daemon that could cause it to ignore
4448 SIGCHLD signals. This is not a major problem, because they are used only to
4449 wake it up if nothing else does.
4451 26. A malformed maildirsize file could cause Exim to calculate negative values
4452 for the mailbox size or file count. Odd effects could occur as a result.
4453 The maildirsize information is now recalculated if the size or filecount
4456 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4457 support for a long time. Removed HAVE_SYS_VFS_H.
4459 28. Installed the latest version of exipick from John Jetmore.
4461 29. In an address list, if the pattern was not a regular expression, an empty
4462 subject address (from a bounce message) matched only if the pattern was an
4463 empty string. Non-empty patterns were not even tested. This was the wrong
4464 because it is perfectly reasonable to use an empty address as part of a
4465 database query. An empty address is now tested by patterns that are
4466 lookups. However, all the other forms of pattern expect the subject to
4467 contain a local part and a domain, and therefore, for them, an empty
4468 address still always fails if the pattern is not itself empty.
4470 30. Exim went into a mad DNS loop when attempting to do a callout where the
4471 host was specified on an smtp transport, and looking it up yielded more
4472 than one IP address.
4474 31. Re-factored the code for checking spool and log partition space into a
4475 function that finds that data and another that does the check. The former
4476 is then used to implement four new variables: $spool_space, $log_space,
4477 $spool_inodes, and $log_inodes.
4479 32. The RFC2047 encoding function was originally intended for short strings
4480 such as real names; it was not keeping to the 75-character limit for
4481 encoded words that the RFC imposes. It now respects the limit, and
4482 generates multiple encoded words if necessary. To be on the safe side, I
4483 have increased the buffer size for the ${rfc2047: expansion operator from
4486 33. It is now permitted to omit both strings after an "if" condition; if the
4487 condition is true, the result is "true". As before, when the second string
4488 is omitted, a false condition yields an empty string. This makes it less
4489 cumbersome to write custom ACL and router conditions.
4491 34. Failure to deliver a bounce message always caused it to be frozen, even if
4492 there was an errors_to setting on the router. The errors_to setting is now
4495 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4496 canonical form (fully expanded) before being placed in
4497 $sender_host_address.
4499 36. The table in the code that translates DNS record types into text (T_A to
4500 "A" for instance) was missing entries for NS and CNAME. It is just possible
4501 that this could have caused confusion if both these types were looked up
4502 for the same domain, because the text type is used as part of Exim's
4503 per-process caching. But the chance of anyone hitting this buglet seems
4506 37. The dnsdb lookup has been extended in a number of ways.
4508 (1) There is a new type, "zns", which walks up the domain tree until it
4509 finds some nameserver records. It should be used with care.
4511 (2) There is a new type, "mxh", which is like "mx" except that it returns
4512 just the host names, not the priorities.
4514 (3) It is now possible to give a list of domains (or IP addresses) to be
4515 looked up. The behaviour when one of the lookups defers can be
4516 controlled by a keyword.
4518 (4) It is now possible to specify the separator character for use when
4519 multiple records are returned.
4521 38. The dnslists ACL condition has been extended: it is now possible to supply
4522 a list of IP addresses and/or domains to be looked up in a particular DNS
4525 39. Added log_selector=+queue_time_overall.
4527 40. When running the queue in the test harness, wait just a tad after forking a
4528 delivery process, to get repeatability of debugging output.
4530 41. Include certificate and key file names in error message when GnuTLS fails
4531 to set them up, because the GnuTLS error message doesn't include the name
4532 of the failing file when there is a problem reading it.
4534 42. Allow both -bf and -bF in the same test run.
4536 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4538 44. The "Exiscan patch" is now merged into the mainline Exim source.
4540 45. Sometimes the final signoff response after QUIT could fail to get
4541 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4542 < 0 before doing a fflush(). This bug looks as though it goes back to the
4543 introduction of TLS in release 3.20, but "sometimes" must have been rare
4544 because the tests only now provoked it.
4546 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4547 (this can affect the format of dates).
4549 47. exim_tidydb, when checking for the continued existence of a message for
4550 which it has found a message-specific retry record, was not finding
4551 messages that were in split spool directories. Consequently, it was
4552 deleting retry records that should have stayed in existence.
4554 48. Steve fixed some bugs in eximstats.
4556 49. The SPA authentication driver was not abandoning authentication and moving
4557 on to the next authenticator when an expansion was forced to fail,
4558 contradicting the general specification for all authenticators. Instead it
4559 was generating a temporary error. It now behaves as specified.
4561 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4562 (the order specifies the preference for clients). The order is now AES256,
4563 AES128, 3DES, ARCFOUR128.
4565 51. Small patch to Sieve code - explicitly set From: when generating an
4568 52. Exim crashed if a remote delivery caused a very long error message to be
4569 recorded - for instance if somebody sent an entire SpamAssassin report back
4570 as a large number of 550 error lines. This bug was coincidentally fixed by
4571 increasing the size of one of Exim's internal buffers (big_buffer) that
4572 happened as part of the Exiscan merge. However, to be on the safe side, I
4573 have made the code more robust (and fixed the comments that describe what
4576 53. Now that there can be additional text after "Completed" in log lines (if
4577 the queue_time_overall log selector is set), a one-byte patch to exigrep
4578 was needed to allow it to recognize "Completed" as not the last thing in
4581 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4582 patch that reportedly fixes this has been added. I am not expert enough to
4583 create a test for it. This is what the patch creator wrote:
4585 "I found a little strange behaviour of ldap code when working with
4586 Windows 2003 AD Domain, where users was placed in more than one
4587 Organization Units. When I tried to give exim partial DN, the exit code
4588 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4589 But simultaneously result of request was absolutely normal ldap result,
4590 so I produce this patch..."
4592 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4593 REFERENCE, so I have modified the code to exclude the patch when that macro
4596 55. Some experimental protocols are using DNS PTR records for new purposes. The
4597 keys for these records are domain names, not reversed IP addresses. The
4598 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4599 leaves it alone. Component reversal etc. now happens only for IP addresses.
4602 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4604 57. Double the size of the debug message buffer (to 2048) so that more of very
4605 long debug lines gets shown.
4607 58. The exicyclog utility now does better if the number of log files to keep
4608 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4610 59. Two changes related to the smtp_active_hostname option:
4612 (1) $smtp_active_hostname is now available as a variable.
4613 (2) The default for smtp_banner uses $smtp_active_hostname instead
4614 of $primary_hostname.
4616 60. The host_aton() function is supposed to be passed a string that is known
4617 to be a valid IP address. However, in the case of IPv6 addresses, it was
4618 not checking this. This is a hostage to fortune. Exim now panics and dies
4619 if the condition is not met. A case was found where this could be provoked
4620 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4621 components; fortuitously, this particular loophole had already been fixed
4622 by change 4.50/55 above.
4624 If there are any other similar loopholes, the new check in host_aton()
4625 itself should stop them being exploited. The report I received stated that
4626 data on the command line could provoke the exploit when Exim was running as
4627 exim, but did not say which command line option was involved. All I could
4628 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4629 running as the user.
4632 61. There was a buffer overflow vulnerability in the SPA authentication code
4633 (which came originally from the Samba project). I have added a test to the
4634 spa_base64_to_bits() function which I hope fixes it.
4637 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4638 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4640 63. The daemon start-up calls getloadavg() while still root for those OS that
4641 need the first call to be done as root, but it missed one case: when
4642 deliver_queue_load_max is set with deliver_drop_privilege. This is
4643 necessary for the benefit of the queue runner, because there is no re-exec
4644 when deliver_drop_privilege is set.
4646 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4647 This has been fixed.
4649 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4650 was in use, was not putting the data itself into the right store pool;
4651 consequently, it could be overwritten for a subsequent message in the same
4652 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4655 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4657 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4658 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4659 the function were treating the return as a boolean value, which happened to
4660 work because 0=false and not-0=true, but is not correct code.
4662 68. The host_aton() function was not handling scoped IPv6 addresses (those
4663 with, for example, "%eth0" on the end) correctly.
4665 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4666 negated items (that is, ~something) in unsigned ints. Some compilers
4667 apparently mutter when there is no cast.
4669 70. If an address verification called from an ACL failed, and did not produce a
4670 user-specific message (i.e. there was only a "system" message), nothing was
4671 put in $acl_verify_message. In this situation, it now puts the system
4674 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4675 session; change 4.31/43 added the unwanted input to the log line - except
4676 that it did not do this in the start of session case. It now does.
4678 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4679 This is wrong and can cause the other end to generate a synchronization
4680 error if it is another Exim or anything else that does the synchronization
4681 check. A QUIT command is no longer sent after a timeout.
4683 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4684 during host lookups.
4686 74. The defer_ok option of callout verification was not working if it was used
4687 when verifying addresses in header lines, that is, for this case:
4689 verify = header_sender/callout=defer_ok
4691 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4692 those file descriptors could be used for SMTP connections. If anything
4693 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4694 could be sent to the SMTP client, causing chaos. The daemon now opens
4695 stdin, stdout, and stderr to /dev/null when it puts itself into the
4698 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4699 log by default. The user can override this with suitable Perl magic.
4701 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4702 the log message when discard triggers, was not working for the DATA ACL or
4703 for the non-SMTP ACL.
4705 78. Error message wording change in sieve.c.
4707 79. If smtp_accept_max_per_host was set, the number of connections could be
4708 restricted to fewer than expected, because the daemon was trying to set up
4709 a new connection before checking whether the processes handling previous
4710 connections had finished. The check for completed processes is now done
4711 earlier. On busy systems, this bug wouldn't be noticed because something
4712 else would have woken the daemon, and it would have reaped the completed
4715 80. If a message was submitted locally by a user whose login name contained one
4716 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4717 It caused a spool format error. I have fixed the spool reading code. A
4718 related problem was that the "from" clause in the Received: line became
4719 illegal because of the space(s). It is now covered by ${quote_local_part.
4721 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4724 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4726 83. Patch from Sieve maintainer to fix unterminated string problem in
4727 "vacation" handling.
4729 84. Some minor changes to the Linux configuration files to help with other
4730 OS variants using glibc.
4732 85. One more patch for Sieve to update vacation handling to latest spec.
4735 ----------------------------------------------------
4736 See the note above about the 4.44 and 4.50 releases.
4737 ----------------------------------------------------
4743 1. Change 4.43/35 introduced a bug that caused file counts to be
4744 incorrectly computed when quota_filecount was set in an appendfile
4747 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4748 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4751 3. Give more explanation in the error message when the command for a transport
4752 filter fails to execute.
4754 4. There are several places where Exim runs a non-Exim command in a
4755 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4756 was being done only for the command run by the queryprogram router. It is
4757 now done for all such subprocesses. The other cases are: ${run, transport
4758 filters, and the commands run by the lmtp and pipe transports.
4760 5. Some older OS have a limit of 256 on the maximum number of file
4761 descriptors. Exim was using setrlimit() to set 1000 as a large value
4762 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4763 systems. I've change it so that if it can't get 1000, it tries for 256.
4765 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4766 was an oversight, and furthermore, ever since the addition of extra
4767 controls (e.g. 4.43/32), the checks on when to allow different forms of
4768 "control" were broken. There should now be diagnostics for all cases when a
4769 control that does not make sense is encountered.
4771 7. $recipients is now available in the predata ACL (oversight).
4773 8. Tidy the search cache before the fork to do a delivery from a message
4774 received from the command line. Otherwise the child will trigger a lookup
4775 failure and thereby defer the delivery if it tries to use (for example) a
4776 cached ldap connection that the parent has called unbind on.
4778 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4779 of $address_data from the recipient verification was clobbered by the
4780 sender verification.
4782 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4783 was its contents. (It was OK if the option was not defined at all.)
4785 11. A "Completed" log line is now written for messages that are removed from
4786 the spool by the -Mrm option.
4788 12. $host_address is now set to the target address during the checking of
4789 ignore_target_hosts.
4791 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4792 being passed; this would have caused $sender_host_name to have been used if
4793 matching the list had actually called for a host name (not very likely,
4794 since this list is usually IP addresses). A host name is now passed as
4797 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4798 code that specifies a non-restarting handler (typically sigaction() in
4799 modern systems) in an attempt to fix a rare and obscure crash bug.
4801 15. Narrowed the window for a race in the daemon that could cause it to ignore
4802 SIGCHLD signals. This is not a major problem, because they are used only to
4803 wake it up if nothing else does.
4805 16. A malformed maildirsize file could cause Exim to calculate negative values
4806 for the mailbox size or file count. Odd effects could occur as a result.
4807 The maildirsize information is now recalculated if the size or filecount
4810 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4811 support for a long time. Removed HAVE_SYS_VFS_H.
4813 18. Updated exipick to current release from John Jetmore.
4815 19. Allow an empty sender to be matched against a lookup in an address list.
4816 Previously the only cases considered were a regular expression, or an
4819 20. Exim went into a mad DNS lookup loop when doing a callout where the
4820 host was specified on the transport, if the DNS lookup yielded more than
4823 21. The RFC2047 encoding function was originally intended for short strings
4824 such as real names; it was not keeping to the 75-character limit for
4825 encoded words that the RFC imposes. It now respects the limit, and
4826 generates multiple encoded words if necessary. To be on the safe side, I
4827 have increased the buffer size for the ${rfc2047: expansion operator from
4830 22. Failure to deliver a bounce message always caused it to be frozen, even if
4831 there was an errors_to setting on the router. The errors_to setting is now
4834 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4835 canonical form (fully expanded) before being placed in
4836 $sender_host_address.
4838 24. Updated eximstats to version 1.33
4840 25. Include certificate and key file names in error message when GnuTLS fails
4841 to set them up, because the GnuTLS error message doesn't include the name
4842 of the failing file when there is a problem reading it.
4844 26. Expand error message when OpenSSL has problems setting up cert/key files.
4847 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4848 (this can affect the format of dates).
4850 28. exim_tidydb, when checking for the continued existence of a message for
4851 which it has found a message-specific retry record, was not finding
4852 messages that were in split spool directories. Consequently, it was
4853 deleting retry records that should have stayed in existence.
4855 29. eximstats updated to version 1.35
4856 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4857 1.35 - bugfix such that pie charts by volume are generated correctly
4859 30. The SPA authentication driver was not abandoning authentication and moving
4860 on to the next authenticator when an expansion was forced to fail,
4861 contradicting the general specification for all authenticators. Instead it
4862 was generating a temporary error. It now behaves as specified.
4864 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4865 (the order specifies the preference for clients). The order is now AES256,
4866 AES128, 3DES, ARCFOUR128.
4868 31. Small patch to Sieve code - explicitly set From: when generating an
4871 32. Exim crashed if a remote delivery caused a very long error message to be
4872 recorded - for instance if somebody sent an entire SpamAssassin report back
4873 as a large number of 550 error lines. This bug was coincidentally fixed by
4874 increasing the size of one of Exim's internal buffers (big_buffer) that
4875 happened as part of the Exiscan merge. However, to be on the safe side, I
4876 have made the code more robust (and fixed the comments that describe what
4879 33. Some experimental protocols are using DNS PTR records for new purposes. The
4880 keys for these records are domain names, not reversed IP addresses. The
4881 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4882 leaves it alone. Component reversal etc. now happens only for IP addresses.
4885 34. The host_aton() function is supposed to be passed a string that is known
4886 to be a valid IP address. However, in the case of IPv6 addresses, it was
4887 not checking this. This is a hostage to fortune. Exim now panics and dies
4888 if the condition is not met. A case was found where this could be provoked
4889 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4890 components; fortuitously, this particular loophole had already been fixed
4891 by change 4.50/55 or 4.44/33 above.
4893 If there are any other similar loopholes, the new check in host_aton()
4894 itself should stop them being exploited. The report I received stated that
4895 data on the command line could provoke the exploit when Exim was running as
4896 exim, but did not say which command line option was involved. All I could
4897 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4898 running as the user.
4901 35. There was a buffer overflow vulnerability in the SPA authentication code
4902 (which came originally from the Samba project). I have added a test to the
4903 spa_base64_to_bits() function which I hope fixes it.
4906 36. The daemon start-up calls getloadavg() while still root for those OS that
4907 need the first call to be done as root, but it missed one case: when
4908 deliver_queue_load_max is set with deliver_drop_privilege. This is
4909 necessary for the benefit of the queue runner, because there is no re-exec
4910 when deliver_drop_privilege is set.
4912 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4913 was in use, was not putting the data itself into the right store pool;
4914 consequently, it could be overwritten for a subsequent message in the same
4915 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4918 38. Sometimes the final signoff response after QUIT could fail to get
4919 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4920 < 0 before doing a fflush(). This bug looks as though it goes back to the
4921 introduction of TLS in release 3.20, but "sometimes" must have been rare
4922 because the tests only now provoked it.
4928 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4929 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4930 Now it returns an integer. A number of calls were still expecting a T/F
4931 return. Fortuitously, in all cases, the tests worked in OK situations,
4932 which is the norm. However, things would have gone wrong on any write
4933 failures on the smtp file descriptor. This function is used when sending
4934 messages over SMTP and also when doing verify callouts.
4936 2. When Exim is called to do synchronous delivery of a locally submitted
4937 message (the -odf or -odi options), it no longer closes stderr before doing
4940 3. Implemented the mua_wrapper option.
4942 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4944 5. Implemented the functions header_remove(), header_testname(),
4945 header_add_at_position(), and receive_remove_recipient(), and exported them
4948 6. If an ACL "warn" statement specified the addition of headers, Exim already
4949 inserted X-ACL-Warn: at the start if there was no header name. However, it
4950 was not making this test for the second and subsequent header lines if
4951 there were newlines in the string. This meant that an invalid header could
4952 be inserted if Exim was badly configured.
4954 7. Allow an ACL "warn" statement to add header lines at the start or after all
4955 the Received: headers, as well as at the end.
4957 8. Added the rcpt_4xx retry error code.
4959 9. Added postmaster_mailfrom=xxx to callout verification option.
4961 10. Added mailfrom=xxxx to the callout verification option, for verify=
4964 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4965 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4967 12. Inserted some casts to stop certain compilers warning when using pointer
4968 differences as field lengths or precisions in printf-type calls (mostly
4969 affecting debugging statements).
4971 13. Added optional readline() support for -be (dynamically loaded).
4973 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4974 same clock tick as a message's arrival, so that its received time was the
4975 same as the "first fail" time on the retry record, and that message
4976 remained on the queue past the ultimate address timeout, every queue runner
4977 would try a delivery (because it was past the ultimate address timeout) but
4978 after another failure, the ultimate address timeout, which should have then
4979 bounced the address, did not kick in. This was a "< instead of <=" error;
4980 in most cases the first failure would have been in the next clock tick
4981 after the received time, and all would be well.
4983 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4984 being recognized when the domain list was tested by the match_domain
4985 condition in an expansion string.
4987 16. Added the ${str2b64: operator.
4989 17. Exim was always calling setrlimit() to set a large limit for the number of
4990 processes, without checking whether the existing limit was already
4991 adequate. (It did check for the limit on file descriptors.) Furthermore,
4992 errors from getrlimit() and setrlimit() were being ignored. Now they are
4993 logged to the main and panic logs, but Exim does carry on, to try to do its
4994 job under whatever limits there are.
4996 18. Imported PCRE 5.0.
4998 19. Trivial typo in log message " temporarily refused connection" (the leading
5001 20. If the log selector return_path_on_delivery was set and an address was
5002 redirected to /dev/null, the delivery process crashed because it assumed
5003 that a return path would always be set for a "successful" delivery. In this
5004 case, the whole delivery is bypassed as an optimization, and therefore no
5007 21. Internal re-arrangement: the function for sending a challenge and reading
5008 a response while authentication was assuming a zero-terminated challenge
5009 string. It's now changed to take a pointer and a length, to allow for
5010 binary data in such strings.
5012 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5014 23. Exim was not respecting finduser_retries when seeking the login of the
5015 uid under which it was called; it was always trying 10 times. (The default
5016 setting of finduser_retries is zero.) Also, it was sleeping after the final
5017 failure, which is pointless.
5019 24. Implemented tls_on_connect_ports.
5021 25. Implemented acl_smtp_predata.
5023 26. If the domain in control=submission is set empty, Exim assumes that the
5024 authenticated id is a complete email address when it generates From: or
5025 Sender: header lines.
5027 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5028 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5029 chown and chgrp in /bin and hostname in /usr/bin.
5031 28. Exim was keeping the "process log" file open after each use, just as it
5032 does for the main log. This opens the possibility of it remaining open for
5033 long periods when the USR1 signal hits a daemon. Occasional processlog
5034 errors were reported, that could have been caused by this. Anyway, it seems
5035 much more sensible not to leave this file open at all, so that is what now
5038 29. The long-running daemon process does not normally write to the log once it
5039 has entered its main loop, and it closes the log before doing so. This is
5040 so that log files can straightforwardly be renamed and moved. However,
5041 there are a couple of unusual error situations where the daemon does write
5042 log entries, and I had neglected to close the log afterwards.
5044 30. The text of an SMTP error response that was received during a remote
5045 delivery was being truncated at 512 bytes. This is too short for some of
5046 the long messages that one sometimes sees. I've increased the limit to
5049 31. It is now possible to make retry rules that apply only when a message has a
5050 specific sender, in particular, an empty sender.
5052 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5053 it possible to be selective about when SMTP synchronization is enforced.
5055 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5057 32. Implemented hosts_connection_nolog.
5059 33. Added an ACL for QUIT.
5061 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5064 35. Added mailbox_size and mailbox_filecount to appendfile.
5066 36. Added control = no_multiline_responses to ACLs.
5068 37. There was a bug in the logic of the code that waits for the clock to tick
5069 in the case where the clock went backwards by a substantial amount such
5070 that the microsecond fraction of "now" was more than the microsecond
5071 fraction of "then" (but the whole seconds number was less).
5073 38. Added support for the libradius Radius client library this is found on
5074 FreeBSD (previously only the radiusclient library was supported).
5080 1. When certain lookups returned multiple values in the form name=value, the
5081 quoting of the values was not always being done properly. Specifically:
5082 (a) If the value started with a double quote, but contained no whitespace,
5084 (b) If the value contained whitespace other than a space character (i.e.
5085 tabs or newlines or carriage returns) it was not quoted.
5086 This fix has been applied to the mysql and pgsql lookups by writing a
5087 separate quoting function and calling it from the lookup code. The fix
5088 should probably also be applied to nisplus, ibase and oracle lookups, but
5089 since I cannot test any of those, I have not disturbed their existing code.
5091 2. A hit in the callout cache for a specific address caused a log line with no
5092 reason for rejecting RCPT. Now it says "Previous (cached) callout
5093 verification failure".
5095 3. There was an off-by-one bug in the queryprogram router. An over-long
5096 return line was truncated at 256 instead of 255 characters, thereby
5097 overflowing its buffer with the terminating zero. As well as fixing this, I
5098 have increased the buffer size to 1024 (and made a note to document this).
5100 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5101 when Exim is waiting for an SMTP response from a remote server, Exim
5102 restarts its select() call on the socket, thereby resetting its timeout.
5103 This is not a problem when such interrupts are rare. Somebody set up a cron
5104 job to run exiwhat every 2 minutes, which is less than the normal select()
5105 timeout (5 or 10 minutes). This meant that the select() timeout never
5106 kicked in because it was always reset. I have fixed this by comparing the
5107 time when an interrupt arrives with the time at the start of the first call
5108 to select(). If more time than the timeout has elapsed, the interrupt is
5109 treated as a timeout.
5111 5. Some internal re-factoring in preparation for the addition of Sieve
5112 extensions (by MH). In particular, the "personal" test is moved to a
5113 separate function, and given an option for scanning Cc: and Bcc: (which is
5114 not set for Exim filters).
5116 6. When Exim created an email address using the login of the caller as the
5117 local part (e.g. when creating a From: or Sender: header line), it was not
5118 quoting the local part when it contained special characters such as @.
5120 7. Installed new OpenBSD configuration files.
5122 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5123 try to make them clearer.
5125 9. Callout options, other than the timeout value, were being ignored when
5126 verifying sender addresses in header lines. For example, when using
5128 verify = header_sender/callout=no_cache
5130 the cache was (incorrectly) being used.
5132 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5133 only the Cygwin environment.
5135 11. When return_path_on_delivery was set as a log selector, if different remote
5136 addresses in the same message used different return paths and parallel
5137 remote delivery occurred, the wrong values would sometimes be logged.
5138 (Whenever a remote delivery process finished, the return path value from
5139 the most recently started remote delivery process was logged.)
5141 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5142 lines when AUTH and/or TLS are in use. This is the "received protocol"
5143 field. Exim used to use "asmtp" for authenticated SMTP, without any
5144 indication (in the protocol name) for TLS use. Now it follows the RFC and
5145 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5146 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5147 names appear in log lines as well as in Received: header lines.
5149 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5150 extensions, and comparison tests, and to fix some bugs.
5152 14. Changes to the "personal" filter test:
5154 (1) The test was buggy in that it was just doing the equivalent of
5155 "contains" tests on header lines. For example, if a user's address was
5156 anne@some.where, the "personal" test would incorrectly be true for
5158 To: susanne@some.where
5160 This test is now done by extracting each address from the header in turn,
5161 and checking the entire address. Other tests that are part of "personal"
5162 are now done using regular expressions (for example, to check local parts
5163 of addresses in From: header lines).
5165 (2) The list of non-personal local parts in From: addresses has been
5166 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5167 taken from the Sieve specification recommendations.
5169 (3) If the message contains any header line starting with "List-" it is
5170 treated as non-personal.
5172 (4) The test for "circular" in the Subject: header line has been removed
5173 because it now seems ill-conceived.
5175 15. Minor typos in src/EDITME comments corrected.
5177 16. Installed latest exipick from John Jetmore.
5179 17. If headers_add on a router specified a text string that was too long for
5180 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5181 of string_sprintf() is now avoided.
5183 18. $message_body_size was not set (it was always zero) when running the DATA
5184 ACL and the local_scan() function.
5186 19. For the "mail" command in an Exim filter, no default was being set for
5187 the once_repeat time, causing a random time value to be used if "once" was
5188 specified. (If the value happened to be <= 0, no repeat happened.) The
5189 default is now 0s, meaning "never repeat". The "vacation" command was OK
5190 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5191 (I found it when inspecting the code).
5193 20. There is now an overall timeout for performing a callout verification. It
5194 defaults to 4 times the callout timeout, which applies to individual SMTP
5195 commands during the callout. The overall timeout applies when there is more
5196 than one host that can be tried. The timeout is checked before trying the
5197 next host. This prevents very long delays if there are a large number of
5198 hosts and all are timing out (e.g. when the network connections are timing
5199 out). The value of the overall timeout can be changed by specifying an
5200 additional sub-option for "callout", called "maxwait". For example:
5202 verify = sender/callout=5s,maxwait=20s
5204 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5205 to the end before writing, but this should make it even safer).
5207 22. Exim was forgetting that it had advertised PIPELINING for the second and
5208 subsequent messages on an SMTP connection. It was also not resetting its
5209 memory on STARTTLS and an internal HELO.
5211 23. When Exim logs an SMTP synchronization error within a session, it now
5212 records whether PIPELINING has been advertised or not.
5214 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5215 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5216 rather than long int.
5218 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5220 26. Added the never_mail option to autoreply.
5226 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5227 crash if the getsockname() call failed; this can happen if a connection is
5228 closed very soon after it is established. The problem was simply in the
5229 order in which certain operations were done, causing Exim to try to write
5230 to the SMTP stream before it had set up the file descriptor. The bug has
5231 been fixed by making things happen in the correct order.
5237 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5238 before the connection was closed, thus losing the rejection response.
5240 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5241 some early Solaris releases, but causes trouble in current releases where
5242 socklen_t is defined.
5244 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5247 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5250 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5251 that releasing the top of it at the end releases what was used for sub-
5252 expansions (unless the block got too big). However, discard this block if
5253 the first thing is a variable or header, so that we can use its block when
5254 it is dynamic (useful for very large $message_headers, for example).
5256 6. Lookups now cache *every* query, not just the most recent. A new, separate
5257 store pool is used for this. It can be recovered when all lookup caches are
5258 flushed. Lookups now release memory at the end of their result strings.
5259 This has involved some general refactoring of the lookup sources.
5261 7. Some code has been added to the store_xxx() functions to reduce the amount
5262 of flapping under certain conditions.
5264 8. log_incoming_interface used to affect only the <= reception log lines. Now
5265 it causes the local interface and port to be added to several more SMTP log
5266 lines, for example "SMTP connection from", and rejection lines.
5268 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5270 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5272 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5273 could be overwritten at the end of the current message (or the start of a
5274 new message if it was set in a HELO ACL). The value is now preserved for
5275 the duration of the SMTP connection.
5277 12. If a transport had a headers_rewrite setting, and a matching header line
5278 contained an unqualified address, that address was qualified, even if it
5279 did not match any rewriting rules. The underlying bug was that the values
5280 of the flags that permit the existence of unqualified sender and recipient
5281 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5282 non-local messages, and by -bnq for local messages) were not being
5283 preserved with the message after it was received.
5285 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5286 "next input=" as part of the text comprising the host identity instead of
5287 the correct text. The code was using the same buffer for two different
5288 strings. However, depending on which order the printing function evaluated
5289 its arguments, the bug did not always show up. Under Linux, for example, my
5290 test suite worked just fine.
5292 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5293 doesn't work with some older versions of Perl. It has been changed to "my",
5294 which in any case is probably the better facility to use.
5296 15. A really picky compiler found some instances of statements for creating
5297 error messages that either had too many or two few arguments for the format
5300 16. The size of the buffer for calls to the DNS resolver has been increased
5301 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5302 for addresses that have a lot of PTR records. This alleviates a problem; it
5303 does not fully solve it.
5305 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5306 buffer now truncates the list and logs the incident, which is the same
5307 action as happens when Exim is looking up a host name and its aliases.
5308 Previously in this situation something unpredictable would happen;
5309 sometimes it was "internal error: store_reset failed".
5311 18. If a server dropped the connection unexpectedly when an Exim client was
5312 using GnuTLS and trying to read a response, the client delivery process
5313 crashed while trying to generate an error log message.
5315 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5316 string, for example:
5318 warn message = H1: something\nH2: something
5320 the text was added as a single header line from Exim's point of view
5321 though it ended up OK in the delivered message. However, searching for the
5322 second and subsequent header lines using $h_h2: did not work. This has been
5323 fixed. Similarly, if a system filter added multiple headers in this way,
5324 the routers could not see them.
5326 20. Expanded the error message when iplsearch is called with an invalid key to
5327 suggest using net-iplsearch in a host list.
5329 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5330 ACLs are no longer actually imposed (and a message to that effect is
5333 22. If a "gecos" field in a passwd entry contained escaped characters, in
5334 particular, if it contained a \" sequence, Exim got it wrong when building
5335 a From: or a Sender: header from that name. A second bug also caused
5336 incorrect handling when an unquoted " was present following a character
5337 that needed quoting.
5339 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5340 was not being matched caselessly.
5342 24. Arranged for all hyphens in the exim.8 source to be escaped with
5345 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5346 properly. Recipient callout cache records were still being keyed to include
5347 the sender, even when use_sender was set false. This led to far more
5348 callouts that were necessary. The sender is no longer included in the key
5349 when use_sender is false.
5351 26. Added "control = submission" modifier to ACLs.
5353 27. Added the ${base62d: operator to decode base 62 numbers.
5355 28. dnsdb lookups can now access SRV records.
5357 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5358 the configuration file.
5360 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5361 (-v) mode. This makes the output for a verbose queue run more intelligible.
5363 31. Added a use_postmaster feature to recipient callouts.
5365 32. Added the $body_zerocount variable, containing the number of binary zero
5366 bytes in the message body.
5368 33. The time of last modification of the "new" subdirectory is now used as the
5369 "mailbox time last read" when there is a quota error for a maildir
5372 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5374 35. Added +ignore_unknown as a special item in host lists.
5376 36. Code for decoding IPv6 addresses in host lists is now included, even if
5377 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5378 address was recognized as an IP address, but was then not correctly decoded
5379 into binary, causing unexpected and incorrect effects when compared with
5386 1. Very minor rewording of debugging text in manualroute to say "list of
5387 hosts" instead of "hostlist".
5389 2. If verify=header_syntax was set, and a header line with an unqualified
5390 address (no domain) and a large number of spaces between the end of the
5391 name and the colon was received, the reception process suffered a buffer
5392 overflow, and (when I tested it) crashed. This was caused by some obsolete
5393 code that should have been removed. The fix is to remove it!
5395 3. When running in the test harness, delay a bit after writing a bounce
5396 message to get a bit more predictability in the log output.
5398 4. Added a call to search_tidyup() just before forking a reception process. In
5399 theory, someone could use a lookup in the expansion of smtp_accept_max_
5400 per_host which, without the tidyup, could leave open a database connection.
5402 5. Added the variables $recipient_data and $sender_data which get set from a
5403 lookup success in an ACL "recipients" or "senders" condition, or a router
5404 "senders" option, similar to $domain_data and $local_part_data.
5406 6. Moved the writing of debug_print from before to after the "senders" test
5409 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5410 problems for message scanning, either using a data ACL, or using
5411 local_scan() because the Received: header was not generated till after they
5412 were called (in order to set the time as the time of reception completion).
5413 I have revised the way this works. The header is now generated after the
5414 body is received, but before the ACL or local_scan() are called. After they
5415 are run, the timestamp in the header is updated.
5421 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5422 before starting a queue runner without re-exec. This happened only when
5423 deliver_drop_privilege was set or when the Exim user was set to root. The
5424 effect of the bug was that timeouts during subsequent deliveries caused
5425 crashes instead of being properly handled. The handler is now left at its
5426 default (and expected) setting.
5428 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5429 message, again when deliver_drop_privilege is set or Exim is run as root.
5430 The bug described in (1) was not present in this case, but the tidying up
5431 of the other signals was missing. I have made the two cases consistent.
5433 3. The ignore_target_hosts setting on a manualroute router was being ignored
5434 for hosts that were looked up using the /MX notation.
5436 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5439 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5440 operated on the sender address. After changing the $sender_address to <>
5441 for the sender address verify, Exim was re-instated it as the original
5442 (before rewriting) address, but remembering that it had rewritten it, so it
5443 wasn't rewriting it again. This bug also had the effect of breaking the
5444 sender address verification caching when the sender address was rewritten.
5446 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5447 This has been changed so that if the ip literal address matches
5448 ignore_target_hosts, the router declines.
5450 7. Added expansion conditions match_domain, match_address, and match_local_
5451 part (NOT match_host).
5453 8. The placeholder for the Received: header didn't have a length field set.
5455 9. Added code to Exim itself and to exim_lock to test for a specific race
5456 condition that could lead to file corruption when using MBX delivery. The
5457 issue is with the lockfile that is created in /tmp. If this file is removed
5458 after a process has opened it but before that process has acquired a lock,
5459 there is the potential for a second process to recreate the file and also
5460 acquire a lock. This could lead to two Exim processes writing to the file
5461 at the same time. The added code performs the same test as UW imapd; it
5462 checks after acquiring the lock that its file descriptor still refers to
5463 the same named file.
5465 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5466 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5469 11. Added the smtp_active_hostname option. If used, this will typically be made
5470 to depend on the incoming interface address. Because $interface_address is
5471 not set up until the daemon has forked a reception process, error responses
5472 that can happen earlier (such as "too many connections") no longer contain
5475 12. If an expansion in a condition on a "warn" statement fails because a lookup
5476 defers, the "warn" statement is abandoned, and the next ACL statement is
5477 processed. Previously this caused the whole ACL to be aborted.
5479 13. Added the iplsearch lookup type.
5481 14. Added ident_timeout as a log selector.
5483 15. Added tls_certificate_verified as a log selector.
5485 16. Added a global option tls_require_ciphers (compare the smtp transport
5486 option of the same name). This controls incoming TLS connections.
5488 17. I finally figured out how to make tls_require_ciphers do a similar thing
5489 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5490 before starting the TLS session.
5492 18. Tabs are now shown as \t in -bP output.
5494 19. If the log selector return_path_on_delivery was set, Exim crashed when
5495 bouncing a message because it had too many Received: header lines.
5497 20. If two routers both had headers_remove settings, and the first one included
5498 a superfluous trailing colon, the final name in the first list and the
5499 first name in the second list were incorrectly joined into one item (with a
5500 colon in the middle).
5506 1. Added -C and -D options to the exinext utility, mainly to make it easier
5507 to include in the automated testing, but these could be helpful when
5508 multiple configurations are in use.
5510 2. The exinext utility was not formatting the output nicely when there was
5511 an alternate port involved in the retry record key, nor when there was a
5512 message id as well (for retries that were specific to a specific message
5513 and a specific host). It was also confused by IPv6 addresses, because of
5514 the additional colons they contain. I have fixed the IPv4 problem, and
5515 patched it up to do a reasonable job for IPv6.
5517 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5518 delivery, the log line now contains "pipelined" if PIPELINING was used.
5520 4. An SMTP transport process used to panic and die if the bind() call to set
5521 an explicit outgoing interface failed. This has been changed; it is now
5522 treated in the same way as a connect() failure.
5524 5. A reference to $sender_host_name in the part of a conditional expansion
5525 that was being skipped was still causing a DNS lookup. This no longer
5528 6. The def: expansion condition was not recognizing references to header lines
5529 that used bh_ and bheader_.
5531 7. Added the _cache feature to named lists.
5533 8. The code for checking quota_filecount in the appendfile transport was
5534 allowing one more file than it should have been.
5536 9. For compatibility with Sendmail, the command line option
5544 and sets the incoming protocol and host name (for trusted callers). The
5545 host name and its colon can be omitted when only the protocol is to be set.
5546 Note the Exim already has two private options, -pd and -ps, that refer to
5547 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5548 "s", but I don't think that's a major issue.
5550 10. A number of refactoring changes to the code, none of which should affect
5553 (a) The number of logging options was getting close to filling up the
5554 32-bit word that was used as a bit map. I have split them into two classes:
5555 those that are passed in the argument to log_write(), and those that are
5556 only ever tested independently outside of that function. These are now in
5557 separate 32-bit words, so there is plenty of room for expansion again.
5558 There is no change in the user interface or the logging behaviour.
5560 (b) When building, for example, log lines, the code previously used a
5561 macro that called string_cat() twice, in order to add two strings. This is
5562 not really sufficiently general. Furthermore, there was one instance where
5563 it was actually wrong because one of the argument was used twice, and in
5564 one call a function was used. (As it happened, calling the function twice
5565 did not affect the overall behaviour.) The macro has been replaced by a
5566 function that can join an arbitrary number of extra strings onto a growing
5569 (c) The code for expansion conditions now uses a table and a binary chop
5570 instead of a serial search (which was left over from when there were very
5571 few conditions). Also, it now recognizes conditions like "pam" even when
5572 the relevant support is not compiled in: a suitably worded error message is
5573 given if an attempt is made to use such a condition.
5575 11. Added ${time_interval:xxxxx}.
5577 12. A bug was causing one of the ddress fields not to be passed back correctly
5578 from remote delivery subprocesses. The field in question was not being
5579 subsequently used, so this caused to problems in practice.
5581 13. Added new log selectors queue_time and deliver_time.
5583 14. Might have fixed a bug in maildirsizefile handling that threw up
5584 "unexpected character" debug warnings, and recalculated the data
5585 unnecessarily. In any case, I expanded the warning message to give more
5588 15. Added the message "Restricted characters in address" to the statements in
5589 the default ACL that block characters like @ and % in local parts.
5591 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5592 Three changes have been made:
5594 (a) There was a serious bug; a negative response to MAIL caused the whole
5595 recipient domain to be cached as invalid, thereby blocking all messages
5596 to all local parts at the same domain, from all senders. This bug has
5597 been fixed. The domain is no longer cached after a negative response to
5598 MAIL if the sender used is not empty.
5600 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5603 (c) A new callout option, "use_sender" has been added for people who want
5604 the modified behaviour.
5610 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5613 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5614 indeed breaks things for older releases.
5616 3. Added additional logging to the case where there is a problem reading data
5617 from a filter that is running in a subprocess using a pipe, in order to
5618 try to track down a specific problem.
5620 4. Testing facility fudge: when running in the test harness and attempting
5621 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5622 getting "No route to host". Convert this to a timeout.
5624 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5627 6. Some OS don't have socklen_t but use size_t instead. This affects the
5628 fifth argument of getsockopt() amongst other things. This is now
5629 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5630 can be set for individual OS. I have set it for SunOS5, OSF1, and
5631 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5632 some earlier ones do not.
5634 7. Change 4.30/15 was not doing the test caselessly.
5636 8. The standard form for an IPv6 address literal was being rejected by address
5637 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5638 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5639 this, as well as the form without the "IPv6" on the front (but only when
5640 address literals are enabled, of course).
5642 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5644 10. Exim crashed if a message with an empty sender address specified by -f
5645 encountered a router with an errors_to setting. This could be provoked only
5646 by a command such as
5650 where an empty string was supplied; "<>" did not hit this bug.
5652 11. Installed PCRE release 4.5.
5654 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5655 remained set. It is now erased.
5657 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5658 times from message ids (which are base 36 rather than the normal 62).
5660 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5661 were being counted as actual protocol errors, and logged if the log
5662 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5663 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5664 and DATA following a set of rejected RCPTs do not count as protocol errors.
5665 In other words, Exim assumes they were pipelined, though this may not
5666 actually be the case. Of course, in all cases the client gets an
5667 appropriate error code.
5669 15. If a lookup fails in an ACL condition, a message about the failure may
5670 be available; it is used if testing the ACL cannot continue, because most
5671 such messages specify what the cause of the deferral is. However, some
5672 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5673 that caused an old message to be retained and used if a later statement
5674 caused a defer, replacing the real cause of the deferral.
5676 16. If an IP address had so many PTR records that the DNS lookup buffer
5677 was not large enough to hold them, Exim could crash while trying to process
5678 the truncated data. It now detects and logs this case.
5680 17. Further to 4.21/58, another change has been made: if (and only if) the
5681 first line of a message (the first header line) ends with CRLF, a bare LF
5682 in a subsequent header line has a space inserted after it, so as not to
5683 terminate the header.
5685 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5686 unnecessarily, used atoi() instead of strtol(), and didn't check the
5687 termination when getting file sizes from file names by regex.
5689 19. Completely re-implemented the support for maildirsize files, in the light
5690 of a number of problems with the previous contributed implementation
5691 (4.30/29). In particular:
5693 . If the quota is zero, the maildirsize file is maintained, but no quota is
5696 . If the maildir directory does not exist, it is created before any attempt
5697 to write a maildirsize file.
5699 . The quota value in the file is just a cache; if the quota is changed in
5700 the transport, the new value overrides.
5702 . A regular expression is available for excluding directories from the
5705 20. The autoreply transport checks the characters in options that define the
5706 message's headers; it allows continued headers, but it was checking with
5707 isspace() after an embedded newline instead of explicitly looking for a
5710 21. If all the "regular" hosts to which an address was routed had passed their
5711 expiry times, and had not reached their retry times, the address was
5712 bounced, even if fallback hosts were defined. Now Exim should go on to try
5715 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5716 equivalent code in the SMTP transport. Some hosts send humungous responses
5717 to HELO/EHLO, more than 1024 it seems.
5719 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5720 gives compiler warnings in some environments. I've now done it "properly",
5723 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5724 (because of problems with the built-in one) was declared to return uschar *
5725 instead of char *, causing compiler failure.
5727 25. Fixed a file descriptor leak when processing alias/forward files.
5729 26. Fixed a minor format string issue in dbfn.c.
5731 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5733 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5734 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5737 29. When a sender address is verified, it is cached, to save repeating the test
5738 when there is more than one recipient in a message. However, when the
5739 verification involves a callout, it is possible for different callout
5740 options to be set for different recipients. It is too complicated to keep
5741 track of this in the cache, so now Exim always runs a verification when a
5742 callout is required, relying on the callout cache for the optimization.
5743 The overhead is duplication of the address routing, but this should not be
5746 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5747 to be verified with callout=postmaster, and the main callout worked but the
5748 postmaster check failed, the verification correctly failed. However, if a
5749 subsequent RCPT command asked for sender verification *without* the
5750 postmaster check, incorrect caching caused this verification also to fail,
5753 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5754 it was not caching the DNS options (qualify_single, search_parents) that
5755 were used when the lookup failed. A subsequent lookup with different
5756 options therefore always gave the same answer, though there were cases
5757 where it should not have. (Example: a "domains = !$mx_any" option on a
5758 dnslookup router: the "domains" option is always processed without any
5759 widening, but the router might have qualify_single set.) Now Exim uses the
5760 cached value only when the same options are set.
5762 32. Added John Jetmore's "exipick" utility to the distribution.
5764 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5765 than a timeout (e.g. a certificate is required, and is not provided), an
5766 Exim server now closes the connection immediately. Previously it waited for
5767 the client to close - but if the client is SSL, it seems that they each
5768 wait for each other, leading to a delay before one of them times out.
5770 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5771 maintained 0.8.x compatibility because I don't think many are using it, and
5772 it is clearly obsolete.
5774 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5777 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5778 client certificate was expired. A simple patch fixes this, though I don't
5779 understand the full logic of why the verify callback is called multiple
5782 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5783 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5784 which causes problems with some clients (such as the Certicom SSL Plus
5785 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5786 disables the coutermeasure allowing Eudora to connect."
5788 38. Exim was not checking that a write() to a log file succeeded. This could
5789 lead to Bad Things if a log got too big, in particular if it hit a file
5790 size limit. Exim now panics and dies if it cannot write to a log file, just
5791 as it does if it cannot open a log file.
5793 39. Modified OS/Makefile-Linux so that it now contains
5795 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5797 The two -D definitions ensure that Exim is compiled with large file
5798 support, which makes it possible to handle log files that are bigger than
5801 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5802 instance) a domain was checked against a named list that involved a lookup,
5803 causing $domain_data to be set, then another domain was checked against the
5804 same list, then the first domain was re-checked, the value of $domain_data
5805 after the final check could be wrong. In particular, if the second check
5806 failed, it could be set empty. This bug probably also applied to
5809 41. The strip_trailing_dot option was not being applied to the address given
5810 with the -f command-line option.
5812 42. The code for reading a message's header from the spool was incrementing
5813 $received_count, but never initializing it. This meant that the value was
5814 incorrect (doubled) while delivering a message in the same process in which
5815 it was received. In the most common configuration of Exim, this never
5816 happens - a fresh exec is done - but it can happen when
5817 deliver_drop_privilege is set.
5819 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5820 it now includes up to 150 characters of the unexpected data in the log
5823 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5824 and building data strings. The size of both of these buffers was 10 000
5825 bytes - far larger than anybody would *ever* want, thought I. Needless to
5826 say, somebody hit the limit. I have increased the maximum line length to
5827 20 000 and the maximum data length of concatenated lines to 100 000. I have
5828 also fixed two bugs, because there was no checking on these buffers. Tsk,
5829 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5830 buffer is too small.
5832 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5833 lsearch lookups. Now it does.
5835 46. When parsing a route_list item in a manualroute router, a fixed-length
5836 buffer was used for the list of hosts. I made this 1024 bytes long,
5837 thinking that nobody would ever have a list of hosts that long. Wrong.
5838 Somebody had a whole pile of complicated expansion conditions, and the
5839 string was silently truncated, leading to an expansion error. It turns out
5840 that it is easier to change to an unlimited length (owing to other changes
5841 that have happened since this code was originally written) than to build
5842 structure for giving a limitation error. The length of the item that
5843 expands into the list of hosts is now unlimited.
5845 47. The lsearch lookup could not handle data where the length of text line was
5846 more than 4095 characters. Such lines were truncated, leading to shortened
5847 data being returned. It should now handle lines of any length.
5849 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5850 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5853 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5854 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5855 no longer happens in comment lines. A list of replacements is now placed
5856 at the head of all of the source files, except those whose only change is
5857 to replace PERL_COMMAND in the very first #! line.
5859 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5860 messages on the queue, with a bottom-up merge sort, using code contributed
5861 by Michael Haardt. This should make operations like -bp somewhat faster on
5862 large queues. It won't affect queue runners, except when queue_run_in_order
5865 51. Installed eximstats 1.31 in the distribution.
5867 52. Added support for SRV lookups to the dnslookup router.
5869 53. If an ACL referred to $message_body or $message_body_end, the value was not
5870 reset for any messages that followed in the same SMTP session.
5872 54. The store-handling optimization for building very long strings was not
5873 differentiating between the different store pools. I don't think this
5874 actually made any difference in practice, but I've tidied it.
5876 55. While running the routers to verify a sender address, $sender_address
5877 was still set to the sender address. This is wrong, because when routing to
5878 send a bounce to the sender, it would be empty. Therefore, I have changed
5879 it so that, while verifying a sender address, $sender_address is set to <>.
5880 (There is no change to what happens when verifying a recipient address.)
5882 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5883 target A or AAAA records (if not already returned) without resetting the
5884 qualify_single or search_parents options of the DNS resolver. These are
5885 inappropriate in this case because the targets of MX and SRV records must
5886 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5887 target that, when qualified, matched something in the local domain. These
5888 two options are now turned off when doing these lookups.
5890 57. It seems that at least some releases of Reiserfs (which does not have the
5891 concept of a fixed number of inodes) returns zero and not -1 for the
5892 number of available inodes. This interacted badly with check_spool_inodes,
5893 which assumed that -1 was the "no such thing" setting. What I have done is
5894 to check that the total number of inodes is greater than zero before doing
5895 the test of how many are available.
5897 58. When a "warn" ACL statement has a log_message modifier, the message is
5898 remembered, and not repeated. This is to avoid a lot of repetition when a
5899 message has many recipients that cause the same warning to be written.
5900 However, Exim was preserving the list of already written lines for an
5901 entire SMTP session, which doesn't seem right. The memory is now reset if a
5902 new message is started.
5904 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5905 debugging output unless log_rewrite was also set.
5907 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5908 of (int)(handle) when we know that handle contains (void *)(-1).
5910 61. The Exim daemon panic-logs an error return when it closes the incoming
5911 connection. However "connection reset by peer" seems to be common, and
5912 isn't really an error worthy of noting specially, so that particular error
5915 62. When Exim is trying to find all the local interfaces, it used to panic and
5916 die if the ioctl to get the interface flags failed. However, it seems that
5917 on at least one OS (Solaris 9) it is possible to have an interface that is
5918 included in the list of interfaces, but for which you get a failure error
5919 for this call. This happens when the interface is not "plumbed" into a
5920 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5921 failure of the "get flags" call assumes that the interface is down.
5923 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5924 makes life easier for people who are doing arithmetic on fields extracted
5925 from dates, where you often get leading zeros that should not be
5926 interpreted as octal.
5928 64. Added qualify_domain to the redirect router, to override the global
5931 65. If a pathologically long header line contained very many addresses (the
5932 report of this problem mentioned 10 000) and each of them was rewritten,
5933 Exim could use up a very large amount of memory. (It kept on making new
5934 copies of the header line as it rewrote, and never released the old ones.)
5935 At the expense of a bit more processing, the header rewriting function has
5936 been changed so that it no longer eats memory in this way.
5938 66. The generation of the Received: header has been moved from the time that a
5939 message starts to be received, to the time that it finishes. The timestamp
5940 in the Received: header should now be very close to that of the <= log
5941 line. There are two side-effects of this change:
5943 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5944 logged header lines no longer include the local Received: line, because
5945 it has not yet been created. The same applies to a copy of the message
5946 that is returned to a non-SMTP sender when a message is rejected.
5948 (b) When a filter file is tested using -bf, no additional Received: header
5949 is added to the test message. After some thought, I decided that this
5952 This change does not affect the value of $received_for. It is still set
5953 after address rewriting, but before local_scan() is called.
5955 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5957 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5958 gave an unhelpful panic error message, and a defer error. I have managed to
5959 change this behaviour so that it now rejects any supplied certificate,
5960 which seems right, as the list of acceptable certificates is empty.
5962 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5963 gave an unhelpful defer error. I have not managed to make this reject any
5964 supplied certificates, but the error message it gives is "no certificate
5965 supplied", which is not helpful.
5967 70. exigrep's output now also includes lines that are not associated with any
5968 message, but which match the given pattern. Implemented by a patch from
5969 Martin Sluka, which also tidied up the Perl a bit.
5971 71. Recipient callout verification, like sender verification, was using <> in
5972 the MAIL FROM command. This isn't really the right thing, since the actual
5973 sender may affect whether the remote host accepts the recipient or not. I
5974 have changed it to use the actual sender in the callout; this means that
5975 the cache record is now keyed on a recipient/sender pair, not just the
5976 recipient address. There doesn't seem to be a real danger of callout loops,
5977 since a callout by the remote host to check the sender would use <>.
5978 [SEE ABOVE: changed after hitting problems.]
5980 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5981 temporary errors. However, in the case of such a code being given after
5982 the end of a data transmission (i.e. after ".") Exim was failing to write
5983 a retry record for the message. (Yes, there was some broken host that was
5984 actually sending 8xx at this point.)
5986 73. An unknown lookup type in a host list could cause Exim to panic-die when
5987 the list was checked. (An example that provoked this was putting <; in the
5988 middle of a list instead of at the start.) If this happened during a DATA
5989 ACL check, a -D file could be left lying around. This kind of configuration
5990 error no longer causes Exim to die; instead it causes a defer error. The
5991 incident is still logged to the main and panic logs.
5993 74. Buglet left over from Exim 3 conversion. The message "too many messages
5994 in one connection" was written to the rejectlog but not the mainlog, except
5995 when address rewriting (yes!) was being logged.
5997 75. Added write_rejectlog option.
5999 76. When a system filter was run not as root (that is, when system_filter_user
6000 was set), the values of the $n variables were not being returned to the
6001 main process; thus, they were not subsequently available in the $sn
6004 77. Added +return_path_on_delivery log selector.
6006 78. A connection timeout was being treated differently from recipients deferred
6007 when testing hosts_max_try with a message that was older than the host's
6008 retry timeout. (The host should not be counted, thus allowing all hosts to
6009 be tried at least once before bouncing.) This may have been the cause of an
6010 occasionally reported bug whereby a message would remain on the queue
6011 longer than the retry timeout, but would be bounced if a delivery was
6012 forced. I say "may" because I never totally pinned down the problem;
6013 setting up timeout/retry tests is difficult. See also the next item.
6015 79. The ultimate address timeout was not being applied to errors that involved
6016 a combination of host plus message (for example, a timeout on a MAIL
6017 command). When an address resolved to a number of possible hosts, and they
6018 were not all tried for each delivery (e.g. because of hosts_max_try), a
6019 message could remain on the queue longer than the retry timeout.
6021 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6024 81. Fixed an obscure SMTP outgoing bug which required at least the following
6025 conditions: (a) there was another message waiting for the same server;
6026 (b) the server returned 5xx to all RCPT commands in the first message so
6027 that the message was not completed; (c) the server dropped the connection
6028 or gave a negative response to the RSET that Exim sends to abort the
6029 transaction. The observed case was a dropped connection after DATA that had
6030 been sent in pipelining mode. That is, the server had advertised PIPELINING
6031 but was not implementing it correctly. The effect of the bug was incorrect
6032 behaviour, such as trying another host, and this could lead to a crash.
6038 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6039 and daemon.c were passed as pointers to ints; they should have been
6040 pointers to socklen_t variables (which are typically unsigned ints).
6042 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6045 3. Fixed a really odd bug that affected only the testing scheme; patching a
6046 certain fixed string in the binary changed the value of another string that
6047 happened to be identical to the end of the original first string.
6049 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6050 name", it returns that address as the IP address. On some operating
6051 systems (e.g. Solaris), it also passes back the IP address string as the
6052 "host name". However, on others (e.g. Linux), it passes back an empty
6053 string. Exim wasn't checking for this, and was changing the host name to an
6054 empty string, assuming it had been canonicalized.
6056 5. Although rare, it is permitted to have more than one PTR record for a given
6057 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6058 all the names associated with an address, because they do in Solaris.
6059 However, it seems that they do not in Linux for data that comes from the
6060 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6061 I found this out when I moved to a new Linux workstation and tried to run
6062 the Exim test suite.
6064 To get round this problem I have changed the code so that it now does its
6065 own call to the DNS to look up PTR records when searching for a host name.
6066 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6067 addresses that are only in /etc/hosts are still found.
6069 This behaviour is, however, controlled by an option called host_lookup_
6070 order, which defaults to "bydns:byaddr". If people want to use the other
6071 order, or indeed, just use one or the other means of lookup, they can
6072 specify it in this variable.
6074 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6075 some operating systems, this comes back from gethostbyaddr() as an empty
6076 string, and this is what Exim used to test for. However, it seems that in
6077 other systems, "." is yielded. Exim now tests for this case too.
6079 7. The values of check_spool_space and check_log_space are now held internally
6080 as a number of kilobytes instead of an absolute number of bytes. If a
6081 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6082 kilobyte. This means that much larger values can be stored.
6084 8. Exim monitor: an attempt to get the action menu when not actually pointing
6085 at a message produces an empty menu entitled "No message selected". This
6086 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6087 no entries in it ("Shell widget menu has zero width and/or height"). So I
6088 have added a single, blank menu entry in this case.
6090 9. Added ${quote_local_part.
6092 10. MIME decoding is now applied to the contents of Subject: header lines when
6095 11. Now that a reference to $sender_host_address automatically causes a reverse
6096 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6097 host lookup before query-style lookups in lists that might use this
6098 variable. This has therefore been abolished, and the "net-" prefix is no
6099 longer necessary for query-style lookups.
6101 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6102 have been a typo for LFLAGS, so it has been changed.
6104 13. The install script calls Exim with "-C /dev/null" in order to find the
6105 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6106 to be output. However, since Exim outputs its version number before the
6107 error, it didn't break the script. It just looked ugly. I fixed this by
6108 always allowing "-C /dev/null" if the caller is root.
6110 14. Ignore overlarge ACL variable number when reading spool file - insurance
6111 against a later release with more variables having written the file.
6113 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6114 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6115 this, as well as the form without the "IPv6" on the front.
6117 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6118 OS/Makefile-Darwin file.
6120 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6121 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6123 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6124 variable. However, some OS use ints for inodes. I've added cast to long int
6125 to get rid of the compiler warning.
6127 19. I had forgotten to lock out "/../" in configuration file names when
6128 ALT_CONFIG_PREFIX was set.
6130 20. Routers used for verification do not need to specify transports. However,
6131 if such a router generated a host list, and callout was configured, Exim
6132 crashed, because it could not find a port number from the (non-existent)
6133 transport. It now assumes port 25 in this circumstance.
6135 21. Added the -t option to exigrep.
6137 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6138 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6139 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6140 them is set, LOOKUP_LSEARCH is forced.
6142 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6144 24. Added sender and host information to the "rejected by local_scan()" log
6145 line; previously there was no indication of these.
6147 25. Added .include_if_exists.
6149 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6150 Linux. It turns out that not all file systems support this. Apparently some
6151 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6152 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6153 not supported on the file descriptor, is now ignored when Exim is trying to
6154 sync a directory. This applies only to Linux.
6156 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6158 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6159 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6162 29. Added support for maildirsize files from supplied patch (modified a bit).
6164 30. The use of :fail: followed by an empty string could lead Exim to respond to
6165 sender verification failures with (e.g.):
6167 550 Verification failed for <xxx>
6168 550 Sender verify failed
6170 where the first response line was missing the '-' that indicates it is not
6171 the final line of the response.
6173 31. The loop for finding the name of the user that called Exim had a hardwired
6174 limit of 10; it now uses the value of finduser_retries, which is used for
6175 all other user lookups.
6177 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6180 33. Exim was neglecting to zero errno before one call of strtol() when
6181 expanding a string and expecting an integer value. On some systems this
6182 resulted in spurious "integer overflow" errors. Also, it was casting the
6183 result into an int without checking.
6185 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6186 did not work. The code looks as if it has *never* worked, though it appears
6187 to have been documented since at least release 1.62. I have made it work.
6189 35. The "timeout_DNS" error in retry rules, also documented since at least
6190 1.62, also never worked. As it isn't clear exactly what this means, and
6191 clearly it isn't a major issue, I have abolished the feature by treating it
6192 as "timeout", and writing a warning to the main and panic logs.
6194 36. The display of retry rules for -brt wasn't always showing the error code
6197 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6198 timeout_connect_A, timeout_connect_MX.
6200 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6201 to the empty sender.
6203 39. The daemon was not analysing the content of -oX till after it had closed
6204 stderr and disconnected from the controlling terminal. This meant that any
6205 syntax errors were only noted on the panic log, and the return code from
6206 the command was 0. By re-arranging the code a little, I've made the
6207 decoding happen first, so such errors now appear on stderr, and the return
6208 code is 1. However, the actual setting up of the sockets still happens in
6209 the disconnected process, so errors there are still only recorded on the
6212 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6213 connections (as happens on some IP stacks) was logged at start up time as
6214 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6215 it from "IPv6 and IPv4", which means that two separate sockets are being
6218 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6219 says whether AF_INET or AF_INET6 was passed as an argument.
6221 42. Exiwhat output was messed up when time zones were included in log
6224 43. Exiwhat now gives more information about the daemon's listening ports,
6225 and whether -tls-on-connect was used.
6227 44. The "port" option of the smtp transport is now expanded.
6229 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6230 silently ignored. Now an error message is written to the main and panic
6233 46. There's a new ACL modifier called "logwrite" which writes to a log file
6234 as soon as it is encountered.
6236 47. Added $local_user_uid and $local_user_gid at routing time.
6238 48. Exim crashed when trying to verify a sender address that was being
6241 49. Exim was recognizing only a space character after ".include". It now also
6242 recognizes a tab character.
6244 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6245 extracting the relevant information from the specification. The man page no
6246 longer contains scrambled data for the -d option, and I've added a section
6247 at the front about calling Exim under different names.
6249 51. Added "extra_headers" argument to the "mail" command in filter files.
6251 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6254 53. Installed eximstats 1.29.
6256 54. Added transport_filter_timeout as a generic transport option.
6258 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6259 Cc: header lines. This was required by RFC 822, but it not required by RFC
6262 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6263 incoming messages that did not have them. Now it does so only if the
6264 message originates locally, that is, if there is no associated remote host
6265 address. When Resent- header lines are present, this applies to the Resent-
6266 lines rather than the non-Resent- lines.
6268 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6269 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6271 58. Messages for configuration errors now include the name of the main
6272 configuration file - useful now that there may be more than one file in a
6273 list (.included file names were always shown).
6275 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6276 for those rare installations that do not start the daemon as root or run it
6277 setuid root. I've cut out the call to initgroups() if the daemon is not
6280 60. The Exim user and group can now be bound into the binary as text strings
6281 that are looked up at the start of Exim's processing.
6283 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6285 62. Added $mailstore_basename variable.
6287 63. Installed patch to sieve.c from Michael Haardt.
6289 64. When Exim failed to open the panic log after failing to open the main log,
6290 the original message it was trying to log was written to stderr and debug
6291 output, but if they were not available (the usual case in production), it
6292 was lost. Now it is written to syslog before the two lines that record the
6293 failures to open the logs.
6295 65. Users' Exim filters run in subprocesses under the user's uid. It is
6296 possible for a "deliver" command or an alias in a "personal" command to
6297 provoke an address rewrite. If logging of address rewriting is configured,
6298 this fails because the process is not running as root or exim. There may be
6299 a better way of dealing with this, but for the moment (because 4.30 needs
6300 to be released), I have disabled address rewrite logging when running a
6301 filter in a non-root, non-exim process.
6307 1. The buildconfig auxiliary program wasn't quoting the value set for
6308 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6309 not defined. This bug was masked in 4.22 by the effect that was fixed in
6312 2. Some messages that were rejected after a message id was allocated were
6313 shown as "incomplete" by exigrep. It no longer does this for messages that
6314 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6316 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6317 have allow_domain_literals set, the ID did not get logged in the <= line.
6318 Domain literals are now always recognized in Message-ID: header lines.
6320 4. The first argument for a ${extract expansion item is the key name or field
6321 number. Leading and trailing spaces in this item were not being ignored,
6322 causing some misleading effects.
6324 5. When deliver_drop_privilege was set, single queue runner processes started
6325 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6326 same command in the process it spins off) were not dropping privilege.
6328 6. When the daemon running as "exim" started a queue runner, it always
6329 re-executed Exim in the spun-off process. This is a waste of effort when
6330 deliver_drop_privilege is set. The new process now just calls the
6331 queue-runner function directly.
6337 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6340 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6341 ignored. Though the use of -oP was forcing the writing of a pid file, it
6342 was always written to the default place.
6344 3. If the message "no IP address found for host xxxx" is generated during
6345 incoming verification, it is now followed by identification of the incoming
6346 connection (so you can more easily find what provoked it).
6348 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6350 5. Added some features to "harden" Exim a bit more against certain attacks:
6352 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6353 be put in Local/Makefile. This is like the never_users runtime option,
6354 but it cannot be overridden. The default setting is "root".
6356 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6357 prefix string with which any file named in a -C command line option
6360 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6361 is retained for -C and -D only if the caller of Exim is root. Without
6362 it, the exim user may also use -C and -D and retain privilege.
6364 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6365 command line option is disabled.
6367 6. Macro names set by the -D option must start with an upper case letter, just
6368 like macro names defined in the configuration file.
6370 7. Added "dereference=" facility to LDAP.
6372 8. Two instances of the typo "uknown" in the source files are fixed.
6374 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6375 the Configure-Makefile script screwed up while processing it.
6377 10. Incorporated PCRE 4.4.
6379 11. The SMTP synchronization check was not operating right at the start of an
6380 SMTP session. For example, it could not catch a HELO sent before the client
6381 waited for the greeting. There is now a check for outstanding input at the
6382 point when the greeting is written. Because of the duplex, asynchronous
6383 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6384 way, but not yet received, when the check is performed.
6386 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6387 on TCP/IP sockets, because this apparently causes some broken clients to
6390 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6391 unchanged) from the Cygwin maintainer.
6393 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6394 was supported instead of testing for mbx. Effectively a typo.
6396 15. The spa authenticator server code was not checking that the input it
6397 received was valid base64.
6399 16. The debug output line for the "set" modifier in ACLs was not showing the
6400 name of the variable that was being set.
6402 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6404 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6405 lookup on its own. Something else was needed to trigger the lookup. For
6406 example, a match in host_lookup or the need for a host name in a host list.
6407 Now, if $sender_host_name is referenced and the host name has not yet been
6408 looked up, a lookup is performed. If the lookup fails, the variable remains
6409 empty, and $host_lookup_failed is set to "1".
6411 19. Added "eqi" as a case-independent comparison operator.
6413 20. The saslauthd authentication condition could segfault if neither service
6414 nor realm was specified.
6416 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6417 error message that was logged was misleading, and incoming SMTP
6418 connections were dropped. The message is now more accurate, and temporary
6419 errors are given to SMTP connections.
6421 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6422 (except RSET) with a 421 error, until QUIT is received. However, it was
6423 failing to send a response to QUIT.
6425 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6426 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6427 ACL is now run after the helo_try_verify_hosts code.
6429 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6430 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6431 case-independent, so other case variants are also recognized). Apparently
6432 some systems use these upper case variants.
6434 25. If more than two messages were waiting for the same host, and a transport
6435 filter was specified for the transport, Exim sent two messages over the
6436 same TCP/IP connection, and then failed with "socket operation on non-
6437 socket" when it tried to send the third.
6439 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6441 27. The extern definition of crypt16() in expand.c was not being excluded when
6442 the OS had its own crypt16() function.
6444 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6445 as a preferred synonym for return_size_limit, both as an option and as an
6448 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6450 30. Changed the default configuration ACL to relax the local part checking rule
6451 for addresses that are not in any local domains. For these addresses,
6452 slashes and pipe symbols are allowed within local parts, but the sequence
6453 /../ is explicitly forbidden.
6455 31. SPA server authentication was not clearing the challenge buffer before
6458 32. log_message in a "warn" ACL statement was writing to the reject log as
6459 well as to the main log, which contradicts the documentation and doesn't
6460 seem right (because no rejection is happening). So I have stopped it.
6462 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6463 However, I am unable to do any testing of this.
6465 34. Fixed an infelicity in the appendfile transport. When checking directories
6466 for a mailbox, to see if any needed to be created, it was accidentally
6467 using path names with one or more superfluous leading slashes; tracing
6468 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6470 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6471 contents are added to the log line that is written for every discarded
6472 recipient. (Previously a log_message setting was ignored.)
6474 36. The ${quote: operator now quotes the string if it is empty.
6476 37. The install script runs exim in order to find its version number. If for
6477 some reason other than non-existence or emptiness, which it checks, it
6478 could not run './exim', it was installing it with an empty version number,
6479 i.e. as "exim-". This error state is now caught, and the installation is
6482 38. An argument was missing from the function that creates an error message
6483 when Exim fails to connect to the socket for saslauthd authentication.
6484 This could cause Exim to crash, or give a corrupted message.
6486 39. Added isip, isip4, and isip6 to ${if conditions.
6488 40. The ACL variables $acl_xx are now saved with the message, and can be
6489 accessed later in routers, transports, and filters.
6491 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6492 strings in the file are not string-expanded.
6494 42. If a MAIL command specified a SIZE value that was too large to fit into an
6495 int variable, the check against message_size_limit failed. Such values are
6496 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6497 day this will have to be increased, but I don't think I want to be around
6498 when emails are that large.
6505 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6506 iconv() is not standard in FreeBSD.
6508 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6509 IPv6 enabled. The observed symptom was a segmentation fault on return from
6510 the function os_common_find_running_interfaces() in src/os.c.
6512 3. In the check_special_case() function in daemon.c I had used "errno" as an
6513 argument name, which causes warnings on some systems. This was basically a
6514 typo, since it was named "eno" in the comments!
6516 4. The code that waits for the clock to tick (at a resolution of some fraction
6517 of a second) so as to ensure message-id uniqueness was always waiting for
6518 at least one whole tick, when it could have waited for less. [This is
6519 almost certainly not relevant at current processor speeds, where it is
6520 unlikely to ever wait at all. But we try to future-proof.]
6522 5. The function that sleeps for a time interval that includes fractions of a
6523 second contained a race. It did not block SIGALRM between setting the
6524 timer, and suspending (a couple of lines later). If the interval was short
6525 and the sigsuspend() was delayed until after it had expired, the suspension
6526 never ended. On busy systems this could lead to processes getting stuck for
6529 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6530 process, before it forks any delivery processes. The open lookup caching
6531 mechanism meant that the open file or database connection was passed into
6532 the delivery process. The problem was that delivery processes always tidy
6533 up cached lookup data. This could cause a problem for the next delivery
6534 process started by the queue runner, because the external queue runner
6535 process does not know about the closure. So the next delivery process
6536 still has data in the lookup cache. In the case of a file lookup, there was
6537 no problem because closing a file descriptor in a subprocess doesn't affect
6538 the parent. However, if the lookup was caching a connection to a database,
6539 the connection was closed, and the second delivery process was likely to
6540 see errors such as "PGSQL: query failed: server closed the connection
6541 unexpectedly". The problem has been fixed by closing all cached lookups
6542 in a queue runner before running a delivery process.
6544 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6545 seem to have the "const" qualifier which it has on other OS. I've
6548 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6549 *of the same type* (client or server) with the same public name that an
6550 error should be diagnosed.
6552 9. When Exim looked up a host name for an IP address, but failed to find the
6553 original IP address when looking up the host name (a safety check), it
6554 output the message "<ip address> does not match any IP for NULL", which was
6555 confusing, to say the least. The bug was that the host name should have
6556 appeared instead of "NULL".
6558 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6559 user that is built into the binary, and the -C or -D options is used, root
6560 privilege is dropped before the configuration file is read. In addition,
6561 logging is switched to stderr instead of the normal log files. If the
6562 configuration then re-defines the Exim user, the unprivileged environment
6563 is probably not what is expected, so Exim logs a panic warning message (but
6566 However, if deliver_drop_privilege is set, the unprivileged state may well
6567 be exactly what is intended, so the warning has been cut out in that case,
6568 and Exim is allowed to try to write to its normal log files.
6574 1. smtp_return_error_details was not giving details for temporary sender
6575 or receiver verification errors.
6577 2. Diagnose a configuration error if two authenticators have the same public
6580 3. Exim used not to create the message log file for a message until the first
6581 delivery attempt. This could be confusing when incoming messages were held
6582 for policy or load reasons. The message log file is now created at the time
6583 the message is received, and an initial "Received" line is written to it.
6585 4. The automatically generated man page for command line options had a minor
6586 bug that caused no ill effects; however, a more serious problem was that
6587 the procedure for building the man page automatically didn't always
6588 operate. Consequently, release 4.20 contains an out-of-date version. This
6589 shouldn't happen again.
6591 5. When building Exim with embedded Perl support, the script that builds the
6592 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6593 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6595 6. The freeze_tell option was not being used for messages that were frozen on
6596 arrival, either by an ACL or by local_scan().
6598 7. Added the smtp_incomplete_transaction log selector.
6600 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6601 was accepting AUTH without a new EHLO.
6603 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6604 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6607 10. Logging of TCP/IP connections (when configured) now happens in the main
6608 daemon process instead of the child process, so that the TCP/IP connection
6609 count is more accurate (but it can never be perfect).
6611 11. The use of "drop" in a nested ACL was not being handled correctly in the
6612 outer ACL. Now, if condition failure induced by the nested "drop" causes
6613 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6614 or "require"), the connection is dropped.
6616 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6617 that yield "discard" can now be used with an "accept" or a "discard" verb,
6618 but an error is generated for any others (because I can't see a useful way
6619 to define what should happen).
6621 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6622 are now processed in the same way as lines in the Exim configuration file.
6623 In particular, continuation lines are supported.
6625 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6627 15. Added -ti meaning -t -i.
6629 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6630 domains, and warn by logging if others are found.
6632 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6633 returned from GIFCONF when Exim is trying to find the list of interfaces on
6634 a host. The code in os.c has been modified to copy each ifreq to an aligned
6635 structure in all cases.
6637 Also, in some cases, the returned ifreq's were being copied to a 'struct
6638 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6639 means the last couple of bytes of an IPv6 address could be chopped if the
6640 ifreq contained only a normal sockaddr (14 bytes storage).
6642 18. Named domain lists were not supported in the hosts_treat_as_local option.
6643 An entry such as +xxxx was not recognized, and was treated as a literal
6646 19. Ensure that header lines added by a DATA ACL are included in the reject log
6647 if the ACL subsequently rejects the message.
6649 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6650 MD5 (which is deprecated).
6652 21. When testing a filter file using -bf, Exim was writing a message when it
6653 took the sender from a "From " line in the message, but it was not doing so
6654 when it took $return_path from a Return-Path: header line. It now does.
6656 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6657 with a valid header line field name (a series of printing characters
6658 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6660 23. Changed "disc" in the source to "disk" to conform to the documentation and
6661 the book and for uniformity.
6663 24. Ignore Sendmail's -Ooption=value command line item.
6665 25. When execve() failed while trying to run a command in a pipe transport,
6666 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6667 could be confused with a return value of 69 from the command itself. This
6668 has been changed to 127, the value the shell returns if it is asked to run
6669 a non-existent command. The wording for the related log line suggests a
6670 non-existent command as the problem.
6672 26. If received_header_text expands to an empty string, do not add a Received:
6673 header line to the message. (Well, it adds a token one on the spool, but
6674 marks it "old" so that it doesn't get used or transmitted.)
6676 27. Installed eximstats 1.28 (addition of -nt option).
6678 28. There was no check for failure on the call to getsockname() in the daemon
6679 code. This can fail if there is a shortage of resources on the system, with
6680 ENOMEM, for example. A temporary error is now given on failure.
6682 29. Contrary to the C standard, it seems that in some environments, the
6683 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6684 program. Exim now does this explicitly; it affects the formatting of
6685 timestamps using strftime().
6687 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6688 complaints. I've now initialized all the variables, to avoid this.
6690 32. Header lines added by a system filter were not being "seen" during
6691 transport-time rewrites.
6693 33. The info_callback() function passed to OpenSSL is set up with type void
6694 (*)(SSL *, int, int), as described somewhere. However, when calling the
6695 function (actually a macro) that sets it up, the type void(*)() is
6696 expected. I've put in a cast to prevent warnings from picky compilers.
6698 34. If a DNS black list lookup found a CNAME record, but there were no A
6699 records associated with the domain it pointed at, Exim crashed.
6701 35. If a DNS black list lookup returned more than one A record, Exim ignored
6702 all but the first. It now scans all returned addresses if a particular IP
6703 value is being sought. In this situation, the contents of the
6704 $dnslist_value variable are a list of all the addresses, separated by a
6707 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6708 to accept a host name and all its aliases if the forward lookup for any of
6709 them yielded the IP address of the incoming connection. Now it accepts only
6710 those names whose forward lookup yields the correct IP address. Any other
6711 names are discarded. This closes a loophole whereby a rogue DNS
6712 administrator could create reverse DNS records to break through a
6713 wildcarded host restriction in an ACL.
6715 37. If a user filter or a system filter that ran in a subprocess used any of
6716 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6717 the wrong values were passed to the pipe command ($thisaddress had the
6718 value of $0, $0 had the value of $1, etc). This bug was introduced by
6719 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6721 38. Improved the line breaking for long SMTP error messages from ACLs.
6722 Previously, if there was no break point between 40 and 75 characters, Exim
6723 left the rest of the message alone. Two changes have been made: (a) I've
6724 reduced the minimum length to 35 characters; (b) if it can't find a break
6725 point between 35 and 75 characters, it looks ahead and uses the first one
6726 that it finds. This may give the occasional overlong line, but at least the
6727 remaining text gets split now.
6729 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6730 file descriptors might be low, and that setting 1000 would always raise it.
6731 It turns out that in some environments, the limit is already over 1000 and
6732 that lowering it causes trouble. So now Exim takes care not to decrease it.
6734 40. When delivering a message, the value of $return_path is set to $sender_
6735 address at the start of routing (routers may change the value). By an
6736 oversight, this default was not being set up when an address was tested by
6737 -bt or -bv, which affected the outcome if any router or filter referred to
6740 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6741 log only when "message" or "log_message" are set. However, if one of the
6742 conditions was an address verification, or a call to a nested ACL, the
6743 messages generated by the underlying test were being passed through. This
6744 no longer happens. The underlying message is available in $acl_verify_
6745 message for both "message" and "log_message" expansions, so it can be
6746 passed through if needed.
6748 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6749 new expansion $bh_ to give the encoded byte string without charset
6750 translation. Translation happens only if iconv() is available; HAVE_ICONV
6751 indicates this at build time. HEADERS_CHARSET gives the charset to
6752 translate to; headers_charset can change it in the configuration, and
6753 "headers charset" can change it in an individual filter file.
6755 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6756 that creates RFC 2047 encoded "words" labels them as that charset instead
6757 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6758 expansion operator; (ii) when Exim creates a From: line for a local
6759 message; (iii) when a header line is rewritten to include a "phrase" part.
6761 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6762 buggy, causing it to skip the first lines of messages whose message ID
6763 ended in 'D'. This would not have bitten before Exim release 4.14, because
6764 message IDs were unlikely to end in 'D' before then. The effect was to have
6765 incorrect size information for certain domains.
6767 45. #include "config.h" was missing at the start of the crypt16.c module. This
6768 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6771 46. If there was a timeout during a "random" callout check, Exim treated it as
6772 a failure of the random address, and carried on sending RSET and the real
6773 address. If the delay was just some slowness somewhere, the response to the
6774 original RCPT would be taken as a response to RSET and so on, causing
6775 mayhem of various kinds.
6777 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6778 when I implemented it. It didn't allow for the fact that some option values
6779 may legitimately be negative (e.g. size_addition), and it didn't even do
6780 the right test for positive values.
6782 48. Domain names in DNS records are case-independent. Exim always looks them up
6783 in lower case. Some resolvers return domain names in exactly the case they
6784 appear in the zone file, that is, they may contain uppercase letters. Not
6785 all resolvers do this - some return always lower case. Exim was treating a
6786 change of case by a resolver as a change of domain, similar to a widening
6787 of a domain abbreviation. This triggered its re-routing code and so it was
6788 trying to route what was effectively the same domain again. This normally
6789 caused routing to fail (because the router wouldn't handle the domain
6790 twice). Now Exim checks for this case specially, and just changes the
6791 casing of the domain that it ultimately uses when it transmits the message
6794 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6797 50. If a filter generated a file delivery with a non-absolute name (possible if
6798 no home directory exists for the router), the forbid_file option was not
6801 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6802 the existing equality matching.
6804 52. Exim was using ints instead of ino_t variables in some places where it was
6805 dealing with inode numbers.
6807 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6808 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6809 called TMPDIR, and if it finds it is different, it changes its value.
6811 54. The smtp_printf() function is now made available to local_scan() so
6812 additional output lines can be written before returning. There is also an
6813 smtp_fflush() function to enable the detection of a dropped connection.
6814 The variables smtp_input and smtp_batched_input are exported to
6817 55. Changed the default runtime configuration: the message "Unknown user"
6818 has been removed from the ACL, and instead placed on the localuser router,
6819 using the cannot_route_message feature. This means that any verification
6820 failures that generate their own messages won't get overridden. Similarly,
6821 the "Unrouteable address" message that was in the ACL for unverifiable
6822 relay addresses has also been removed.
6824 56. Added hosts_avoid_esmtp to the smtp transport.
6826 57. The exicyclog script was not checking for the esoteric option
6827 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6828 will work only if exicyclog is run under the appropriate euid.
6830 58. Following a discussion on the list, the rules by which Exim recognises line
6831 endings on incoming messages have been changed. The -dropcr and drop_cr
6832 options are now no-ops, retained only for backwards compatibility. The
6833 following line terminators are recognized: LF CRLF CR. However, special
6834 processing applies to CR:
6836 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6837 nor a local message in the state where . is a terminator.
6839 (ii) If a bare CR is encountered in a header line, an extra space is added
6840 after the line terminator so as not to end the header. The reasoning
6841 behind this is that bare CRs in header lines are most likely either
6842 to be mistakes, or people trying to play silly games.
6844 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6845 was being incorrectly given as 18 bytes larger than it should have been.
6846 This is a VOB (very old bug).
6848 60. This may never have affected anything current, but just in case it has:
6849 When the local host is found other than at the start of a list of hosts,
6850 the local host, those with the same MX, and any that follow, are discarded.
6851 When the list in question was part of a longer list of hosts, the following
6852 hosts (not currently being processed) were also being discarded. This no
6853 longer happens. I'm not sure if this situation could ever has previously
6856 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6859 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6860 header. This is something that is recommended in a new Internet Draft, and
6861 is something that is documented as being done by Sendmail. There are two
6862 possible values. For messages generated by the autoreply transport, Exim
6865 Auto-Submitted: auto-replied
6867 whereas for all other generated messages (e.g. bounces) it adds
6869 Auto-Submitted: auto-generated
6871 63. The "personal" condition in filters now includes a test for the
6872 Auto-Submitted: header. If it contains the string "auto-" the message it
6873 not considered personal.
6875 64. Added rcpt_include_affixes as a generic transport option.
6877 65. Added queue_only_override (default true).
6879 66. Added the syslog_duplication option.
6881 67. If what should have been the first header line of a message consisted of
6882 a space followed by a colon, Exim was mis-interpreting it as a header line.
6883 It isn't of course - it is syntactically invalid and should therefore be
6884 treated as the start of the message body. The misbehaviour could have
6885 caused a number of strange effects, including loss of data in subsequent
6886 header lines, and spool format errors.
6888 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6889 client host had authenticated. This control can now be exercised by an ACL
6890 for more flexibility.
6892 69. By default, callouts do not happen when testing with -bh. There is now a
6893 variant, -bhc, which does actually run the callout code, including
6894 consulting and updating the callout cache.
6896 70. Added support for saslauthd authentication, courtesy of Alexander
6899 71. If statvfs() failed on the spool or log directories while checking their
6900 size for availability, Exim confusingly gave the error "space shortage".
6901 Furthermore, in debugging mode it crashed with a floating point exception.
6902 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6903 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6904 serious problem, Exim now writes to the main and panic logs when this
6905 happens, with details of the failure. It then refuses to accept the
6906 incoming message, giving the message "spool directory problem" or "log
6907 directory problem" with a 421 code for SMTP messages.
6909 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6910 0, 1, and 2 exist, because some OS complain for execs without them (see
6911 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6912 descriptors. However, the code omitted to check that the open succeeded,
6913 causing mysterious errors if for some reason the permissions on /dev/null
6914 got screwed. Now Exim writes a message to the main and panic logs, and
6915 bombs out if it can't open /dev/null.
6917 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6918 interact so that it is all more flexible. It is supposed to remain
6919 backwards compatible. Also added extra_local_interfaces.
6921 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6922 to bomb out with an assertion failure - to the client this appears as a
6923 connection drop. This problem occurs in the part of the code that was taken
6924 from the Samba project. Fortunately, the assertion is in a very simple
6925 function, so I have fixed this by reproducing the function inline in the
6926 one place where it is called, and arranging for authentication to fail
6927 instead of killing the process with assert().
6929 75. The SPA client code was not working when the server requested OEM rather
6930 than Unicode encoding.
6932 76. Added code to make require_files with a specific uid setting more usable in
6933 the case where statting the file as root fails - usually a non-root-mounted
6934 NFS file system. When this happens and the failure is EACCES, Exim now
6935 forks a subprocess and does the per-uid checking as the relevant uid.
6937 77. Added process_log_path.
6939 78. If log_file_path was not explicitly set, a setting of check_log_space or
6940 check_log_inodes was ignored.
6942 79. If a space check for the spool or log partitions fails, the incident is now
6943 logged. Of course, in the latter case the data may get lost...
6945 80. Added the %p formatting code to string_format() so that it can be used to
6946 print addresses in debug_print(). Adjusted all the address printing in the
6947 debugging in store.c to use %p rather than %d.
6949 81. There was a concern that a line of code in smtp_in.c could overflow a
6950 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6951 initially expressed, the concern was not well-founded, because trailing
6952 spaces are removed early. However, if the trailing spaces were followed by
6953 a NULL, they did not get removed, so the overflow was possible. Two fixes
6956 (a) I re-wrote the offending code in a cleaner fashion.
6957 (b) If an incoming SMTP command contains a NULL character, it is rejected
6960 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6961 runs initgroups(), so that if the Exim user is in any additional groups,
6962 they will be used during message reception.
6968 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob