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
212 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
213 than -2003 did; needs libidn2 in addition to libidn.
215 JH/02 The path option on a pipe transport is now expanded before use.
217 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
218 Patch provided by "Björn", documentation fix added too.
220 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
221 missing a wire-to-host endian conversion.
223 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
224 close after a BDAT command line could be taken as a following command,
225 giving a synch failure. Fix by only checking for synch immediately
226 before acknowledging the chunk.
228 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
229 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
230 macro. Patches provided by Josh Soref.
232 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
233 Previously we did not; the RFC seems ambiguous and VRFY is not listed
234 by IANA as a service extension. However, John Klensin suggests that we
237 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
238 the dkim code may be unix-mode line endings rather than smtp wire-format
239 CRLF, so prepend a CR to any bare LF.
241 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
242 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
244 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
245 the first were themselves being wrongly included in the feed into dkim
246 processing; with most chunk sizes in use this resulted in an incorrect
247 body hash calculated value.
249 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
250 DKIM signature block, for verification. Although advised against by
251 standards it is specifically not ruled illegal.
253 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
255 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
256 missing a body hash (the bh= tag).
258 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
259 It seems that HAProxy sends the Proxy Protocol information in clear and
260 only then does a TLS startup, so do the same.
262 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
263 TCP connections (such as for Spamd) unless the daemon successfully set
264 Fast Open mode on its listening sockets. This fixes breakage seen on
265 too-old kernels or those not configured for Fast Open, at the cost of
266 requiring both directions being enabled for TFO, and TFO never being used
267 by non-daemon-related Exim processes.
269 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
270 endings, at least on the first header line. Try to canonify any that get
271 past that check, despite the cost.
273 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
274 now limited to an arbitrary five deep, while parsing addresses with the
275 strip_excess_angle_brackets option enabled.
277 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
278 instead leave the unprompted TLS handshake in socket buffer for the
279 TLS library to consume.
281 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
283 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
285 JH/16 Drop variables when they go out of scope. Memory management drops a whole
286 region in one operation, for speed, and this leaves assigned pointers
287 dangling. Add checks run only under the testsuite which checks all
288 variables at a store-reset and panics on a dangling pointer; add code
289 explicitly nulling out all the variables discovered. Fixes one known
290 bug: a transport crash, where a dangling pointer for $sending_ip_address
291 originally assigned in a verify callout, is re-used.
293 PP/06 Drop '.' from @INC in various Perl scripts.
295 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
297 PP/08 Reduce a number of compilation warnings under clang; building with
298 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
299 should be warning-free.
301 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
303 HS/01 Fix portability problems introduced by PP/08 for platforms where
304 realloc(NULL) is not equivalent to malloc() [SunOS et al].
306 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
307 chunk. This allows us to accept broken chunked messages. We need a more
308 general solution here.
310 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
311 already-broken messages in the queue.
313 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
315 JH/19 Fix reference counting bug in routing-generated-address tracking.
321 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
322 supports it and a size is available (ie. the sending peer gave us one).
324 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
325 years of being deprecated). The replacements are the ACLs
326 acl_smtp_mime and acl_not_smtp_mime.
328 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
329 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
330 or even an in-clear connection were permitted. Now, if the host lookup
331 was dnssec and dane was requested then the host is only used if the
332 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
333 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
334 if one fails this test.
335 This means that a poorly-configured remote DNS will make it incommunicado;
336 but it protects against a DNS-interception attack on it.
338 JH/04 Bug 1810: make continued-use of an open smtp transport connection
339 non-noisy when a race steals the message being considered.
341 JH/05 If main configuration option tls_certificate is unset, generate a
342 self-signed certificate for inbound TLS connections.
344 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
345 in rewrites and routers.
347 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
348 and logged a warning sing 4.83; now they are a configuration file error.
350 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
351 (lacking @domain). Apply the same qualification processing as RCPT.
353 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
355 JH/10 Support ${sha256:} applied to a string (as well as the previous
358 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
359 a cutthrough deliver is pending, as we always want to make a connection.
360 This also avoids re-routing the message when later placing the cutthrough
361 connection after a verify cache hit.
362 Do not update it with the verify result either.
364 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
365 when routing results in more than one destination address.
367 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
368 signing (which inhibits the cutthrough capability). Previously only
369 the presence of an option was tested; now an expansion evaluating as
370 empty is permissible (obviously it should depend only on data available
371 when the cutthrough connection is made).
373 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
374 the relevant preceding SMTP command did not note the pipelining mode.
376 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
377 Previously they were not counted.
379 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
380 as one having no matching records. Previously we deferred the message
381 that needed the lookup.
383 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
384 distinguished as "(=".
386 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
387 for missing MX records. Previously it only worked for missing A records.
389 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
391 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
392 after the data-go-ahead and data-ack. Patch from Jason Betts.
394 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
395 even for a "none" policy. Patch from Tony Meyer.
397 JH/22 Fix continued use of a connection for further deliveries. If a port was
398 specified by a router, it must also match for the delivery to be
401 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
402 When one of the recipients of a message was unsuitable for the connection
403 (has no matching addresses), we lost track of needing to mark it
404 deferred. As a result mail would be lost.
406 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
408 JH/25 Decoding ACL controls is now done using a binary search; the source code
409 takes up less space and should be simpler to maintain. Merge the ACL
410 condition decode tables also, with similar effect.
412 JH/26 Fix problem with one_time used on a redirect router which returned the
413 parent address unchanged. A retry would see the parent address marked as
414 delivered, so not attempt the (identical) child. As a result mail would
417 JH/27 Fix a possible security hole, wherein a process operating with the Exim
418 UID can gain a root shell. Credit to http://www.halfdog.net/ for
419 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
422 JH/28 Enable {spool,log} filesystem space and inode checks as default.
423 Main config options check_{log,spool}_{inodes,space} are now
424 100 inodes, 10MB unless set otherwise in the configuration.
426 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
427 Previously it only applied to the main-section connection policy
430 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
432 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
433 by me. Added RFC7919 DH primes as an alternative.
435 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
436 are not in the system include path.
438 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
439 GnuTLS, when a session startup failed (eg because the client disconnected)
440 Exim did stdio operations after fclose. This was exposed by a recent
441 change which nulled out the file handle after the fclose.
443 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
444 signed directly by the cert-signing cert, rather than an intermediate
445 OCSP-signing cert. This is the model used by LetsEncrypt.
447 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
449 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
450 an incoming connection.
452 HS/02 Bug 1802: Do not half-close the connection after sending a request
455 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
456 fallback to "prime256v1".
458 JH/34 SECURITY: Use proper copy of DATA command in error message.
459 Could leak key material. Remotely exploitable. CVE-2016-9963.
465 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
466 and 3.4.4 - once the server is enabled to respond to an OCSP request
467 it does even when not requested, resulting in a stapling non-aware
468 client dropping the TLS connection.
470 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
471 support variable-length bit vectors. No functional change.
473 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
474 The I= interface field on outgoing lines is now after the H= remote
475 host field, same as incoming lines. There is a separate
476 outgoing_interface log selector which allows you to disable the
479 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
480 If not running log_selector +smtp_connection the mainlog would be held
481 open indefinitely after a "too many connections" event, including to a
482 deleted file after a log rotate. Leave the per net connection logging
483 leaving it open for efficiency as that will be quickly detected by the
484 check on the next write.
486 HS/01 Bug 1671: Fix post transport crash.
487 Processing the wait-<transport> messages could crash the delivery
488 process if the message IDs didn't exist for some reason. When
489 using 'split_spool_directory=yes' the construction of the spool
490 file name failed already, exposing the same netto behaviour.
492 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
493 mime_regex ACL conditions.
495 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
496 to DSN fail messages (bounces): remote IP, remote greeting, remote response
497 to HELO, local diagnostic string.
499 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
500 log line to debug. Even when configured with a tls authenticator many
501 client connections are expected to not authenticate in this way, so
502 an authenticate fail is not an error.
504 HS/02 Add the Exim version string to the process info. This way exiwhat
505 gives some more detail about the running daemon.
507 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
508 matter for fast-change records such as DNSBLs.
510 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
511 retry record, even if constant. There may be multiple transports with
512 different interface settings and the retry behaviour needs to be kept
515 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
517 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
519 JH/10 Bug 840: fix log_defer_output option of pipe transport
521 JH/11 Bug 830: use same host for all RCPTS of a message, even under
522 hosts_randomize. This matters a lot when combined with mua_wrapper.
524 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
525 ${quote_pgsql:<string>} operator.
527 JH/13 Bug 1708: avoid misaligned access in cached lookup.
529 JH/14 Change header file name for freeradius-client. Relevant if compiling
530 with Radius support; from the Gentoo tree and checked under Fedora.
532 JH/15 Bug 1712: Introduce $prdr_requested flag variable
534 JH/16 Bug 1714: Permit an empty string as expansion result for transport
535 option transport_filter, meaning no filtering.
537 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
539 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
540 defaults to "*" (all hosts). The variable is now available when not built
541 with TLS, default unset, mainly to enable keeping the testsuite sane.
542 If a server certificate is not supplied (via tls_certificate) an error is
543 logged, and clients will find TLS connections fail on startup. Presumably
544 they will retry in-clear.
545 Packagers of Exim are strongly encouraged to create a server certificate
546 at installation time.
548 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
549 with the $config_file variable.
551 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
552 in transport context, after the attempt, and per-recipient. The latter type
553 is per host attempted. The event data is the error message, and the errno
554 information encodes the lookup type (A vs. MX) used for the (first) host,
555 and the trailing two digits of the smtp 4xx response.
557 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
558 to write to mainlog (or rejectlog, paniclog) in the window between file
559 creation and permissions/ownership being changed. Particularly affects
560 installations where exicyclog is run as root, rather than exim user;
561 result is that the running daemon panics and dies.
563 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
565 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
566 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
567 "pri" and "weight". Note that the previous implicit priority given by the
568 list order is no longer honoured.
570 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
573 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
574 by defining SUPPORT_SOCKS.
576 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
577 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
578 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
579 variables are renamed to proxy_{local,external}_{address,port}.
581 JH/27 Move Internationalisation support from Experimental to mainline, enabled
582 for a build by defining SUPPORT_I18N
584 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
585 of the query string, and make ${quote_redis:} do that quoting.
587 JH/29 Move Events support from Experimental to mainline, enabled by default
588 and removable for a build by defining DISABLE_EVENT.
590 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
592 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
593 cached by the daemon.
595 JH/32 Move Redis support from Experimental to mainline, enabled for a build
596 by defining LOOKUP_REDIS. The libhiredis library is required.
598 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
599 keys are given for lookup.
601 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
602 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
603 only supported when built with TLS support. The PolarSSL SHA routines
604 are still used when the TLS library is too old for convenient support.
606 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
607 openssl_options), for security. OpenSSL forces this from version 1.1.0
608 server-side so match that on older versions.
610 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
611 allocation for $value could be released as the expansion processing
612 concluded, but leaving the global pointer active for it.
614 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
615 and to use the domains and local_parts ACL conditions.
617 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
618 incorrectly not doubled on cutthrough transmission, hence seen as a
619 body-termination at the receiving system - resulting in truncated mails.
620 Commonly the sender saw a TCP-level error, and retransmitted the message
621 via the normal store-and-forward channel. This could result in duplicates
622 received - but deduplicating mailstores were liable to retain only the
623 initial truncated version.
625 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
627 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
629 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
630 we're in there, support oversigning also; bug 1309.
632 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
634 HS/04 Add support for keep_environment and add_environment options.
636 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
637 either intentional arithmetic overflow during PRNG, or testing config-
640 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
641 delivery resulted in actual delivery. Cancel cutthrough before DATA
644 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
645 rejects a recipient: pass the reject to the originator.
647 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
648 Many were false-positives and ignorable, but it's worth fixing the
651 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
652 for the new environment-manipulation done at startup. Move the routines
653 from being local to tls.c to being global via the os.c file.
655 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
656 an extract embedded as result-arg for a map, the first arg for extract
657 is unavailable so we cannot tell if this is a numbered or keyed
658 extraction. Accept either.
664 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
667 JH/02 The smtp transport option "multi_domain" is now expanded.
669 JH/03 The smtp transport now requests PRDR by default, if the server offers
672 JH/04 Certificate name checking on server certificates, when exim is a client,
673 is now done by default. The transport option tls_verify_cert_hostnames
674 can be used to disable this per-host. The build option
675 EXPERIMENTAL_CERTNAMES is withdrawn.
677 JH/05 The value of the tls_verify_certificates smtp transport and main options
678 default to the word "system" to access the system default CA bundle.
679 For GnuTLS, only version 3.0.20 or later.
681 JH/06 Verification of the server certificate for a TLS connection is now tried
682 (but not required) by default. The verification status is now logged by
683 default, for both outbound TLS and client-certificate supplying inbound
686 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
689 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
690 Status Notification (bounce) messages are now MIME format per RFC 3464.
691 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
692 under the control of the dsn_advertise_hosts option, and routers may
693 have a dsn_lasthop option.
695 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
696 default, modifiable by a malware= option. The list separator for
697 the options can now be changed in the usual way. Bug 68.
699 JH/10 The smtp_receive_timeout main option is now expanded before use.
701 JH/11 The incoming_interface log option now also enables logging of the
702 local interface on delivery outgoing connections.
704 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
705 if the interface and destination host and port all match.
707 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
710 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
711 Patch from Andrew Lewis.
713 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
714 now supports optional time-restrictions, weighting, and priority
715 modifiers per server. Patch originally by <rommer@active.by>.
717 JH/16 The spamd_address main option now supports a mixed list of local
718 and remote servers. Remote servers can be IPv6 addresses, and
719 specify a port-range.
721 JH/17 Bug 68: The spamd_address main option now supports an optional
722 timeout value per server.
724 JH/18 Bug 1581: Router and transport options headers_add/remove can
725 now have the list separator specified.
727 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
730 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
733 JH/21 Support for the A6 type of dns record is withdrawn.
735 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
736 rather than the verbs used.
738 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
739 from 255 to 1024 chars.
741 JH/24 Verification callouts now attempt to use TLS by default.
743 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
744 are generic router options now. The defaults didn't change.
746 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
747 Original patch from Alexander Shikoff, worked over by JH.
749 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
750 files if ZCAT_COMMAND is not executable.
752 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
754 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
756 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
757 Normally benign, it bites when the pair was led to by a CNAME;
758 modern usage is to not canonicalize the domain to a CNAME target
759 (and we were inconsistent anyway for A-only vs AAAA+A).
761 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
763 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
764 when evaluating $sender_host_dnssec.
766 JH/31 Check the HELO verification lookup for DNSSEC, adding new
767 $sender_helo_dnssec variable.
769 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
771 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
773 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
775 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
776 documented as working, but never had. Support all but $spam_report.
778 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
779 added for tls authenticator.
781 HS/03 Add perl_taintmode main config option
787 TL/01 When running the test suite, the README says that variables such as
788 no_msglog_check are global and can be placed anywhere in a specific
789 test's script, however it was observed that placement needed to be near
790 the beginning for it to behave that way. Changed the runtest perl
791 script to read through the entire script once to detect and set these
792 variables, reset to the beginning of the script, and then run through
793 the script parsing/test process like normal.
795 TL/02 The BSD's have an arc4random API. One of the functions to induce
796 adding randomness was arc4random_stir(), but it has been removed in
797 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
798 function when detected.
800 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
801 cause callback expansion.
803 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
804 syntax errors in an expansion can be treated as a string instead of
805 logging or causing an error, due to the internal use of bool_lax
806 instead of bool when processing it.
808 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
809 server certificates when making smtp deliveries.
811 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
813 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
815 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
817 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
818 Merged patch from Sebastian Wiedenroth.
820 JH/05 Fix results-pipe from transport process. Several recipients, combined
821 with certificate use, exposed issues where response data items split
822 over buffer boundaries were not parsed properly. This eventually
823 resulted in duplicates being sent. This issue only became common enough
824 to notice due to the introduction of connection certificate information,
825 the item size being so much larger. Found and fixed by Wolfgang Breyha.
827 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
828 size buffer was used, resulting in syntax errors when an expansion
831 JH/07 Add support for directories of certificates when compiled with a GnuTLS
832 version 3.3.6 or later.
834 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
835 is EXPERIMENTAL_EVENT, the main-configuration and transport options
836 both become "event_action", the variables become $event_name, $event_data
837 and $event_defer_errno. There is a new variable $verify_mode, usable in
838 routers, transports and related events. The tls:cert event is now also
839 raised for inbound connections, if the main configuration event_action
842 TL/06 In test suite, disable OCSP for old versions of openssl which contained
843 early OCSP support, but no stapling (appears to be less than 1.0.0).
845 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
846 server certificate names available under the smtp transport option
847 "tls_verify_cert_hostname" now do not permit multi-component wildcard
850 JH/10 Time-related extraction expansions from certificates now use the main
851 option "timezone" setting for output formatting, and are consistent
852 between OpenSSL and GnuTLS compilations. Bug 1541.
854 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
855 encoded parameter in the incoming message. Bug 1558.
857 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
858 include certificate info, eximon was claiming there were spoolfile
861 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
863 JH/14 Log delivery-related information more consistently, using the sequence
864 "H=<name> [<ip>]" wherever possible.
866 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
867 are problematic for Debian distribution, omit them from the release
870 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
872 JH/16 Fix string representation of time values on 64bit time_t architectures.
875 JH/17 Fix a null-indirection in certextract expansions when a nondefault
876 output list separator was used.
881 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
882 checkers that were complaining about end of non-void function with no
885 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
886 This was a regression introduced in 4.83 by another bugfix.
888 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
890 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
891 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
897 TF/01 Correctly close the server side of TLS when forking for delivery.
899 When a message was received over SMTP with TLS, Exim failed to clear up
900 the incoming connection properly after forking off the child process to
901 deliver the message. In some situations the subsequent outgoing
902 delivery connection happened to have the same fd number as the incoming
903 connection previously had. Exim would try to use TLS and fail, logging
904 a "Bad file descriptor" error.
906 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
907 utilities have not been installed.
909 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
910 temporary space as the ACL may create new global variables.
912 TL/01 LDAP support uses per connection or global context settings, depending
913 upon the detected version of the libraries at build time.
915 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
916 to extract and use the src ip:port in logging and expansions as if it
917 were a direct connection from the outside internet. PPv2 support was
918 updated based on HAProxy spec change in May 2014.
920 JH/02 Add ${listextract {number}{list}{success}{fail}}.
922 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
923 Properly escape header and check for NULL return.
925 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
928 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
930 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
931 characters in header names, implemented as a verify condition.
932 Contributed by Michael Fischer v. Mollard.
934 TL/05 Rename SPF condition results err_perm and err_temp to standardized
935 results permerror and temperror. Previous values are deprecated but
936 still accepted. In a future release, err_perm and err_temp will be
937 completely removed, which will be a backward incompatibility if the
938 ACL tests for either of these two old results. Patch contributed by
939 user bes-internal on the mailing list.
941 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
943 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
944 selectors, in both main and reject logs.
946 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
949 JH/07 Add malware type "sock" for talking to simple daemon.
951 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
953 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
954 routers/transports under cutthrough routing.
956 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
957 numbers. Touch up "bool" conditional to keep the same definition.
959 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
961 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
963 JH/12 Expand items in router/transport headers_add or headers_remove lists
964 individually rather than the list as a whole. Bug 1452.
966 Required for reasonable handling of multiple headers_ options when
967 they may be empty; requires that headers_remove items with embedded
968 colons must have them doubled (or the list-separator changed).
970 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
971 view the policy declared in the DMARC record. Currently, $dmarc_status
972 is a combined value of both the record presence and the result of the
975 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
977 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
978 dnslookup router and the smtp transport (applying to the forward
981 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
982 of ldap servers used for a specific lookup. Patch provided by Heiko
985 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
986 New variable $lookup_dnssec_authenticated for observability.
988 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
989 Patch submitted by Lars Timman.
991 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
993 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
994 Requires trusted mode and valid format message id, aborts otherwise.
995 Patch contributed by Heiko Schlichting.
997 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
998 certextract with support for various fields. Bug 1358.
1000 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1001 is requested by default, modifiable by smtp transport option
1004 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1005 operate on certificate variables to give certificate fingerprints
1006 Also new ${sha256:cert_variable}.
1008 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1010 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1011 Christian Aistleitner.
1013 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1015 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1016 file. Patch from Wolfgang Breyha.
1018 JH/25 Expand the coverage of the delivery $host and $host_address to
1019 client authenticators run in verify callout. Bug 1476.
1021 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1022 align with daemon_smtp_ports. Bug 72.
1024 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1025 support and error reporting did not work properly.
1027 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1028 and is readable. Patch from Andrew Colin Kissa.
1030 TL/14 Enhance documentation of ${run expansion and how it parses the
1031 commandline after expansion, particularly in the case when an
1032 unquoted variable expansion results in an empty value.
1034 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1036 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1037 by renaming away from it. Addresses GH issue 15, reported by
1040 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1042 TL/15 SECURITY: prevent double expansion in math comparison functions
1043 (can expand unsanitized data). Not remotely exploitable.
1050 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1052 PP/02 Make -n do something, by making it not do something.
1053 When combined with -bP, the name of an option is not output.
1055 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1058 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1059 $sender_host_name and config options to manage this, and basic check
1062 PP/05 DSCP support for outbound connections and control modifier for inbound.
1064 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1065 (Only plugin which currently uses this is kerberos4, which nobody should
1066 be using, but we should make it available and other future plugins might
1067 conceivably use it, even though it would break NAT; stuff *should* be
1068 using channel bindings instead).
1070 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1071 name; added for Sendmail compatibility; requires admin caller.
1072 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1073 just ignore it); requires trusted caller.
1074 Also parse but ignore: -Ac -Am -X<logfile>
1077 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1079 TL/02 Add +smtp_confirmation as a default logging option.
1081 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1082 Patch by Magnus Holmgren from 2007-02-20.
1084 TL/04 Bugzilla 1281 - Spec typo.
1085 Bugzilla 1283 - Spec typo.
1086 Bugzilla 1290 - Spec grammar fixes.
1088 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1090 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1092 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1093 mailing list by by Dmitry Isaikin.
1095 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1097 JH/02 Support "G" suffix to numbers in ${if comparisons.
1099 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1101 NM/01 Bugzilla 1197 - Spec typo
1102 Bugzilla 1196 - Spec examples corrections
1104 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1106 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1107 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1110 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1111 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1113 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1114 "acl {{name}{arg}...}", and optional args on acl condition
1117 JH/05 Permit multiple router/transport headers_add/remove lines.
1119 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1121 JH/07 Avoid using a waiting database for a single-message-only transport.
1122 Performance patch from Paul Fisher. Bugzilla 1262.
1124 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1127 JH/09 Add $headers_added variable, with content from use of ACL modifier
1128 add_header (but not yet added to the message). Bugzilla 199.
1130 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1131 Pulled from Bugzilla 817 by Wolfgang Breyha.
1133 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1135 (nb: this is the same fix as in Exim 4.80.1)
1137 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1140 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1143 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1145 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1146 advertise SMTP AUTH mechanism to us, instead of a generic
1147 protocol violation error. Also, make Exim more robust to bad
1148 data from the Dovecot auth socket.
1150 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1152 When a queue runner is handling a message, Exim first routes the
1153 recipient addresses, during which it prunes them based on the retry
1154 hints database. After that it attempts to deliver the message to
1155 any remaining recipients. It then updates the hints database using
1158 So if a recipient address works intermittently, it can get repeatedly
1159 deferred at routing time. The retry hints record remains fresh so the
1160 address never reaches the final cutoff time.
1162 This is a fairly common occurrence when a user is bumping up against
1163 their storage quota. Exim had some logic in its local delivery code
1164 to deal with this. However it did not apply to per-recipient defers
1165 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1167 This change adds a proper retry rule check during routing so that the
1168 final cutoff time is checked against the message's age. We only do
1169 this check if there is an address retry record and there is not a
1170 domain retry record; this implies that previous attempts to handle
1171 the address had the retry_use_local_parts option turned on. We use
1172 this as an approximation for the destination being like a local
1173 delivery, as in LMTP.
1175 I suspect this new check makes the old local delivery cutoff check
1176 redundant, but I have not verified this so I left the code in place.
1178 TF/02 Correct gecos expansion when From: is a prefix of the username.
1180 Test 0254 submits a message to Exim with the header
1184 When I ran the test suite under the user fanf2, Exim expanded
1185 the header to contain my full name, whereas it should have added
1186 a Resent-Sender: header. It erroneously treats any prefix of the
1187 username as equal to the username.
1189 This change corrects that bug.
1191 GF/01 DCC debug and logging tidyup
1192 Error conditions log to paniclog rather than rejectlog.
1193 Debug lines prefixed by "DCC: " to remove any ambiguity.
1195 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1197 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1198 Bug spotted by Jeremy Harris; was flawed since initial commit.
1199 Would have resulted in OCSP responses post-SNI triggering an Exim
1200 NULL dereference and crash.
1202 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1204 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1205 Bug detection, analysis and fix by Samuel Thibault.
1206 Bugzilla 1331, Debian bug #698092.
1208 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1210 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1211 Server implementation by Todd Lyons, client by JH.
1212 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1213 config variable "prdr_enable" controls whether the server
1214 advertises the facility. If the client requests PRDR a new
1215 acl_data_smtp_prdr ACL is called once for each recipient, after
1216 the body content is received and before the acl_smtp_data ACL.
1217 The client is controlled by both of: a hosts_try_prdr option
1218 on the smtp transport, and the server advertisement.
1219 Default client logging of deliveries and rejections involving
1220 PRDR are flagged with the string "PRDR".
1222 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1223 fclose(). Diagnosis by Todd Lyons.
1225 PP/17 Update configure.default to handle IPv6 localhost better.
1226 Patch by Alain Williams (plus minor tweaks).
1229 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1230 This is now consistent with GnuTLS, and is now documented: the
1231 previous undocumented portable approach to treating the option as
1232 unset was to force an expansion failure. That still works, and
1233 an empty string is now equivalent.
1235 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1236 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1237 not performing validation itself.
1239 PP/20 Added force_command boolean option to pipe transport.
1240 Patch from Nick Koston, of cPanel Inc.
1242 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1245 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1247 PP/21 Fix eximon continuous updating with timestamped log-files.
1248 Broken in a format-string cleanup in 4.80, missed when I repaired the
1249 other false fix of the same issue.
1250 Report and fix from Heiko Schlichting.
1253 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1254 Report from Prashanth Katuri.
1256 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1257 It's SecureTransport, so affects any MacOS clients which use the
1258 system-integrated TLS libraries, including email clients.
1260 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1261 using a MIME ACL for non-SMTP local injection.
1262 Report and assistance in diagnosis by Warren Baker.
1264 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1266 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1268 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1269 last id that failed so it may be referenced in subsequent ACL's.
1271 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1274 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1275 ldap library initialization, allowing self-signed CA's to be
1276 used. Also properly sets require_cert option later in code by
1277 using NULL (global ldap config) instead of ldap handle (per
1278 session). Bug diagnosis and testing by alxgomz.
1280 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1281 the src/util/ subdirectory.
1283 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1284 renamed to Transport Post Delivery Action by Jeremy Harris, as
1287 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1288 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1289 redis_servers = needs to be configured which will be used by the redis
1290 lookup. Patch from Warren Baker, of The Packet Hub.
1292 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1294 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1295 hostname or reverse DNS when processing a host list. Used suggestions
1296 from multiple comments on this bug.
1298 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1300 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1301 Missed a few lines, added it to make the runtest require no keyboard
1304 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1305 contains upper case chars. Make router use caseful_local_part.
1307 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1308 support when GnuTLS has been built with p11-kit.
1314 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1316 This, or similar/improved, will also be change PP/11 of 4.82.
1322 PP/01 Handle short writes when writing local log-files.
1323 In practice, only affects FreeBSD (8 onwards).
1324 Bugzilla 1053, with thanks to Dmitry Isaikin.
1326 NM/01 Bugzilla 949 - Documentation tweak
1328 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1331 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1333 PP/02 Implemented gsasl authenticator.
1335 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1337 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1338 `pkg-config foo` for cflags/libs.
1340 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1341 with rest of GSASL and with heimdal_gssapi.
1343 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1344 `pkg-config foo` for cflags/libs for the TLS implementation.
1346 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1347 properties get this fed in as external SSF. A number of robustness
1348 and debugging improvements to the cyrus_sasl authenticator.
1350 PP/08 cyrus_sasl server now expands the server_realm option.
1352 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1353 Patch by Jeremy Harris.
1355 PP/10 Added dbmjz lookup type.
1357 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1359 PP/12 MAIL args handles TAB as well as SP, for better interop with
1360 non-compliant senders.
1361 Analysis and variant patch by Todd Lyons.
1363 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1364 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1365 Patch from Dirk Mueller <dmueller@suse.com>
1367 PP/13 tls_peerdn now print-escaped for spool files.
1368 Observed some $tls_peerdn in wild which contained \n, which resulted
1369 in spool file corruption.
1371 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1372 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1373 or write after TLS renegotiation, which otherwise led to messages
1376 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1377 as a tracking header (ie: a signed header comes before the signature).
1378 Patch from Wolfgang Breyha.
1380 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1381 comma-sep list; embedded commas doubled.
1383 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1385 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1387 Report and patch from Dmitry Banschikov.
1389 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1390 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1391 failures. We appear to now support TLS1.1+ with Exim.
1393 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1394 lets Exim select keys and certificates based upon TLS SNI from client.
1395 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1396 before an outbound SMTP session. New log_selector, +tls_sni.
1398 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1399 NULL dereference. Report and patch from Alun Jones.
1401 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1402 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1403 Not seeing resolver debug output on NetBSD, but suspect this is a
1404 resolver implementation change.
1406 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1407 Left warnings. Added "eximon gdb" invocation mode.
1409 PP/21 Defaulting "accept_8bitmime" to true, not false.
1411 PP/22 Added -bw for inetd wait mode support.
1413 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1414 locate the relevant includes and libraries. Made this the default.
1416 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1417 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1419 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1420 This may cause build issues on older platforms.
1422 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1423 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1424 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1425 Added SNI support via GnuTLS too.
1426 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1428 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1430 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1431 Applied second patch from Janne, implementing suggestion to default
1432 multiple-strings-in-record handling to match SPF spec.
1434 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1436 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1437 read-only, out of scope).
1438 Patch from Wolfgang Breyha, report from Stuart Northfield.
1440 PP/29 Fix three issues highlighted by clang analyser static analysis.
1441 Only crash-plausible issue would require the Cambridge-specific
1442 iplookup router and a misconfiguration.
1443 Report from Marcin Mirosław.
1445 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1447 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1448 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1449 As part of this, removing so much warning spew let me fix some minor
1450 real issues in debug logging.
1452 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1453 assignment on my part. Fixed.
1455 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1456 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1457 Janne Snabb (who went above and beyond: thank you).
1459 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1460 string otherwise requires a connection and a bunch more work and it's
1461 relatively easy to get wrong. Should also expose TLS library linkage
1464 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1465 64-bit ${eval} (JH/03).
1467 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1468 GNU libc to support some of the 64-bit stuff, should not lead to
1469 conflicts. Defined before os.h is pulled in, so if a given platform
1470 needs to override this, it can.
1472 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1473 protection layer was required, which is not implemented.
1474 Bugzilla 1254, patch from Wolfgang Breyha.
1476 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1477 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1478 tls_dhparam take prime identifiers. Also unbreak combination of
1479 OpenSSL+DH_params+TLSSNI.
1481 PP/39 Disable SSLv2 by default in OpenSSL support.
1487 PP/01 Solaris build fix for Oracle's LDAP libraries.
1488 Bugzilla 1109, patch from Stephen Usher.
1490 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1492 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1495 TF/02 Fix a couple more cases where we did not log the error message
1496 when unlink() failed. See also change 4.74-TF/03.
1498 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1499 lock up or crash if it happened to be inside a call to libc when it
1500 got a SIGUSR1 from exiwhat.
1502 The SIGUSR1 handler appends the current process status to the process
1503 log which is later printed by exiwhat. It used to use the general
1504 purpose logging code to do this, but several functions it calls are
1505 not safe for signals.
1507 The new output code in the SIGUSR1 handler is specific to the process
1508 log, and simple enough that it's easy to inspect for signal safety.
1509 Removing some special cases also simplifies the general logging code.
1510 Removing the spurious timestamps from the process log simplifies
1513 TF/04 Improved ratelimit ACL condition.
1515 The /noupdate option has been deprecated in favour of /readonly which
1516 has clearer semantics. The /leaky, /strict, and /readonly update modes
1517 are mutually exclusive. The update mode is no longer included in the
1518 database key; it just determines when the database is updated. (This
1519 means that when you upgrade Exim will forget old rate measurements.)
1521 Exim now checks that the per_* options are used with an update mode that
1522 makes sense for the current ACL. For example, when Exim is processing a
1523 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1524 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1525 must specify per_mail/readonly. If you omit the update mode it defaults to
1526 /leaky where that makes sense (as before) or /readonly where required.
1528 The /noupdate option is now undocumented but still supported for
1529 backwards compatibility. It is equivalent to /readonly except that in
1530 ACLs where /readonly is required you may specify /leaky/noupdate or
1531 /strict/noupdate which are treated the same as /readonly.
1533 A useful new feature is the /count= option. This is a generalization
1534 of the per_byte option, so that you can measure the throughput of other
1535 aggregate values. For example, the per_byte option is now equivalent
1536 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1538 The per_rcpt option has been generalized using the /count= mechanism
1539 (though it's more complicated than the per_byte equivalence). When it is
1540 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1541 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1542 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1543 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1544 Note that using per_rcpt with a non-readonly update mode in more than
1545 one ACL will cause the recipients to be double-counted. (The per_mail
1546 and per_byte options don't have this problem.)
1548 The handling of very low rates has changed slightly. If the computed rate
1549 is less than the event's count (usually one) then this event is the first
1550 after a long gap. In this case the rate is set to the same as this event's
1551 count, so that the first message of a spam run is counted properly.
1553 The major new feature is a mechanism for counting the rate of unique
1554 events. The new per_addr option counts the number of different
1555 recipients that someone has sent messages to in the last time period. It
1556 behaves like per_rcpt if all the recipient addresses are different, but
1557 duplicate recipient addresses do not increase the measured rate. Like
1558 the /count= option this is a general mechanism, so the per_addr option
1559 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1560 example, measure the rate that a client uses different sender addresses
1561 with the options per_mail/unique=$sender_address. There are further
1562 details in the main documentation.
1564 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1566 TF/06 Removed a few PCRE remnants.
1568 TF/07 Automatically extract Exim's version number from tags in the git
1569 repository when doing development or release builds.
1571 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1572 Bugzilla 879. Patch from Paul Fisher.
1574 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1575 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1578 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1580 PP/05 Variable $av_failed, true if the AV scanner deferred.
1581 Bugzilla 1078. Patch from John Horne.
1583 PP/06 Stop make process more reliably on build failure.
1584 Bugzilla 1087. Patch from Heiko Schlittermann.
1586 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1587 Bugzilla 1089. Patch from Heiko Schlittermann.
1589 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1590 Bugzilla 1131. Patch from Holger Weiß.
1592 PP/09 Handle IPv6 addresses with SPF.
1593 Bugzilla 860. Patch from Wolfgang Breyha.
1595 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1597 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1600 PP/11 match_* no longer expand right-hand-side by default.
1601 New compile-time build option, EXPAND_LISTMATCH_RHS.
1602 New expansion conditions, "inlist", "inlisti".
1604 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1606 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1608 PP/14 fix log_write() format string regression from TF/03.
1609 Bugzilla 1152. Patch from Dmitry Isaikin.
1615 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1617 PP/02 Harmonised TLS library version reporting; only show if debugging.
1618 Layout now matches that introduced for other libraries in 4.74 PP/03.
1620 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1622 PP/04 New "dns_use_edns0" global option.
1624 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1627 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1628 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1630 TK/01 Updated PolarSSL code to 0.14.2.
1631 Bugzilla 1097. Patch from Andreas Metzler.
1633 PP/07 Catch divide-by-zero in ${eval:...}.
1634 Fixes bugzilla 1102.
1636 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1639 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1640 format-string attack -- SECURITY: remote arbitrary code execution.
1642 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1643 time unintentionally subject to list matching rules, letting the header
1644 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1645 arbitrary string expansion). This allowed for information disclosure.
1647 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1648 INT_MIN/-1 -- value coerced to INT_MAX.
1654 NM/01 Workaround for PCRE version dependency in version reporting
1657 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1658 This fixes portability to compilers other than gcc, notably
1659 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1661 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1662 makefiles for portability to HP-UX and POSIX correctness.
1664 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1665 Also via indented variable definition in the Makefile.
1666 (Debugging by Oliver Heesakkers).
1668 PP/02 Restore caching of spamd results with expanded spamd_address.
1669 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1671 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1672 Improves build reliability. Fix from: Frank Elsner
1674 NM/02 Fix wide character breakage in the rfc2047 coding
1675 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1677 NM/03 Allow underscore in dnslist lookups
1678 Fixes bug 1026. Patch from Graeme Fowler
1680 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1681 Code patches from Adam Ciarcinski of NetBSD.
1683 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1686 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1687 is logged, to avoid truncation. Patch from John Horne.
1689 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1690 Patch from Jakob Hirsch.
1692 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1693 SQL string expansion failure details.
1694 Patch from Andrey Oktyabrski.
1696 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1697 Patch from Simon Arlott.
1699 PP/09 New lookups functionality failed to compile on old gcc which rejects
1700 extern declarations in function scope.
1701 Patch from Oliver Fleischmann
1703 PP/10 Use sig_atomic_t for flags set from signal handlers.
1704 Check getgroups() return and improve debugging.
1705 Fixed developed for diagnosis in bug 927 (which turned out to be
1708 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1709 Patch from Mark Zealey.
1711 PP/12 Bugzilla 1056: Improved spamd server selection.
1712 Patch from Mark Zealey.
1714 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1715 Based on patch from Heiko Schlittermann.
1717 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1718 Patch from Uwe Doering, sign-off by Michael Haardt.
1720 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1721 variable declaration deep within a block. Bug and patch from
1724 PP/15 lookups-Makefile IRIX compatibility coercion.
1726 PP/16 Make DISABLE_DKIM build knob functional.
1728 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1729 Patch by Simon Arlott
1731 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1732 variable argument macros. Our copy now differs from upstream.
1738 TF/01 Failure to get a lock on a hints database can have serious
1739 consequences so log it to the panic log.
1741 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1742 controlled using the smtp_confirmation log selector.
1744 TF/03 Include the error message when we fail to unlink a spool file.
1746 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1747 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1748 for maintaining out-of-tree patches for some time.
1750 PP/01 Bugzilla 139: Documentation and portability issues.
1751 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1752 Handle per-OS dynamic-module compilation flags.
1754 PP/02 Let /dev/null have normal permissions.
1755 The 4.73 fixes were a little too stringent and complained about the
1756 permissions on /dev/null. Exempt it from some checks.
1757 Reported by Andreas M. Kirchwitz.
1759 PP/03 Report version information for many libraries, including
1760 Exim version information for dynamically loaded libraries. Created
1761 version.h, now support a version extension string for distributors
1762 who patch heavily. Dynamic module ABI change.
1764 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1765 privilege escalation vulnerability whereby the Exim run-time user
1766 can cause root to append content of the attacker's choosing to
1769 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1772 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1773 If dropping privileges for untrusted macros, we disabled normal logging
1774 on the basis that it would fail; for the Exim run-time user, this is not
1775 the case, and it resulted in successful deliveries going unlogged.
1776 Fixed. Reported by Andreas Metzler.
1782 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1783 only prepend for the Resent-* case. Fixes regression introduced in
1784 Exim 4.70 by NM/22 for Bugzilla 607.
1786 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1787 increasing numbers of administrators be bitten by this.
1789 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1791 PP/03 Bugzilla 994: added openssl_options main configuration option.
1793 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1795 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1797 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1799 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1800 without a peer certificate, leading to a segfault because of an
1801 assumption that peers always have certificates. Be a little more
1802 paranoid. Problem reported by Martin Tscholak.
1804 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1805 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1806 NB: ClamAV planning to remove STREAM in "middle of 2010".
1807 CL also introduces -bmalware, various -d+acl logging additions and
1808 more caution in buffer sizes.
1810 PP/09 Implemented reverse_ip expansion operator.
1812 PP/10 Bugzilla 937: provide a "debug" ACL control.
1814 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1816 PP/12 Bugzilla 973: Implement --version.
1818 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1820 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1822 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1824 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1825 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1826 ignore trailing whitespace.
1828 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1830 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1833 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1834 Notification from Dr Andrew Aitchison.
1836 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1837 ExtendedDetectionInfo response format.
1838 Notification from John Horne.
1840 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1843 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1844 XSL and documented dependency on system catalogs, with examples of how
1847 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1850 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1851 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1852 configuration file which is writeable by the Exim user or group.
1854 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1855 of configuration files to cover files specified with the -C option if
1856 they are going to be used with root privileges, not just the default
1859 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1860 option (effectively making it always true).
1862 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1863 files to be used while preserving root privileges.
1865 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1866 that rogue child processes cannot use them.
1868 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1869 run-time user, instead of root.
1871 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1872 Exim run-time user without dropping privileges.
1874 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1875 result string, instead of calling string_vformat() twice with the same
1878 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1879 for other users. Others should always drop root privileges if they use
1880 -C on the command line, even for a whitelisted configure file.
1882 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1884 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1890 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1891 $data_path, and $header_path variables; fixed documentation bugs and
1894 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1895 exipick to access non-standard spools, including the "frozen" queue
1898 NM/01 Bugzilla 965: Support mysql stored procedures.
1899 Patch from Alain Williams
1901 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1903 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1904 Patch from Andreas Metzler
1906 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1907 Patch from Kirill Miazine
1909 NM/05 Bugzilla 671: Added umask to procmail example.
1911 JJ/03 installed exipick 20100323.0, fixing doc bug
1913 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1914 directory. Notification and patch from Dan Rosenberg.
1916 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1918 TK/02 Improve log output when DKIM signing operation fails.
1920 MH/01 Treat the transport option dkim_domain as a colon separated
1921 list, not as a single string, and sign the message with each element,
1922 omitting multiple occurences of the same signer.
1924 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1925 Bugzilla 985, 986. Patch by Simon Arlott
1927 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1928 Patch by Simon Arlott
1930 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1931 MBX locking. Notification from Dan Rosenberg.
1937 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1939 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1941 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1943 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1945 NM/04 Bugzilla 915: Flag broken perl installation during build.
1951 TK/01 Added patch by Johannes Berg that expands the main option
1952 "spamd_address" if it starts with a dollar sign.
1954 TK/02 Write list of recipients to X-Envelope-Sender header when building
1955 the mbox-format spool file for content scanning (suggested by Jakob
1958 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1959 (http://www.dcc-servers.net/) support via dccifd. Activated by
1960 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1962 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1963 by Mark Daniel Reidel <mr@df.eu>.
1965 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1966 When building exim an external PCRE library is now needed -
1967 PCRE is a system library on the majority of modern systems.
1968 See entry on PCRE_LIBS in EDITME file.
1970 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1971 conversation. Added nologin parameter to request.
1972 Patch contributed by Kirill Miazine.
1974 TF/01 Do not log submission mode rewrites if they do not change the address.
1976 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1978 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1979 log files in place. Contributed by Roberto Lima.
1981 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1983 TF/03 Bugzilla 615: When checking the local_parts router precondition
1984 after a local_part_suffix or local_part_prefix option, Exim now
1985 does not use the address's named list lookup cache, since this
1986 contains cached lookups for the whole local part.
1988 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1989 Robert Millan. Documentation is in experimental-spec.txt.
1991 TF/04 Bugzilla 668: Fix parallel build (make -j).
1993 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1995 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1996 Patch provided by Jan Srzednicki.
1998 TF/05 Leading white space used to be stripped from $spam_report which
1999 wrecked the formatting. Now it is preserved.
2001 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2002 that they are available at delivery time.
2004 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2006 TF/08 TLS error reporting now respects the incoming_interface and
2007 incoming_port log selectors.
2009 TF/09 Produce a more useful error message if an SMTP transport's hosts
2010 setting expands to an empty string.
2012 NM/06 Bugzilla 744: EXPN did not work under TLS.
2013 Patch provided by Phil Pennock.
2015 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2016 Patch provided by Richard Godbee.
2018 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2019 acl_smtp_notquit, added index entry.
2021 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2022 Patch provided by Eugene Bujak.
2024 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2025 accept(). Patch provided by Maxim Dounin.
2027 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2028 Patch provided by Phil Pennock.
2030 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2032 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2033 Patch provided by Brad "anomie" Jorsch.
2035 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2036 Patch provided by Dean Brooks.
2038 TK/05 Add native DKIM support (does not depend on external libraries).
2040 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2041 Patch provided by Graeme Fowler.
2043 NM/16 Bugzilla 851: Documentation example syntax fix.
2045 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2047 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2050 NM/19 Bugzilla 745: TLS version reporting.
2051 Patch provided by Phil Pennock.
2053 NM/20 Bugzilla 167: bool: condition support.
2054 Patch provided by Phil Pennock.
2056 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2057 clients. Patch provided by Phil Pennock.
2059 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2060 Patch provided by Brad "anomie" Jorsch.
2062 NM/23 Bugzilla 687: Fix misparses in eximstats.
2063 Patch provided by Heiko Schlittermann.
2065 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2066 Patch provided by Heiko Schlittermann.
2068 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2069 plus update to original patch.
2071 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2073 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2074 Patch provided by David Brownlee.
2076 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2078 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2080 NM/30 Bugzilla 888: TLS documentation bugfixes.
2082 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2084 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2085 Unlike the original bugzilla I have changed all shell scripts in src tree.
2087 NM/33 Bugzilla 898: Transport filter timeout fix.
2088 Patch by Todd Rinaldo.
2090 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2091 Patch by Serge Demonchaux.
2093 NM/35 Bugzilla 39: Base64 decode bug fixes.
2094 Patch by Jakob Hirsch.
2096 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2098 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2100 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2102 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2108 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2109 ALT-N's libdkim that I have put here:
2110 http://duncanthrax.net/exim-experimental/
2112 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2113 were called 'true' and it seems that C99 defines that as a reserved
2114 keyword to be used with 'bool' variable types. That means you could
2115 not include C99-style headers which use bools without triggering
2116 build errors in sieve.c.
2118 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2119 as mailq or other aliases. Changed the --help handling significantly
2120 to do whats expected. exim_usage() emits usage/help information.
2122 SC/01 Added the -bylocaldomain option to eximstats.
2124 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2126 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2128 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2134 PH/01 Another patch from the Sieve maintainer.
2136 PH/02 When an IPv6 address is converted to a string for single-key lookup
2137 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2138 dots are used instead of colons so that keys in lsearch files need not
2139 contain colons. This was done some time before quoting was made available
2140 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2141 (notated using the quote facility) so as to distinguish them from IPv4
2142 keys. This meant that lookups for IP addresses in host lists did not work
2143 for iplsearch lookups.
2145 This has been fixed by arranging for IPv6 addresses to be expressed with
2146 colons if the lookup type is iplsearch. This is not incompatible, because
2147 previously such lookups could never work.
2149 The situation is now rather anomalous, since one *can* have colons in
2150 ordinary lsearch keys. However, making the change in all cases is
2151 incompatible and would probably break a number of configurations.
2153 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2156 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2157 conversion specification without a maximum field width, thereby enabling
2158 a rogue spamd server to cause a buffer overflow. While nobody in their
2159 right mind would setup Exim to query an untrusted spamd server, an
2160 attacker that gains access to a server running spamd could potentially
2161 exploit this vulnerability to run arbitrary code as the Exim user.
2163 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2164 $primary_hostname instead of what libspf2 thinks the hosts name is.
2166 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2167 a directory entry by the name of the lookup key. Previously, if a
2168 symlink pointed to a non-existing file or a file in a directory that
2169 Exim lacked permissions to read, a lookup for a key matching that
2170 symlink would fail. Now it is enough that a matching directory entry
2171 exists, symlink or not. (Bugzilla 503.)
2173 PH/03 The body_linecount and body_zerocount variables are now exported in the
2176 PH/04 Added the $dnslist_matched variable.
2178 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2179 This means they are set thereafter only if the connection becomes
2182 PH/06 Added the client_condition to authenticators so that some can be skipped
2183 by clients under certain conditions.
2185 PH/07 The error message for a badly-placed control=no_multiline_responses left
2186 "_responses" off the end of the name.
2188 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2190 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2191 (without spaces) instead of just copying the configuration text.
2193 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2195 PH/11 Added $max_received_linelength.
2197 PH/12 Added +ignore_defer and +include_defer to host lists.
2199 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2200 way in which PCRE > 7.0 is built.
2202 PH/14 Implemented queue_only_load_latch.
2204 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2205 MAIL command. The effect was to mangle the value on 64-bit systems.
2207 PH/16 Another patch from the Sieve maintainer.
2209 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2211 PH/18 If a system quota error occurred while trying to create the file for
2212 a maildir delivery, the message "Mailbox is full" was not appended to the
2213 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2214 only to a quota excession during the actual writing of the file.
2216 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2217 characters?) which causes problems in log lines. The DN values are now
2218 passed through string_printing() before being added to log lines.
2220 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2221 and InterBase are left for another time.)
2223 PH/21 Added message_body_newlines option.
2225 PH/22 Guard against possible overflow in moan_check_errorcopy().
2227 PH/23 POSIX allows open() to be a macro; guard against that.
2229 PH/24 If the recipient of an error message contained an @ in the local part
2230 (suitably quoted, of course), incorrect values were put in $domain and
2231 $local_part during the evaluation of errors_copy.
2237 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2238 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2241 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2242 issue a MAIL command.
2244 PH/02 In an ACL statement such as
2246 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2248 if a client was not listed at all, or was listed with a value other than
2249 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2250 the condition was not true (as it should be), so access was not denied.
2251 The bug was that the ! inversion was incorrectly passed on to the second
2252 item. This has been fixed.
2254 PH/03 Added additional dnslists conditions == and =& which are different from
2255 = and & when the dns lookup returns more than one IP address.
2257 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2258 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2260 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2261 FSYNC, which compiles an option called disable_fsync that allows for
2262 bypassing fsync(). The documentation is heavily laced with warnings.
2264 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2266 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2267 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2268 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2269 including adding "make clean"; (3) Added -fPIC when compiling the test
2270 dynamically loaded module, to get rid of a warning.
2272 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2273 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2274 = 0s. The bug is otherwise harmless.
2276 PH/07 There was a bug in the dovecot authenticator such that the value of
2277 $auth1 could be overwritten, and so not correctly preserved, after a
2278 successful authentication. This usually meant that the value preserved by
2279 the server_setid option was incorrect.
2281 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2283 PH/09 Installed PCRE release 7.0.
2285 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2286 run for batched SMTP input. It is now run at the start of every message
2287 in the batch. While fixing this I discovered that the process information
2288 (output by running exiwhat) was not always getting set for -bs and -bS
2289 input. This is fixed, and it now also says "batched" for BSMTP.
2291 PH/11 Added control=no_pipelining.
2293 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2294 patch, slightly modified), and move the expansion of helo_data till after
2295 the connection is made in the smtp transport (so it can use these
2298 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2300 PH/14 Added log_selector = +pid.
2302 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2304 PH/16 Add ${if forany and ${if forall.
2306 PH/17 Added dsn_from option to vary the From: line in DSNs.
2308 PH/18 Flush SMTP output before performing a callout, unless control =
2309 no_callout_flush is set.
2311 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2312 was true (the default) a successful delivery failed to delete the retry
2313 item, thus causing premature timeout of the address. The bug is now
2316 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2318 PH/21 Long custom messages for fakedefer and fakereject are now split up
2319 into multiline responses in the same way that messages for "deny" and
2320 other ACL rejections are.
2322 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2323 with slight modification.
2325 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2326 draft, changing the syntax and factoring some duplicate code".
2328 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2329 for deliveries of the second and subsequent messages over the same SMTP
2332 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2333 ${reduce, with only minor "tidies".
2335 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2337 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2338 expansion side effects.
2340 PH/27 When a message times out after an over-quota error from an Exim-imposed
2341 quota, the bounce message says "mailbox is full". This message was not
2342 being given when it was a system quota that was exceeded. It now should
2345 MH/03 Made $recipients available in local_scan(). local_scan() already has
2346 better access to the recipient list through recipients_list[], but
2347 $recipients can be useful in postmaster-provided expansion strings.
2349 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2350 in the case of a MAIL command with additional options following the
2351 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2352 were accidentally chopped off.
2354 PH/29 SMTP synchronization checks are implemented when a command is read -
2355 there is a check that no more input is waiting when there shouldn't be
2356 any. However, for some commands, a delay in an ACL can mean that it is
2357 some time before the response is written. In this time, more input might
2358 arrive, invalidly. So now there are extra checks after an ACL has run for
2359 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2360 pipelining has not been advertised.
2362 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2364 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2365 returned in the SMTP response when smtp_return_error_details was false.
2366 This has been fixed.
2368 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2369 instead of the C I/O that was originally supplied, because problems were
2370 reported on Solaris.
2372 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2373 Exim which did not show up earlier: it was assuming that a call to
2374 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2375 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2376 was a macro that became an assignment, so it seemed to work. This has
2377 changed to a proper function call with a void return, hence the compile
2378 error. Exim's code has been fixed.
2380 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2383 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2385 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2387 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2388 because it turns out to be too big or there is a timeout) there is no
2389 "Completed" line in the log. When some messages of this type were
2390 selected by exigrep, they were listed as "not completed". Others were
2391 picked up by some special patterns. I have improved the selection
2392 criteria to be more general.
2394 PH/38 The host_find_failed option in the manualroute router can now be set
2395 to "ignore", to completely ignore a host whose IP address cannot be
2396 found. If all hosts are ignored, the behaviour is controlled by the new
2397 host_all_ignored option.
2399 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2400 homing or because of multiple MX records with /mx) generated more than
2401 one IP address, and the following item turned out to be the local host,
2402 all the secondary addresses of the first item were incorrectly removed
2403 from the list, along with the local host and any following hosts (which
2404 is what is supposed to happen).
2406 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2407 whoever called Exim into the -H file. In the case of the daemon it was
2408 behaving confusingly. When first started, it used values for whoever
2409 started the daemon, but after a SIGHUP it used the Exim user (because it
2410 calls itself on a restart). I have changed the code so that it now always
2413 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2414 message are rejected with the same error (e.g. no authentication or bad
2415 sender address), and a DATA command is nevertheless sent (as can happen
2416 with PIPELINING or a stupid MUA), the error message that was given to the
2417 RCPT commands is included in the rejection of the DATA command. This is
2418 intended to be helpful for MUAs that show only the final error to their
2421 PH/42 Another patch from the Sieve maintainer.
2423 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2424 Eximstats - Fixed some broken HTML links and added missing column headers
2426 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2427 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2429 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2431 PH/43 Yet another patch from the Sieve maintainer.
2433 PH/44 I found a way to check for a TCP/IP connection going away before sending
2434 the response to the final '.' that terminates a message, but only in the
2435 case where the client has not sent further data following the '.'
2436 (unfortunately, this is allowed). However, in many cases there won't be
2437 any further data because there won't be any more messages to send. A call
2438 to select() can be used: if it shows that the input is "ready", there is
2439 either input waiting, or the socket has been closed. An attempt to read
2440 the next input character can distinguish the two cases. Previously, Exim
2441 would have sent an OK response which the client would never have see.
2442 This could lead to message repetition. This fix should cure that, at
2443 least in a lot of common cases.
2445 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2446 advertised in response to EHLO.
2452 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2453 fixed by 4.65/MH/01 (is this a record?) are fixed:
2455 (i) An empty string was always treated as zero by the numeric comparison
2456 operators. This behaviour has been restored.
2458 (ii) It is documented that the numeric comparison operators always treat
2459 their arguments as decimal numbers. This was broken in that numbers
2460 starting with 0 were being interpreted as octal.
2462 While fixing these problems I realized that there was another issue that
2463 hadn't been noticed. Values of message_size_limit (both the global option
2464 and the transport option) were treated as octal if they started with 0.
2465 The documentation was vague. These values are now always treated as
2466 decimal, and I will make that clear in the documentation.
2472 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2473 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2476 MH/01 Don't check that the operands of numeric comparison operators are
2477 integers when their expansion is in "skipping" mode (fixes bug
2478 introduced by 4.64-PH/07).
2480 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2481 child addresses, Exim now panics and dies. Previously, because the count
2482 is held in a short int, deliveries were likely to be lost. As such a
2483 large number of recipients for a single message is ridiculous
2484 (performance will be very, very poor), I have chosen to impose a limit
2485 rather than extend the field.
2491 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2492 leftover -K file (the existence of which was triggered by #402).
2493 While we were at it, introduced process PID as part of the -K
2494 filename. This should rule out race conditions when creating
2497 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2498 processing considerably. Previous code took too long for large mails,
2499 triggering a timeout which in turn triggers #401.
2501 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2502 in the DK code in transports.c. sendfile() is not really portable,
2503 hence the _LINUX specificness.
2505 TF/01 In the add_headers option to the mail command in an Exim filter,
2506 there was a bug that Exim would claim a syntax error in any
2507 header after the first one which had an odd number of characters
2510 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2511 callout verification, Exim cached a "reject" for the entire domain. This
2512 is correct for most verifications, but it is not correct for a recipient
2513 verification with use_sender or use_postmaster set, because in that case
2514 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2515 case of MAIL FROM:<> rejection from other early rejections (e.g.
2516 rejection of HELO). When verifying a recipient using a non-null MAIL
2517 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2518 Whatever the result of the callout, the value of the domain cache is
2519 left unchanged (for any other kind of callout, getting as far as trying
2520 RCPT means that the domain itself is ok).
2522 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2525 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2526 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2527 session, the daemon ignores ECONNECT errors and logs others; it now
2528 ignores EPIPE as well.
2530 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2531 (quoted-printable decoding).
2533 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2534 later the small subsequent patch to fix an introduced bug.
2536 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2538 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2540 PH/08 An error is now given if message_size_limit is specified negative.
2542 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2543 to be given (somewhat) arbitrary names.
2545 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2548 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2549 miscellaneous code fixes
2551 PH/10 Added the log_reject_target ACL modifier to specify where to log
2554 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2555 hostname. This is wrong, because it relates to the incoming message (and
2556 probably the interface on which it is arriving) and not to the outgoing
2557 callout (which could be using a different interface). This has been
2558 changed to use the value of the helo_data option from the smtp transport
2559 instead - this is what is used when a message is actually being sent. If
2560 there is no remote transport (possible with a router that sets up host
2561 addresses), $smtp_active_hostname is used.
2563 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2564 tweaks were necessary in order to get it to work (see also 21 below):
2565 (a) The code assumed that strncpy() returns a negative number on buffer
2566 overflow, which isn't the case. Replaced with Exim's string_format()
2568 (b) There were several signed/unsigned issues. I just did the minimum
2569 hacking in of casts. There is scope for a larger refactoring.
2570 (c) The code used strcasecmp() which is not a standard C function.
2571 Replaced with Exim's strcmpic() function.
2572 (d) The code set only $1; it now sets $auth1 as well.
2573 (e) A simple test gave the error "authentication client didn't specify
2574 service in request". It would seem that Dovecot has changed its
2575 interface. Fortunately there's a specification; I followed it and
2576 changed what the client sends and it appears to be working now.
2578 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2581 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2582 address A is aliased to B and C, where B exists and C does not. Without
2583 -v the output is "A verified" because verification stops after a
2584 successful redirection if more than one address is generated. However,
2585 with -v the child addresses are also verified. Exim was outputting "A
2586 failed to verify" and then showing the successful verification for C,
2587 with its parentage. It now outputs "B failed to verify", showing B's
2588 parentage before showing the successful verification of C.
2590 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2591 look up a TXT record in a specific list after matching in a combined
2594 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2595 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2596 they consult the DNS. I had assumed they would set it the way they
2597 wanted; and indeed my experiments on Linux seem to show that in some
2598 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2599 To be on the safe side, however, I have now made the interface to
2600 host_find_byname() similar to host_find_bydns(), with an argument
2601 containing the DNS resolver options. The host_find_byname() function now
2602 sets these options at its start, just as host_find_bydns() does. The smtp
2603 transport options dns_qualify_single and dns_search_parents are passed to
2604 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2605 of host_find_byname() use the default settings of RES_DEFNAMES
2606 (qualify_single) but not RES_DNSRCH (search_parents).
2608 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2609 spool_read_header() do less string testing, by means of a preliminary
2610 switch on the second character of optional "-foo" lines. (This is
2611 overdue, caused by the large number of possibilities that now exist.
2612 Originally there were few.) While I was there, I also converted the
2613 str(n)cmp tests so they don't re-test the leading "-" and the first
2614 character, in the hope this might squeeze out yet more improvement.
2616 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2617 flag allowing group syntax was set by the header_syntax check but not
2618 turned off, possible causing trouble later; (2) The flag was not being
2619 set at all for the header_verify test, causing "group"-style headers to
2620 be rejected. I have now set it in this case, and also caused header_
2621 verify to ignore an empty address taken from a group. While doing this, I
2622 came across some other cases where the code for allowing group syntax
2623 while scanning a header line wasn't quite right (mostly, not resetting
2624 the flag correctly in the right place). These bugs could have caused
2625 trouble for malformed header lines. I hope it is now all correct.
2627 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2628 with the "reply" argument non-NULL. The code, however (which originally
2629 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2630 but it didn't always do it. This confused somebody who was copying the
2631 code for some other use. I have removed all the tests.
2633 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2634 feature that was used to support insecure browsers during the U.S. crypto
2635 embargo. It requires special client support, and Exim is probably the
2636 only MTA that supported it -- and would never use it because real RSA is
2637 always available. This code has been removed, because it had the bad
2638 effect of slowing Exim down by computing (never used) parameters for the
2639 RSA_EXPORT functionality.
2641 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2642 authenticator to fail if there's a tab character in the incoming data
2643 (there should never be unless someone is messing about, as it's supposed
2644 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2645 if the connection is using TLS or if the remote IP is the same as the
2646 local IP, and the "valid-client-cert option" if a client certificate has
2649 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2650 authenticators. This can be used for authorization after authentication
2651 succeeds. (In the case of plaintext, it servers for both authentication
2654 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2655 if any retry times were supplied.
2657 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2658 connection, where there is no client IP address to check. In this
2659 situation, the verify now always succeeds.
2661 PH/25 Applied John Jetmore's -Mset patch.
2663 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2665 PH/27 In a string expansion for a processed (not raw) header when multiple
2666 headers of the same name were present, leading whitespace was being
2667 removed from all of them, but trailing whitespace was being removed only
2668 from the last one. Now trailing whitespace is removed from each header
2669 before concatenation. Completely empty headers in a concatenation (as
2670 before) are ignored.
2672 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2673 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2675 PH/29 [Removed. This was a change that I later backed out, and forgot to
2676 correct the ChangeLog entry (that I had efficiently created) before
2677 committing the later change.]
2679 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2680 address errors (4xx response to RCPT) over the same connection as other
2681 messages routed to the same hosts. Such deliveries are always "forced",
2682 so retry times are not inspected. This resulted in far too many retries
2683 for the affected addresses. The effect occurred only when there were more
2684 hosts than the hosts_max_try setting in the smtp transport when it had
2685 the 4xx errors. Those hosts that it had tried were not added to the list
2686 of hosts for which the message was waiting, so if all were tried, there
2687 was no problem. Two fixes have been applied:
2689 (i) If there are any address or message errors in an SMTP delivery, none
2690 of the hosts (tried or untried) are now added to the list of hosts
2691 for which the message is waiting, so the message should not be a
2692 candidate for sending over the same connection that was used for a
2693 successful delivery of some other message. This seems entirely
2694 reasonable: after all the message is NOT "waiting for some host".
2695 This is so "obvious" that I'm not sure why it wasn't done
2696 previously. Hope I haven't missed anything, but it can't do any
2697 harm, as the worst effect is to miss an optimization.
2699 (ii) If, despite (i), such a delivery is accidentally attempted, the
2700 routing retry time is respected, so at least it doesn't keep
2701 hammering the server.
2703 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2704 in ${readsocket because some servers need this prod.
2706 PH/32 Added some extra debug output when updating a wait-xxx database.
2708 PH/33 The hint "could be header name not terminated by colon", which has been
2709 given for certain expansion errors for a long time, was not being given
2710 for the ${if def:h_colon_omitted{... case.
2712 PH/34 The spec says: "With one important exception, whenever a domain list is
2713 being scanned, $domain contains the subject domain." There was at least
2714 one case where this was not true.
2716 PH/35 The error "getsockname() failed: connection reset by peer" was being
2717 written to the panic log as well as the main log, but it isn't really
2718 panic-worthy as it just means the connection died rather early on. I have
2719 removed the panic log writing for the ECONNRESET error when getsockname()
2722 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2723 runs only) independently of the message's sender address. This meant
2724 that, if the 4xx error was in fact related to the sender, a different
2725 message to the same recipient with a different sender could confuse
2726 things. In particular, this can happen when sending to a greylisting
2727 server, but other circumstances could also provoke similar problems.
2728 I have changed the default so that the retry time for these errors is now
2729 based a combination of the sender and recipient addresses. This change
2730 can be overridden by setting address_retry_include_sender=false in the
2733 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2734 remote server are returned as part of bounce messages. This was not
2735 happening for LMTP over a pipe (the lmtp transport), but now it is the
2736 same for both kinds of LMTP.
2738 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2739 in header lines that were in fact CNAMEs. This is no longer the case.
2741 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2742 and queue runs started by the daemon processed all messages. This has
2743 been fixed so that -R and -S can now usefully be given with -q<time>.
2745 PH/40 Import PCRE release 6.7 (fixes some bugs).
2747 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2749 PH/42 Give an error if -q is specified more than once.
2751 PH/43 Renamed the variables $interface_address and $interface_port as
2752 $received_ip_address and $received_port, to make it clear that these
2753 values apply to message reception, and not to the outgoing interface when
2754 a message is delivered. (The old names remain recognized, of course.)
2756 PH/44 There was no timeout on the connect() call when using a Unix domain
2757 socket in the ${readsocket expansion. There now is.
2759 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2760 be meaningful with "accept".
2762 SC/01 Eximstats V1.43
2763 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2765 SC/02 Eximstats V1.44
2766 Use a glob alias rather than an array ref in the generated
2767 parser. This improves both readability and performance.
2769 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2770 Collect SpamAssassin and rejection statistics.
2771 Don't display local sender or destination tables unless
2772 there is data to show.
2773 Added average volumes into the top table text output.
2775 SC/04 Eximstats V1.46
2776 Collect data on the number of addresses (recipients)
2777 as well as the number of messages.
2779 SC/05 Eximstats V1.47
2780 Added 'Message too big' to the list of mail rejection
2781 reasons (thanks to Marco Gaiarin).
2783 SC/06 Eximstats V1.48
2784 Mainlog lines which have GMT offsets and are too short to
2785 have a flag are now skipped.
2787 SC/07 Eximstats V1.49 (Alain Williams)
2788 Added the -emptyok flag.
2790 SC/08 Eximstats V1.50
2791 Fixes for obtaining the IP address from reject messages.
2793 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2794 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2795 whitespace changes from 4.64-PH/27
2797 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2800 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2801 are found, allow negative numbers in numeric criteria)
2803 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2805 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2806 to match changes made in 4.64-PH/43
2808 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2810 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2811 "domain" is actually an IP address, and force a failure. This locks out
2812 those revolvers/nameservers that support "A-for-A" lookups, in
2813 contravention of the specifications.
2815 PH/48 When a host name was looked up from an IP address, and the subsequent
2816 forward lookup of the name timed out, the host name was left in
2817 $sender_host_name, contrary to the specification.
2819 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2820 restricted to single-key lookups, Exim was not diagnosing an error if
2821 * or *@ was used with a query-style lookup.
2823 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2825 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2826 long ago, but noone interested enough thought of it. Let's just say that
2827 the "1.1" means that there are some new functions that weren't there at
2828 some point in the past.
2830 PH/51 Error processing for expansion failure of helo_data from an smtp
2831 transport during callout processing was broken.
2833 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2834 tested/used via the -bh/-bhc/-bs options.
2836 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2837 bug, fixed in subsequent PCRE releases).
2839 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2840 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2842 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2848 SC/01 Use a glob alias rather than an array ref in eximstats generated
2849 parser. This improves both readability and performance.
2851 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2852 Don't display local sender or destination tables in eximstats unless
2853 there is data to show.
2854 Added average volumes into the eximstats top table text output.
2856 SC/03 Collect data on the number of addresses (recipients) as well
2857 as the number of messages in eximstats.
2859 TF/01 Correct an error in the documentation for the redirect router. Exim
2860 does (usually) call initgroups() when daemonizing.
2862 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2863 with consistent privilege compared to when running as a daemon.
2865 TF/03 Note in the spec that $authenticated_id is not set for local
2866 submissions from trusted users.
2868 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2869 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2871 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2872 by adding some example configuration directives to the default
2873 configuration file. A little bit of work is required to uncomment the
2874 directives and define how usernames and passwords are checked, but
2875 there is now a framework to start from.
2877 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2878 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2879 without this. I don't know how relevant this is to other LDAP libraries.
2881 PH/02 Add the verb name to the "unknown ACL verb" error.
2883 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2885 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2887 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2888 directory not expanded when it should be if an expanded home directory
2889 was set for the address (which is overridden by the transport).
2891 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2894 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2895 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2896 because it is too late at that time, and has no effect.
2898 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2899 security issue with \' (bugzilla #107). I could not use the
2900 PQescapeStringConn() function, because it needs a PGconn value as one of
2903 PH/08 When testing addresses using -bt, indicate those final addresses that
2904 are duplicates that would not cause an additional delivery. At least one
2905 person was confused, thinking that -bt output corresponded to deliveries.
2906 (Suppressing duplicates isn't a good idea as you lose the information
2907 about possibly different redirections that led to the duplicates.)
2909 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2910 systems where poll() doesn't work, in particular OS X.
2912 PH/10 Added more information to debugging output for retry time not reached.
2914 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2915 operations in malware.c.
2917 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2920 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2921 syslog_duplication set false, log lines that would normally be written
2922 both the the main log and to the reject log were not written to syslog at
2925 PH/14 In the default configuration, change the use of "message" in ACL warn
2926 statements to "add_header".
2928 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2929 not followed by a command (e.g. "seen endif").
2931 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2932 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2935 PH/17 Added extra conditions to the default value of delay_warning_condition
2939 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2940 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2941 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2944 The Auto-Submitted: and various List- headers are standardised, whereas I
2945 don't think Precedence: ever was.
2947 PH/18 Refactored debugging code in route_finduser() to show more information,
2948 in particular, the error code if getpwnam() issues one.
2950 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2951 This is apparently needed in addition to the PH/07 change above to avoid
2952 any possible encoding problems.
2954 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2955 but not after initializing Perl.
2957 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2958 output them only if debugging. By default they are written stderr,
2959 apparently, which is not desirable.
2961 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2964 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2967 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2969 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2970 authenticated or an ident call has been made. Suppress the default
2971 values for $authenticated_id and $authenticated_sender (but permit -oMai
2972 and -oMas) when testing with -bh.
2974 PH/24 Re-jigged the order of the tests in the default configuration so that the
2975 tests for valid domains and recipients precede the DNS black list and CSA
2976 tests, on the grounds that those ones are more expensive.
2978 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2979 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2980 This bug exists in every version of Exim that I still have, right back to
2983 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2984 However, an attempt to turn on case-sensitivity in a regex key by
2985 including (?-i) didn't work because the subject string was already
2986 lowercased, and the effects were non-intuitive. It turns out that a
2987 one-line patch can be used to allow (?-i) to work as expected.
2993 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2994 other effects) broke the use of negated acl sub-conditions.
2996 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2999 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3000 "Deny" causes Exim to reject the incoming connection with a 554 error.
3001 Unfortunately, if there is a major crisis, such as a disk failure,
3002 tcp-wrappers gives "deny", whereas what one would like would be some
3003 kind of temporary error. A kludge has been added to help with this.
3004 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3005 554 error is used if errno is still zero or contains ENOENT (which occurs
3006 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3009 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3011 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3012 errors. Otherwise a message that provokes a temporary error (when other
3013 messages do not) can cause a whole host to time out.
3015 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3016 addresses were routed directly to files or pipes from a redirect router.
3017 File deliveries just didn't batch; pipe deliveries might have suffered
3020 PH/06 A failure to get a lock for a hints database would erroneously always say
3021 "Failed to get write lock", even when it was really a read lock.
3023 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3024 of 0600. This has been changed to use the value of the lockfile_mode
3025 option (which defaults to 0600).
3027 PH/08 Applied small patch from the Sieve maintainer.
3029 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3030 folder from quota calculations, a direct delivery into this folder messed
3031 up the contents of the maildirsize file. This was because the regex was
3032 used only to exclude .Trash (or whatever) when the size of the mailbox
3033 was calculated. There was no check that a delivery was happening into an
3034 excluded directory. This bug has been fixed by ignoring all quota
3035 processing for deliveries into excluded directories.
3037 PH/10 Added the maildirfolder_create_regex option to appendfile.
3043 PH/01 The code for finding all the local interface addresses on a FreeBSD
3044 system running IPv6 was broken. This may well have applied to all BSD
3045 systems, as well as to others that have similar system calls. The broken
3046 code found IPv4 interfaces correctly, but gave incorrect values for the
3047 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3048 that it would not match correctly against @[] and not recognize the IPv6
3051 PH/02 The ipliteral router was not recognizing addresses of the form user@
3052 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3054 PH/03 Added disable_ipv6.
3056 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3057 decoded form, because it is most often used to construct To: headers
3058 lines in autoreplies, and the decoded form may well be syntactically
3059 invalid. However, $reply_address has leading white space removed, and all
3060 newlines turned into spaces so that the autoreply transport does not
3063 PH/05 If group was specified without a user on a router, and no group or user
3064 was specified on a transport, the group from the router was ignored.
3066 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3067 for visible compile-time settings that can be used to change these
3068 numbers, for those that want even more. Backwards compatibility with old
3069 spool files has been maintained. However, going back to a previous Exim
3070 release will lost any variables that are in spool files.
3072 PH/07 Two small changes when running in the test harness: increase delay when
3073 passing a TCP/IP connection to a new process, in case the original
3074 process has to generate a bounce, and remove special handling of
3075 127.0.0.2 (sic), which is no longer necessary.
3077 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3078 be the same on different OS.
3080 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3083 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3084 whether --show-vars was specified or not
3086 JJ/02 exipick: Added support for new ACL variable spool format introduced
3089 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3090 syntactically invalid From: or Reply-to: line, and a filter used this to
3091 generate an autoreply, and therefore failed to obtain an address for the
3092 autoreply, Exim could try to deliver to a non-existent relative file
3093 name, causing unrelated and misleading errors. What now happens is that
3094 it logs this as a hard delivery error, but does not attempt to create a
3097 PH/11 The exinext utility has a -C option for testing purposes, but although
3098 the given file was scanned by exinext itself; it wasn't being passed on
3099 when Exim was called.
3101 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3102 an end-of-file indication when reading a command response.
3104 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3105 compiled. In many other places in Exim, IPv6 addresses are always
3106 recognized, so I have changed this. It also means that IPv4 domain
3107 literals of the form [IPV4:n.n.n.n] are now always recognized.
3109 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3110 used if the router is not running as root, for example, when verifying at
3111 ACL time, or when using -bh. The debugging output from this situation was
3112 non-existent - all you got was a failure to exec. I have made two
3115 (a) Failures to set uid/gid, the current directory, or a process leader
3116 in a subprocess such as that created by queryprogram now generate
3117 suitable debugging output when -d is set.
3119 (b) The queryprogram router detects when it is not running as root,
3120 outputs suitable debugging information if -d is set, and then runs
3121 the subprocess without attempting to change uid/gid.
3123 PH/15 Minor change to Makefile for building test_host (undocumented testing
3126 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3127 additional section of a DNS packet that returns MX or SRV records.
3128 Instead, it always explicitly searches for A/AAAA records. This avoids
3129 major problems that occur when a DNS server includes only records of one
3130 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3131 fixed another bug: if SRV records were looked up and the corresponding
3132 address records were *not* found in the additional section, the port
3133 values from the SRV records were lost.
3135 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3136 using the correct key (the original address) when searching the retry
3137 rules in order to find which one to use for generating the retry hint.
3139 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3140 adding the default one. Similarly, if it contains a Reply-To: header, the
3141 errors_reply_to option, if set, is not used.
3143 PH/19 When calculating a retry time, Exim used to measure the "time since
3144 failure" by looking at the "first failed" field in the retry record. Now
3145 it does not use this if it is later than than the arrival time of the
3146 message. Instead it uses the arrival time. This makes for better
3147 behaviour in cases where some deliveries succeed, thus re-setting the
3148 "first failed" field. An example is a quota failure for a huge message
3149 when small messages continue to be delivered. Without this change, the
3150 "time since failure" will always be short, possible causing more frequent
3151 delivery attempts for the huge message than are intended.
3152 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3154 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3155 $1, $2, $3) because the numerical variables can be reset during some
3156 expansion items (e.g. "match"), thereby losing the authentication data.
3158 PH/21 Make -bV show the size of off_t variables so that the test suite can
3159 decide whether to run tests for quotas > 2G.
3161 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3162 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3163 filecount value is greater than 2G or if a quota value is greater than 2G
3164 on a system where the size of off_t is not greater than 4, a panic error
3167 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3168 never match. The debug and -bh output now contains an explicit error
3169 message indicating a malformed IPv4 address or mask.
3171 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3172 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3173 PH/23 above applies.
3175 PH/25 Do not write to syslog when running in the test harness. The only
3176 occasion when this arises is a failure to open the main or panic logs
3177 (for which there is an explicit test).
3179 PH/26 Added the /no_tell option to "control=freeze".
3181 PH/27 If a host name lookup failed very early in a connection, for example, if
3182 the IP address matched host_lookup and the reverse lookup yielded a name
3183 that did not have a forward lookup, an error message of the form "no IP
3184 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3185 could be logged. Now it outputs the IP address instead of "NULL".
3187 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3188 allows the sender and the authenticated sender to be set when
3189 submitting a message from within Exim. Since child_open_exim() is
3190 documented for local_scan(), the new function should be too.
3192 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3193 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3194 results in an empty string is now treated as unset.
3196 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3198 PH/31 Added sender_verify_fail logging option.
3200 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3201 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3202 tidied the source and removed it altogether.
3204 PH/33 When a queue run was abandoned because the load average was too high, a
3205 log line was always written; now it is written only if the queue_run log
3206 selector is set. In addition, the log line for abandonment now contains
3207 information about the queue run such as the pid. This is always present
3208 in "start" and "stop" lines but was omitted from the "abandon" line.
3210 PH/34 Omit spaces between a header name and the colon in the error message that
3211 is given when verify = headers_syntax fails (if there are lots of them,
3212 the message gets confusing).
3214 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3215 names, as there are now some PTR records that contain slashes. This check
3216 is only to protect against broken name servers that fall over on strange
3217 characters, so the fact that it applies to all lookups doesn't matter.
3219 PH/36 Now that the new test suite is complete, we can remove some of the
3220 special code in Exim that was needed for the old test suite. For example,
3221 sorting DNS records because real resolvers return them in an arbitrary
3222 order. The new test suite's fake resolver always returns records in the
3225 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3226 bounces) except when queue_only is set, to avoid logging races between
3227 the different processes.
3229 PH/38 Panic-die if .include specifies a non-absolute path.
3231 PH/39 A tweak to the "H" retry rule from its user.
3233 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3234 a label. They prevented compilation on older perls.
3236 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3237 a warning to be raised on newish perls.
3239 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3240 on queue. Changes to match documented behaviour of showing count of
3241 messages matching specified criteria.
3243 PH/40 Changed the default ident timeout from 30s to 5s.
3245 PH/41 Added support for the use of login_cap features, on those BSD systems
3246 that have them, for controlling the resources used by pipe deliveries.
3248 PH/42 The content-scanning code uses fopen() to create files in which to put
3249 message data. Previously it was not paying any attention to the mode of
3250 the files. Exim runs with umask(0) because the rest of the code creates
3251 files with open(), and sets the required mode explicitly. Thus, these
3252 files were ending up world-writeable. This was not a big issue, because,
3253 being within the spool directory, they were not world-accessible. I have
3254 created a function called modefopen, which takes an additional mode
3255 argument. It sets umask(777), creates the file, chmods it to the required
3256 mode, then resets the umask. All the relevant calls to fopen() in the
3257 content scanning code have been changed to use this function.
3259 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3260 to 24 hours. This avoids potential overflow problems when processing G
3261 and H retry rules. I suspect nobody ever tinkers with this value.
3263 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3265 PH/45 When the plaintext authenticator is running as a client, the server's
3266 challenges are checked to ensure they are valid base64 strings. By
3267 default, the authentication attempt is cancelled if an invalid string is
3268 received. Setting client_ignore_invalid_base64 true ignores these errors.
3269 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3270 they are received. Thus, the responses can be made to depend on the
3271 challenges. If an invalid string is ignored, an empty string is placed in
3274 PH/46 Messages that are created by the autoreply transport now contains a
3275 References: header, in accordance with RFCs 2822 and 3834.
3277 PH/47 Added authenticated_sender_force to the smtp transport.
3279 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3281 PH/49 Installed latest patch from the Sieve maintainer.
3283 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3284 was also set, the appendfile transport was unnecessarily scanning a
3285 directory of message files (e.g. for maildir delivery) to find the count
3286 of files (along with the size), even though it did not need this
3287 information. It now does the scan only if it needs to find either the
3288 size of the count of files.
3290 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3292 PH/52 Two bugs concerned with error handling when the smtp transport is
3295 (i) Exim was not creating retry information for temporary errors given
3296 for individual recipients after the DATA command when the smtp transport
3297 was used in LMTP mode. This meant that they could be retried too
3298 frequently, and not timed out correctly.
3300 (ii) Exim was setting the flag that allows error details to be returned
3301 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3302 recipients that were returned after the DATA command.
3304 PH/53 This is related to PH/52, but is more general: for any failing address,
3305 when detailed error information was permitted to be returned to the
3306 sender, but the error was temporary, then after the final timeout, only
3307 "retry timeout exceeded" was returned. Now it returns the full error as
3308 well as "retry timeout exceeded".
3310 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3311 do this, and (what is worse) MTAs that accept it.
3313 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3314 will now be deprecated.
3316 PH/56 New os.c-cygwin from the Cygwin maintainer.
3318 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3319 formats (previously only available in exim formats via -bpr, -bpru,
3320 and -bpra. Now also available in native and exiqgrep formats)
3322 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3323 with very large, slow to parse queues
3325 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3327 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3329 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3330 responses to authentication challenges, though it was showing the
3331 challenges; (ii) I've removed the CR characters from the debug output for
3334 PH/58 Allow for the insertion of a newline as well as a space when a string
3335 is turned into more than one encoded-word during RFC 2047 encoding. The
3336 Sieve code now uses this.
3338 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3339 data_4xx, lost_connection, tls_required.
3341 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3342 message was being sent as an SMTP response.
3344 PH/61 Add -l and -k options to exicyclog.
3346 PH/62 When verifying, if an address was redirected to one new address, so that
3347 verification continued, and the new address failed or deferred after
3348 having set something in $address_data, the value of $address_data was not
3349 passed back to the ACL. This was different to the case when no
3350 redirection occurred. The value is now passed back in both cases.
3352 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3353 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3354 use login_cap.h, so on its own it isn't the distinguishing feature. The
3355 new name refers directly to the setclassresources() function.
3357 PH/65 Added configuration files for NetBSD3.
3359 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3361 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3362 is preferred over IPv4.
3364 PH/68 The bounce_return_message and bounce_return_body options were not being
3365 honoured for bounces generated during the reception of non-SMTP messages.
3366 In particular, this applied to messages rejected by the ACL. This bug has
3367 been fixed. However, if bounce_return_message is true and bounce_return_
3368 body is false, the headers that are returned for a non-SMTP message
3369 include only those that have been read before the error was detected.
3370 (In the case of an ACL rejection, they have all been read.)
3372 PH/69 The HTML version of the specification is now built in a directory called
3373 spec_html instead of spec.html, because the latter looks like a path with
3374 a MIME-type, and this confuses some software.
3376 PH/70 Catch two compiler warnings in sieve.c.
3378 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3379 function verify_get_ident() calls ip_connect() to connect a socket, but
3380 if the "connect()" function timed out, ip_connect() used to close the
3381 socket. However, verify_get_ident() also closes the socket later, and in
3382 between Exim writes to the log, which may get opened at this point. When
3383 the socket was closed in ip_connect(), the log could get the same file
3384 descriptor number as the socket. This naturally causes chaos. The fix is
3385 not to close the socket in ip_connect(); the socket should be closed by
3386 the function that creates it. There was only one place in the code where
3387 this was missing, in the iplookup router, which I don't think anybody now
3388 uses, but I've fixed it anyway.
3390 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3391 well as to direct DNS lookups. Otherwise the handling of names in host
3392 lists is inconsistent and therefore confusing.
3398 PH/01 Two changes to the default runtime configuration:
3400 (1) Move the checks for relay_from_hosts and authenticated clients from
3401 after to before the (commented out) DNS black list checks.
3403 (2) Add control=submission to the relay_from_hosts and authenticated
3404 clients checks, on the grounds that messages accepted by these
3405 statements are most likely to be submissions.
3407 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3409 (1) Generate an error if the third argument for the ${prvs expansion is
3412 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3415 (3) Reset the variables that are obtained from the first argument of
3416 ${prvscheck and used in the second argument before leaving the code,
3417 because their memory is reclaimed, so using them afterwards may do
3420 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3421 one (it's much easier than Tom thought :-).
3423 (5) Because of (4), we can now allow for the use of $prvscheck_result
3424 inside the third argument.
3426 PH/03 For some reason, the default setting of PATH when running a command from
3427 a pipe transport was just "/usr/bin". I have changed it to
3430 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3431 anything to be listed in the output from -bV.
3433 PH/05 When a filter generated an autoreply, the entire To: header line was
3434 quoted in the delivery log line, like this:
3436 => >A.N.Other <ano@some.domain> <original@ddress> ...
3438 This has been changed so that it extracts the operative address. There
3439 may be more than one such address. If so, they are comma-separated, like
3442 => >ano@some.domain,ona@other.domain <original@ddress> ...
3444 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3445 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3446 not being looked up in the rDNS to get a host name, Exim was showing the
3447 IP address twice in Received: lines, even though the IP addresses were
3448 identical. For example:
3450 Received: from [1.2.3.4] (helo=[1.2.3.4])
3452 However, if the real host name was known, it was omitting the HELO data
3453 if it matched the actual IP address. This has been tidied up so that it
3454 doesn't show the same IP address twice.
3456 PH/07 When both +timestamp and +memory debugging was on, the value given by
3457 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3458 called by the debug printing, thereby overwriting the timestamp buffer.
3459 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3461 PH/08 When the original message was included in an autoreply transport, it
3462 always said "this is a copy of the message, including all the headers",
3463 even if body_only or headers_only was set. It now gives an appropriate
3466 PH/09 Applied a patch from the Sieve maintainer which:
3468 o fixes some comments
3469 o adds the (disabled) notify extension core
3470 o adds some debug output for the result of if/elsif tests
3471 o points to the current vacation draft in the documentation
3472 and documents the missing references header update
3476 o fixes a bug in processing the envelope test (when testing
3477 multiple envelope elements, the last element determined the
3480 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3481 Electronic Mail") by including:
3483 Auto-submitted: auto-generated
3485 in the messages that it generates (bounce messages and others, such as
3486 warnings). In the case of bounce messages for non-SMTP messages, there was
3487 also a typo: it was using "Auto_submitted" (underscore instead of
3488 hyphen). Since every message generated by Exim is necessarily in response
3489 to another message, thes have all been changed to:
3491 Auto-Submitted: auto-replied
3493 in accordance with these statements in the RFC:
3495 The auto-replied keyword:
3497 - SHOULD be used on messages sent in direct response to another
3498 message by an automatic process,
3500 - MUST NOT be used on manually-generated messages,
3502 - MAY be used on Delivery Status Notifications (DSNs) and Message
3503 Disposition Notifications (MDNs),
3505 - MUST NOT be used on messages generated by automatic or periodic
3506 processes, except for messages which are automatic responses to
3509 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3510 to the default Received: header definition.
3512 PH/12 Added log selector acl_warn_skipped (default on).
3514 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3515 variables because (a) they are in the wrong storage pool and (b) even if
3516 they were copied, it wouldn't work properly because of the caching.
3518 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3519 checking when decoding. Apparently there are clients that generate
3520 overlong encoded strings. Why am I not surprised?
3522 PH/15 If the first argument of "${if match_address" was not empty, but did not
3523 contain an "@" character, Exim crashed. Now it writes a panic log message
3524 and treats the condition as false.
3526 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3528 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3529 extension "envelope-auth". The code is finished and in agreement with
3530 other implementations, but there is no documentation so far and in fact,
3531 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3532 not changing the active code.
3534 Print executed "if" and "elsif" statements when debugging is used. This
3535 helps a great deal to understand what a filter does.
3537 Document more things not specified clearly in RFC3028. I had all this
3538 sorted out, when out of a sudden new issues came to my mind. Oops."
3540 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3543 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3544 canonical form (as documented). However, after a host name lookup from
3545 the IP address, check_host() was doing a simple string comparison with
3546 addresses acquired from the DNS when checking that the found name did
3547 have the original IP as one of its addresses. Since any found IPv6
3548 addresses are likely to be in abbreviated form, the comparison could
3549 fail. Luckily, there already exists a function for doing the comparison
3550 by converting both addresses to binary, so now that is used instead of
3551 the text comparison.
3553 PH/20 There was another similar case to PH/19, when a complete host name was
3554 given in a host list; looking up its IP address could give an abbreviated
3555 form, whereas the current host's name might or might not be abbreviated.
3556 The same fix has been applied.
3562 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3563 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3566 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3567 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3569 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3571 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3572 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3573 Also, for "Auto-Submitted", treat anything other than "no" as
3574 non-personal, in accordance with RFC 3834. (Previously it treated
3575 anything starting "auto-" as non-personal.)
3577 TF/01 The control=submission/name=... option had a problem with syntax
3578 errors if the name included a slash character. The /name= option
3579 now slurps the rest of the string, so it can include any characters
3580 but it must come last in the list of options (after /sender_retain
3583 PH/05 Some modifications to the interface to the fake nameserver for the new
3591 TK/01 Added the "success_on_redirect" address verification option. See
3592 NewStuff for rationale and an example.
3594 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3596 PH/02 Patch to exigrep to allow it to work on syslog lines.
3598 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3599 fread() to skip over the body file's header line, because in Cygwin the
3600 header line is locked and is inaccessible.
3602 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3603 co-exist for some time) to make it clear that it is the Exim ID that is
3604 referenced, not the Message-ID: header line.
3606 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3607 string_format() function, because snprintf() does not exist on all
3610 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3611 ${stat: expansion item.
3613 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3614 protocol synchronization error", to keep the pedants happy.
3616 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3617 well as for IRIX systems, when gcc is being used. See the host.c source
3620 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3622 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3625 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3626 transport and to the smtp transport in LMTP mode.
3628 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3630 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3631 run a filter in a subprocess. This could lead to confusion in subsequent
3632 lookups in the parent process. There should also be a search_tidyup() at
3633 the end of the subprocess.
3635 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3636 only if the host matched helo_try_verify_hosts, which caused the
3637 verification to occur when the EHLO/HELO command was issued. The ACL just
3638 tested the remembered result. Now, if a previous verification attempt has
3639 not happened, "verify = helo" does it there and then.
3641 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3643 TK/03 Fix log output including CR from clamd.
3645 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3646 exist provoked a memory error which could cause a segfault.
3648 PH/15 Installed PCRE 6.2
3650 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3652 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3653 of the problem. Specifically, suggested +O2 rather than +O1 for the
3656 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3658 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3659 record, the port was not correctly logged when the outgoing_port log
3660 selector was set (it logged the transort's default port).
3662 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3663 fallback_hosts, and "hosts" in the smtp transport.
3665 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3666 host errors such as "Connection refused".
3668 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3669 authentication with radiusclient 0.4.9:
3671 - Error returned from rc_read_config was caught wrongly
3672 - Username/password not passed on to radius server due to wrong length.
3674 The presumption is that some radiusclient API changes for 4.51/PH/17
3675 were not taken care of correctly. The code is still untested by me (my
3676 Linux distribution still has 0.3.2 of radiusclient), but it was
3677 contributed by a Radius user.
3679 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3680 expanding the "port" option of the smtp transport.
3682 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3683 while reading a MIME header. Thanks to Tom Hughes for a patch.
3685 PH/24 Include config.h inside local_scan.h so that configuration settings are
3688 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3689 that in an ACL for RCPT (for example), you can examine exactly what was
3692 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3693 commands, but it was not correctly comparing the address with the actual
3694 client host address. Thus, it would show the EHLO address in Received:
3695 header lines when this was not necessary.
3697 PH/27 Added the % operator to ${eval:}.
3699 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3700 it should be ignoring failures (because with -C, for example, it has lost
3701 privilege). It wasn't ignoring creation failures other than "already
3704 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3707 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3708 because an input line was too long, either on its own, or by virtue of
3709 too many continuations, the temporary file was not being removed, and the
3710 return code was incorrect.
3712 PH/31 Missing "BOOL" in function definition in filtertest.c.
3714 PH/32 Applied Sieve patches from the maintainer.
3716 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3718 PH/33 Added "verify = not_blind".
3720 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3721 Local/Makefile (with some defaults set). These are used in built scripts
3722 such as exicyclog, but they have never been used in the exim_install
3723 script (though there are many overriding facilities there). I have
3724 arranged that the exim_install script now takes note of these two
3727 PH/35 Installed configuration files for Dragonfly.
3729 PH/36 When a locally submitted message by a trusted user did not contain a
3730 From: header, and the sender address was obtained from -f or from an SMTP
3731 MAIL command, and the trusted user did not use -F to supply a sender
3732 name, $originator_name was incorrectly used when constructing a From:
3733 header. Furthermore, $originator_name was used for submission mode
3734 messages from external hosts without From: headers in a similar way,
3735 which is clearly wrong.
3737 PH/37 Added control=suppress_local_fixups.
3739 PH/38 When log_selector = +received_sender was set, and the addition of the
3740 sender made the log line's construction buffer exactly full, or one byte
3741 less than full, an overflow happened when the terminating "\n" was
3744 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3745 when the result of a list match is failure because a DNS lookup failed.
3747 PH/40 RM_COMMAND is now used in the building process.
3749 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3750 the "build-* directories that it finds.
3752 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3753 address in a domain literal was a prefix of an interface address.
3755 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3756 when verifying a sender address, unless rewrite_headers is false.
3758 PH/44 Wrote a long comment about why errors_to addresses are verified as
3759 recipients, not senders.
3761 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3762 the ratelimit ACL was added.
3764 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3766 PH/46 Added extra information about PostgreSQL errors to the error string.
3768 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3769 avoiding the need to install special zones in a real server. This is
3770 backwards compatible; if it can't find the fake resolver, it drops back.
3771 Thus, both old and new test suites can be run.
3773 TF/02 Added util/ratelimit.pl
3775 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3776 clock is set back in time.
3778 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3779 Candler <B.Candler@pobox.com>.
3781 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3782 OK for addresses that are the result of redirections.
3784 PH/48 A number of further additions for the benefit of the new test suite,
3785 including a fake gethostbyname() that interfaces to the fake DNS resolver
3788 TF/06 The fix for widen_domains has also been applied to qualify_single and
3789 search_parents which are the other dnslookup options that can cause
3792 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3795 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3797 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3798 service" if it encountered an MX record with an empty target hostname.
3799 The message is now "an MX or SRV record indicated no SMTP service".
3801 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3802 if the DNS of the sending site is misconfigured. This is quite a
3803 common situation. This change restores the behaviour of treating a
3804 helo verification defer as a failure.
3806 PH/51 If self=fail was set on a router, the bounce message did not include the
3807 actual error message.
3813 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3815 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3816 command itself ended in error, the underlying message about the transport
3817 filter timeout was being overwritten with the pipe command error. Now the
3818 underlying error message should be appended to the second error message.
3820 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3822 PH/02 Reduce the amount of output that "make" produces by default. Full output
3823 can still be requested.
3825 PH/03 The warning log line about a condition test deferring for a "warn" verb
3826 was being output only once per connection, rather than after each
3827 occurrence (because it was using the same function as for successful
3828 "warn" verbs). This seems wrong, so I have changed it.
3830 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3831 it should not have, which might have caused a crash in the right
3832 circumstances, but probably never did.
3834 PH/04 Installed a modified version of Tony Finch's patch to make submission
3835 mode fix the return path as well as the Sender: header line, and to
3836 add a /name= option so that you can make the user's friendly name appear
3839 TF/03 Added the control = fakedefer ACL modifier.
3841 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3842 Mark Lowes for thorough testing.
3844 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3846 TK/03 Merged latest SRS patch from Miles Wilton.
3848 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3849 with the definition in sysexits.h (which is #included earlier).
3850 Fortunately, Exim does not actually use EX_OK. The code used to try to
3851 preserve the sysexits.h value, by assuming that macro definitions were
3852 scanned for macro replacements. I have been disabused of this notion,
3853 so now the code just undefines EX_OK before #including unistd.h.
3855 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3856 in the smtp transport. When a block could not be written in a single
3857 write() function, the timeout was being re-applied to each part-write.
3858 This seems wrong - if the receiver was accepting one byte at a time it
3859 would take for ever. The timeout is now adjusted when this happens. It
3860 doesn't have to be particularly precise.
3862 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3863 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3865 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3866 without a domain if the check to <postmaster@domain> fails.
3868 SC/01 Eximstats: added -xls and the ability to specify output files
3869 (patch written by Frank Heydlauf).
3871 SC/02 Eximstats: use FileHandles for outputting results.
3873 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3875 SC/04 Eximstats: fixed display of large numbers with -nvr option
3877 SC/05 Eximstats: fixed merging of reports with empty tables.
3879 SC/06 Eximstats: added the -include_original_destination flag
3881 SC/07 Eximstats: removed tabs and trailing whitespace.
3883 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3885 TK/06 MBOX spool code: Add real "From " MBOX separator line
3886 so the .eml file is really in mbox format (even though
3887 most programs do not really care). Patch from Alex Miller.
3889 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3890 The latter is generated from $received_to and is only set if the
3891 message has one envelope recipient. SA can use these headers,
3892 obviously out-of-the-box. Patch from Alex Miller.
3894 PH/08 The ${def test on a variable was returning false if the variable's
3895 value was "0", contrary to what the specification has always said!
3896 The result should be true unless the variable is empty.
3898 PH/09 The syntax error of a character other than { following "${if
3899 def:variable_name" (after optional whitespace) was not being diagnosed.
3900 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3901 accidental colon was present, for example, could give incorrect results.
3903 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3904 result is used (not including appendfile, where other changes are about
3907 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3908 This involved changing a lot of size variables from int to off_t. It
3909 should work with maildirs and everything.
3911 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3912 spamd dying while we are connected to it.
3914 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3917 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3918 function for BDB 4.3.
3920 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3922 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3923 the sources for PCRE have been re-arranged and more files are now
3926 PH/15 The code I had for printing potentially long long variables in PH/11
3927 above was not the best (it lost precision). The length of off_t variables
3928 is now inspected at build time, and an appropriate printing format (%ld
3929 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3930 to be "long long int" or "long int". This is needed for the internal
3931 formatting function string_vformat().
3933 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3934 the configuration file to be ":syslog", then the script "guesses" where
3935 the logs files are, rather than using the compiled in default. In our
3936 case the guess is not the same as the compiled default, so the script
3937 suddenly stopped working when I started to use syslog. The patch checks
3938 to see if log_file_path is "". If so, it attempts to read it from exim
3939 with no configuration file to get the compiled in version, before it
3940 falls back to the previous guessing code."
3942 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3943 implementing BATV in an Exim configuration. See NewStuff for the gory
3946 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3947 Makefile that are specific to HP-UX.
3949 PH/18 If the "use_postmaster" option was set for a recipient callout together
3950 with the "random" option, the postmaster address was used as the MAIL
3951 FROM address for the random test, but not for the subsequent recipient
3952 test. It is now used for both.
3954 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3955 patch removes a few documentation additions to RFC 3028, because the
3956 latest draft now contains them. It adds the new en;ascii-case comparator
3957 and a new error check for 8bit text in MIME parts. Comparator and
3958 require names are now matched exactly. I enabled the subaddress
3959 extension, but it is not well tested yet (read: it works for me)."
3961 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3962 rework some of the code of TK/09 above to avoid the hardwired use of
3963 "%lld" and "long long". Replaced the call to snprintf() with a call to
3966 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3967 records point to non-existent hosts", "retry timeout exceeded", and
3968 "retry time not reached for any host after a long failure period".
3970 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3971 experimental DomainKeys support:
3973 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3974 (2) On an error such as an illegally used "control", the wrong name for
3975 the control was given.
3977 These problems did NOT occur unless DomainKeys support was compiled.
3979 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3981 PH/24 Added ${if match_ip condition.
3983 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3984 databases so that it will be absolutely obvious if a crash occurs in the
3985 DB library. This is a regular occurrence (often caused by mis-matched
3988 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3989 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3990 picked up on a user's system that detects such things. There doesn't seem
3991 to be a gcc warning option for this - only an attribute that has to be
3992 put on the function's prototype. It seems that in Fedora Core 4 they have
3993 set this on a number of new functions. No doubt there will be more in due
3996 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3997 specify a transport. However, if an address that was verified by such a
3998 router was the subject of a callout, Exim crashed because it tried to
3999 read the rcpt_include_affixes from the non-existent transport. Now it
4000 just assumes that the setting of that option is false. This bug was
4001 introduced by 4.51/PH/31.
4003 PH/28 Changed -d+all to exclude +memory, because that information is very
4004 rarely of interest, but it makes the output a lot bigger. People tend to
4005 do -d+all out of habit.
4007 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4008 code in os-type was giving problems when libc.so lives in lib64, like on
4011 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4012 aren't the modern standard, and it seems that some systems' include files
4013 don't always have them. Exim was already checking for some of the newer
4014 ones like T_AAAA, and defining it itself. I've added checks for all the
4015 record types that Exim uses.
4017 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4018 not automatically generating a new one, as it is supposed to. This
4019 prevented TLS from working. If the file did exist, but contained invalid
4020 data, a new version was generated, as expected. It was only the case of a
4021 non-existent file that was broken.
4023 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4024 with a change in libdomainkeys > 0.64.
4026 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4027 from DNS. If the selector record carries the flag, it now has
4028 precedence over the domain-wide flag.
4030 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4032 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4033 the use of a port name that isn't defined in /etc/services) occurred, the
4034 message was deferred as in a normal delivery, and thus remained on the
4035 spool, instead of being failed because of the mua_wrapper setting. This
4036 is now fixed, and I tidied up some of the mua_wrapper messages at the
4039 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4040 the messages in a hash of arrays rather than using individual hashes.
4041 This is a bit cleaner and results in dramatic memory savings, albeit
4042 at a slight CPU cost.
4044 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4045 as requested by Marc Sherman.
4047 SC/10 Eximstats: added histograms for user specified patterns as requested
4050 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4052 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4053 fopen() in the content-scanning modules that did not already have it.
4059 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4060 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4062 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4064 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4066 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4067 to test Sieve filters that use "vacation".
4069 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4070 that changes the way the GnuTLS parameters are stored in the cache file.
4071 The new format can be generated externally. For backward compatibility,
4072 if the data in the cache doesn't make sense, Exim assumes it has read an
4073 old-format file, and it generates new data and writes a new file. This
4074 means that you can't go back to an older release without removing the
4077 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4078 work if there are any delivery delays because "one_time" forces the
4079 parent to be marked "delivered", so its unseen clone is never tried
4080 again. For this reason, Exim now forbids the simultaneous setting of
4083 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4084 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4085 entry if you want to know the details. The fix, however, neglected to
4086 consider the case when local delivery batching is involved. The test for
4087 "previously delivered" was not happening when checking to see if an
4088 address could be batched with a previous (undelivered) one; under
4089 certain circumstances this could lead to multiple deliveries to the same
4092 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4093 in its include files, and this causes problems building Exim.
4095 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4096 header_syntax) but Exim was just ignoring anything given after a slash.
4097 In particular, this caused confusion with an attempt to use "verify =
4098 reverse_host_lookup/defer_ok". An error is now given when options are
4099 supplied for verify items that do not have them. (Maybe reverse_host_
4100 lookup should have a defer_ok option, but that's a different point.)
4102 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4103 defined by RFC 821) to 2048, because there were problems with some AUTH
4104 commands, and RFC 1869 says the size should be increased for extended
4105 SMTP commands that take arguments.
4107 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4110 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4111 "unknown" error; now it says that the functionality isn't in the binary.
4113 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4114 an address' error message when a string expansion fails (syntax or
4115 whatever). Otherwise the password may appear in the log. Following change
4116 PH/42 below, there is no longer a chance of it appearing in a bounce
4119 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4121 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4122 of its addresses was ever tried. (Bugzilla bug #2.)
4124 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4125 the result incorrectly in the debug output. (It correctly added a newline
4126 to what was transported.)
4128 TF/01 Added $received_time.
4130 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4131 commented out examples of how to interface to a virus scanner and to
4132 SpamAssassin. Also added commented examples of av_scanner and
4133 spamd_address settings.
4135 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4136 and controls are allowed in which ACLs. There were a couple of minor
4137 errors. Some of the entries in the conditions table (which is a table of
4138 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4139 negation of where the condition IS allowed.
4141 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4143 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4144 header file does not have a version number, so I've had to invent a new
4145 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4146 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4147 radiusclient), but it was contributed by a Radius user.
4149 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4150 files or directories, for OpenSSL.
4152 PH/19 When an Exim process that is running as root has to create an Exim log
4153 file, it does so in a subprocess that runs as exim:exim so as to get the
4154 ownership right at creation (otherwise, other Exim processes might see
4155 the file with the wrong ownership). There was no test for failure of this
4156 fork() call, which would lead to the process getting stuck as it waited
4157 for a non-existent subprocess. Forks do occasionally fail when resources
4158 run out. I reviewed all the other calls to fork(); they all seem to check
4161 PH/20 When checking for unexpected SMTP input at connect time (before writing
4162 the banner), Exim was not dealing correctly with a non-positive return
4163 from the read() function. If the client had disconnected by this time,
4164 the result was a log entry for a synchronization error with an empty
4165 string after "input=" when read() returned zero. If read() returned -1
4166 (an event I could not check), uninitialized data bytes were printed.
4167 There were reports of junk text (parts of files, etc) appearing after
4170 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4172 PH/22 Added support for macro redefinition, and (re)definition in between
4173 driver and ACL definitions.
4175 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4176 forgetting to use the resulting value; it was using the unexpanded value.
4178 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4179 hadn't been configured. The fix is from Juergen Kreileder, who
4180 understands it better than I do:
4182 "Here's what I see happening with three configured cyrus_sasl
4183 authenticators configured (plain, login, cram-md5):
4185 On startup auth_cyrus_sasl_init() gets called for each of these.
4186 This means three calls to sasl_listmech() without a specified mech_list.
4187 => SASL tests which mechs of all available mechs actually work
4188 => three warnings about OTP not working
4189 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4191 With the patch, sasl_listmech() also gets called three times. But now
4192 SASL's mech_list option is set to the server_mech specified in the the
4193 authenticator. Or in other words, the answer from sasl_listmech()
4194 gets limited to just the mech you're testing for (which is different
4196 => the return list contains just 'plain' or 'login', 'cram-md5' or
4197 nothing depending on the value of ob->server_mech.
4199 I've just tested the patch: Authentication still works fine,
4200 unavailable mechs specified in the exim configuration are still
4201 caught, and the auth.log warnings about OTP are gone."
4203 PH/25 When debugging is enabled, the contents of the command line are added
4204 to the debugging output, even when log_selector=+arguments is not
4207 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4208 answer is "GNU", and only if the return is "GNU/something" is the answer
4211 PH/27 $acl_verify_message is now set immediately after the failure of a
4212 verification in an ACL, and so is available in subsequent modifiers. In
4213 particular, the message can be preserved by coding like this:
4215 warn !verify = sender
4216 set acl_m0 = $acl_verify_message
4218 Previously, $acl_verify_message was set only while expanding "message"
4219 and "log_message" when a very denied access.
4221 PH/28 Modified OS/os.c-Linux with
4223 -#ifndef OS_LOAD_AVERAGE
4224 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4226 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4227 nomenclature these days.)
4229 PH/29 Installed patch from the Sieve maintainer that adds the options
4230 sieve_useraddress and sieve_subaddress to the redirect router.
4232 PH/30 In these circumstances:
4233 . Two addresses routed to the same list of hosts;
4234 . First host does not offer TLS;
4235 . First host accepts first address;
4236 . First host gives temporary error to second address;
4237 . Second host offers TLS and a TLS session is established;
4238 . Second host accepts second address.
4239 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4240 and peerdn, if requested) that were in fact used only for the second
4243 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4244 attention to any local part prefix or suffix that was matched by the
4245 router that accepted the address. It now behaves in the same way as it
4246 does for delivery: the affixes are removed from the local part unless
4247 rcpt_include_affixes is set on the transport.
4249 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4250 timeout during the DATA phase of an incoming message.
4252 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4253 applied a patch sanctioned by the Sieve maintainer.
4255 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4256 the uid or gid is negative. A case of a negative gid caused this to be
4257 noticed. The fix allows for either to be negative.
4259 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4260 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4263 PH/36 The change PH/12 above was broken. Fixed it.
4265 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4266 the grounds that routing the same address twice would always produce the
4267 same answer. This might have been true once, but it is certainly no
4268 longer true now. Routing a child address may depend on the previous
4269 routing that produced that child. Some complicated redirection strategies
4270 went wrong when messages had multiple recipients, and made Exim's
4271 behaviour dependent on the order in which the addresses were given.
4273 I have moved the duplicate checking until after the routing is complete.
4274 Exim scans the addresses that are assigned to local and remote
4275 transports, and removes any duplicates. This means that more work will be
4276 done, as duplicates will always all be routed, but duplicates are
4277 presumably rare, so I don't expect this is of any significance.
4279 For deliveries to pipes, files, and autoreplies, the duplicate checking
4280 still happens during the routing process, since they are not going to be
4283 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4284 It corrects a timeout issue with spamd. This is Ian's comment: "The
4285 background is that sometimes spamd either never reads data from a
4286 connection it has accepted, or it never writes response data. The exiscan
4287 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4288 blindly assumes that writes won't block so it may never time out."
4290 PH/39 Allow G after quota size as well as K and M.
4292 PH/40 The value set for $authenticated_id in an authenticator may not contain
4293 binary zeroes or newlines because the value is written to log lines and
4294 to spool files. There was no check on this. Now the value is run through
4295 the string_printing() function so that such characters are converted to
4296 printable escape sequences.
4298 PH/41 $message_linecount is a new variable that contains the total number of
4299 lines in the message. Compare $body_linecount, which is the count for the
4302 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4303 bounce and delay warning messages, except in certain special cases, which
4306 (a) An SMTP error message from a remote host;
4307 (b) A message specified in a :fail: redirection;
4308 (c) A message specified in a "fail" command in a system filter;
4309 (d) A message specified in a FAIL return from the queryprogram router;
4310 (e) A message specified by the cannot_route_message router option.
4312 In these cases only, Exim does include the error details in bounce and
4313 warning messages. There are also a few cases where bland messages such
4314 as "unrouteable address" or "local delivery error" are given.
4316 PH/43 $value is now also set for the "else" part of a ${run expansion.
4318 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4319 being worked on, but at least Exim now implements the latest version to
4322 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4323 process to complete was treated as a delivery failure, a timeout while
4324 writing the message to the pipe was logged, but erroneously treated as a
4325 successful delivery. Such timeouts include transport filter timeouts. For
4326 consistency with the overall process timeout, these timeouts are now
4327 treated as errors, giving rise to delivery failures by default. However,
4328 there is now a new Boolean option for the pipe transport called
4329 timeout_defer, which, if set TRUE, converts the failures into defers for
4330 both kinds of timeout. A transport filter timeout is now identified in
4333 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4334 systems where "make" and "gmake" are different, calling "gmake" at top
4335 level broke things. I've arranged for the value of $(MAKE) to be passed
4336 from the Makefile to this script so that it can call the same version of
4340 A note about Exim versions 4.44 and 4.50
4341 ----------------------------------------
4343 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4344 changes of various kinds. As a consequence, a big documentation update was
4345 needed. This delayed the release for rather longer than seemed good, especially
4346 in the light of a couple of (minor) security issues. Therefore, the changes
4347 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4348 release. So 4.44 and 4.50 are in effect two different branches that both start
4351 I have left the 4.50 change log unchanged; it contains all the changes since
4352 4.43. The change log for 4.44 is below; many of its items are identical to
4353 those for 4.50. This seems to be the most sensible way to preserve the
4354 historical information.
4360 1. Minor wording change to the doc/README.SIEVE file.
4362 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4363 computation of the current number of files was incorrect.
4365 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4366 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4369 4. Give more explanation in the error message when the command for a transport
4370 filter fails to execute.
4372 5. There are several places where Exim runs a non-Exim command in a
4373 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4374 was being done only for the command run by the queryprogram router. It is
4375 now done for all such subprocesses. The other cases are: ${run, transport
4376 filters, and the commands run by the lmtp and pipe transports.
4378 6. Added CONFIGURE_GROUP build-time option.
4380 7. Some older OS have a limit of 256 on the maximum number of file
4381 descriptors. Exim was using setrlimit() to set 1000 as a large value
4382 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4383 systems. I've change it so that if it can't get 1000, it tries for 256.
4385 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4386 was an oversight, and furthermore, ever since the addition of extra
4387 controls (e.g. 4.43/32), the checks on when to allow different forms of
4388 "control" were broken. There should now be diagnostics for all cases when a
4389 control that does not make sense is encountered.
4391 9. Added the /retain_sender option to "control=submission".
4393 10. $recipients is now available in the predata ACL (oversight).
4395 11. Tidy the search cache before the fork to do a delivery from a message
4396 received from the command line. Otherwise the child will trigger a lookup
4397 failure and thereby defer the delivery if it tries to use (for example) a
4398 cached ldap connection that the parent has called unbind on.
4400 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4401 of $address_data from the recipient verification was clobbered by the
4402 sender verification.
4404 13. The value of address_data from a sender verification is now available in
4405 $sender_address_data in subsequent conditions in the ACL statement.
4407 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4409 15. Added a new option "connect=<time>" to callout options, to set a different
4412 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4413 was its contents. (It was OK if the option was not defined at all.)
4415 17. A "Completed" log line is now written for messages that are removed from
4416 the spool by the -Mrm option.
4418 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4419 information about exactly what failed.
4421 19. Added -dd to debug only the daemon process.
4423 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4424 handles timeouts, both on the server side and network timeouts. Renamed the
4425 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4427 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4428 It is now set to "smtps".
4430 22. $host_address is now set to the target address during the checking of
4431 ignore_target_hosts.
4433 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4434 being passed; this would have caused $sender_host_name to have been used if
4435 matching the list had actually called for a host name (not very likely,
4436 since this list is usually IP addresses). A host name is now passed as
4439 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4440 code that specifies a non-restarting handler (typically sigaction() in
4441 modern systems) in an attempt to fix a rare and obscure crash bug.
4443 25. Narrowed the window for a race in the daemon that could cause it to ignore
4444 SIGCHLD signals. This is not a major problem, because they are used only to
4445 wake it up if nothing else does.
4447 26. A malformed maildirsize file could cause Exim to calculate negative values
4448 for the mailbox size or file count. Odd effects could occur as a result.
4449 The maildirsize information is now recalculated if the size or filecount
4452 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4453 support for a long time. Removed HAVE_SYS_VFS_H.
4455 28. Installed the latest version of exipick from John Jetmore.
4457 29. In an address list, if the pattern was not a regular expression, an empty
4458 subject address (from a bounce message) matched only if the pattern was an
4459 empty string. Non-empty patterns were not even tested. This was the wrong
4460 because it is perfectly reasonable to use an empty address as part of a
4461 database query. An empty address is now tested by patterns that are
4462 lookups. However, all the other forms of pattern expect the subject to
4463 contain a local part and a domain, and therefore, for them, an empty
4464 address still always fails if the pattern is not itself empty.
4466 30. Exim went into a mad DNS loop when attempting to do a callout where the
4467 host was specified on an smtp transport, and looking it up yielded more
4468 than one IP address.
4470 31. Re-factored the code for checking spool and log partition space into a
4471 function that finds that data and another that does the check. The former
4472 is then used to implement four new variables: $spool_space, $log_space,
4473 $spool_inodes, and $log_inodes.
4475 32. The RFC2047 encoding function was originally intended for short strings
4476 such as real names; it was not keeping to the 75-character limit for
4477 encoded words that the RFC imposes. It now respects the limit, and
4478 generates multiple encoded words if necessary. To be on the safe side, I
4479 have increased the buffer size for the ${rfc2047: expansion operator from
4482 33. It is now permitted to omit both strings after an "if" condition; if the
4483 condition is true, the result is "true". As before, when the second string
4484 is omitted, a false condition yields an empty string. This makes it less
4485 cumbersome to write custom ACL and router conditions.
4487 34. Failure to deliver a bounce message always caused it to be frozen, even if
4488 there was an errors_to setting on the router. The errors_to setting is now
4491 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4492 canonical form (fully expanded) before being placed in
4493 $sender_host_address.
4495 36. The table in the code that translates DNS record types into text (T_A to
4496 "A" for instance) was missing entries for NS and CNAME. It is just possible
4497 that this could have caused confusion if both these types were looked up
4498 for the same domain, because the text type is used as part of Exim's
4499 per-process caching. But the chance of anyone hitting this buglet seems
4502 37. The dnsdb lookup has been extended in a number of ways.
4504 (1) There is a new type, "zns", which walks up the domain tree until it
4505 finds some nameserver records. It should be used with care.
4507 (2) There is a new type, "mxh", which is like "mx" except that it returns
4508 just the host names, not the priorities.
4510 (3) It is now possible to give a list of domains (or IP addresses) to be
4511 looked up. The behaviour when one of the lookups defers can be
4512 controlled by a keyword.
4514 (4) It is now possible to specify the separator character for use when
4515 multiple records are returned.
4517 38. The dnslists ACL condition has been extended: it is now possible to supply
4518 a list of IP addresses and/or domains to be looked up in a particular DNS
4521 39. Added log_selector=+queue_time_overall.
4523 40. When running the queue in the test harness, wait just a tad after forking a
4524 delivery process, to get repeatability of debugging output.
4526 41. Include certificate and key file names in error message when GnuTLS fails
4527 to set them up, because the GnuTLS error message doesn't include the name
4528 of the failing file when there is a problem reading it.
4530 42. Allow both -bf and -bF in the same test run.
4532 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4534 44. The "Exiscan patch" is now merged into the mainline Exim source.
4536 45. Sometimes the final signoff response after QUIT could fail to get
4537 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4538 < 0 before doing a fflush(). This bug looks as though it goes back to the
4539 introduction of TLS in release 3.20, but "sometimes" must have been rare
4540 because the tests only now provoked it.
4542 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4543 (this can affect the format of dates).
4545 47. exim_tidydb, when checking for the continued existence of a message for
4546 which it has found a message-specific retry record, was not finding
4547 messages that were in split spool directories. Consequently, it was
4548 deleting retry records that should have stayed in existence.
4550 48. Steve fixed some bugs in eximstats.
4552 49. The SPA authentication driver was not abandoning authentication and moving
4553 on to the next authenticator when an expansion was forced to fail,
4554 contradicting the general specification for all authenticators. Instead it
4555 was generating a temporary error. It now behaves as specified.
4557 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4558 (the order specifies the preference for clients). The order is now AES256,
4559 AES128, 3DES, ARCFOUR128.
4561 51. Small patch to Sieve code - explicitly set From: when generating an
4564 52. Exim crashed if a remote delivery caused a very long error message to be
4565 recorded - for instance if somebody sent an entire SpamAssassin report back
4566 as a large number of 550 error lines. This bug was coincidentally fixed by
4567 increasing the size of one of Exim's internal buffers (big_buffer) that
4568 happened as part of the Exiscan merge. However, to be on the safe side, I
4569 have made the code more robust (and fixed the comments that describe what
4572 53. Now that there can be additional text after "Completed" in log lines (if
4573 the queue_time_overall log selector is set), a one-byte patch to exigrep
4574 was needed to allow it to recognize "Completed" as not the last thing in
4577 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4578 patch that reportedly fixes this has been added. I am not expert enough to
4579 create a test for it. This is what the patch creator wrote:
4581 "I found a little strange behaviour of ldap code when working with
4582 Windows 2003 AD Domain, where users was placed in more than one
4583 Organization Units. When I tried to give exim partial DN, the exit code
4584 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4585 But simultaneously result of request was absolutely normal ldap result,
4586 so I produce this patch..."
4588 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4589 REFERENCE, so I have modified the code to exclude the patch when that macro
4592 55. Some experimental protocols are using DNS PTR records for new purposes. The
4593 keys for these records are domain names, not reversed IP addresses. The
4594 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4595 leaves it alone. Component reversal etc. now happens only for IP addresses.
4598 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4600 57. Double the size of the debug message buffer (to 2048) so that more of very
4601 long debug lines gets shown.
4603 58. The exicyclog utility now does better if the number of log files to keep
4604 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4606 59. Two changes related to the smtp_active_hostname option:
4608 (1) $smtp_active_hostname is now available as a variable.
4609 (2) The default for smtp_banner uses $smtp_active_hostname instead
4610 of $primary_hostname.
4612 60. The host_aton() function is supposed to be passed a string that is known
4613 to be a valid IP address. However, in the case of IPv6 addresses, it was
4614 not checking this. This is a hostage to fortune. Exim now panics and dies
4615 if the condition is not met. A case was found where this could be provoked
4616 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4617 components; fortuitously, this particular loophole had already been fixed
4618 by change 4.50/55 above.
4620 If there are any other similar loopholes, the new check in host_aton()
4621 itself should stop them being exploited. The report I received stated that
4622 data on the command line could provoke the exploit when Exim was running as
4623 exim, but did not say which command line option was involved. All I could
4624 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4625 running as the user.
4628 61. There was a buffer overflow vulnerability in the SPA authentication code
4629 (which came originally from the Samba project). I have added a test to the
4630 spa_base64_to_bits() function which I hope fixes it.
4633 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4634 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4636 63. The daemon start-up calls getloadavg() while still root for those OS that
4637 need the first call to be done as root, but it missed one case: when
4638 deliver_queue_load_max is set with deliver_drop_privilege. This is
4639 necessary for the benefit of the queue runner, because there is no re-exec
4640 when deliver_drop_privilege is set.
4642 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4643 This has been fixed.
4645 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4646 was in use, was not putting the data itself into the right store pool;
4647 consequently, it could be overwritten for a subsequent message in the same
4648 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4651 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4653 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4654 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4655 the function were treating the return as a boolean value, which happened to
4656 work because 0=false and not-0=true, but is not correct code.
4658 68. The host_aton() function was not handling scoped IPv6 addresses (those
4659 with, for example, "%eth0" on the end) correctly.
4661 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4662 negated items (that is, ~something) in unsigned ints. Some compilers
4663 apparently mutter when there is no cast.
4665 70. If an address verification called from an ACL failed, and did not produce a
4666 user-specific message (i.e. there was only a "system" message), nothing was
4667 put in $acl_verify_message. In this situation, it now puts the system
4670 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4671 session; change 4.31/43 added the unwanted input to the log line - except
4672 that it did not do this in the start of session case. It now does.
4674 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4675 This is wrong and can cause the other end to generate a synchronization
4676 error if it is another Exim or anything else that does the synchronization
4677 check. A QUIT command is no longer sent after a timeout.
4679 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4680 during host lookups.
4682 74. The defer_ok option of callout verification was not working if it was used
4683 when verifying addresses in header lines, that is, for this case:
4685 verify = header_sender/callout=defer_ok
4687 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4688 those file descriptors could be used for SMTP connections. If anything
4689 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4690 could be sent to the SMTP client, causing chaos. The daemon now opens
4691 stdin, stdout, and stderr to /dev/null when it puts itself into the
4694 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4695 log by default. The user can override this with suitable Perl magic.
4697 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4698 the log message when discard triggers, was not working for the DATA ACL or
4699 for the non-SMTP ACL.
4701 78. Error message wording change in sieve.c.
4703 79. If smtp_accept_max_per_host was set, the number of connections could be
4704 restricted to fewer than expected, because the daemon was trying to set up
4705 a new connection before checking whether the processes handling previous
4706 connections had finished. The check for completed processes is now done
4707 earlier. On busy systems, this bug wouldn't be noticed because something
4708 else would have woken the daemon, and it would have reaped the completed
4711 80. If a message was submitted locally by a user whose login name contained one
4712 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4713 It caused a spool format error. I have fixed the spool reading code. A
4714 related problem was that the "from" clause in the Received: line became
4715 illegal because of the space(s). It is now covered by ${quote_local_part.
4717 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4720 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4722 83. Patch from Sieve maintainer to fix unterminated string problem in
4723 "vacation" handling.
4725 84. Some minor changes to the Linux configuration files to help with other
4726 OS variants using glibc.
4728 85. One more patch for Sieve to update vacation handling to latest spec.
4731 ----------------------------------------------------
4732 See the note above about the 4.44 and 4.50 releases.
4733 ----------------------------------------------------
4739 1. Change 4.43/35 introduced a bug that caused file counts to be
4740 incorrectly computed when quota_filecount was set in an appendfile
4743 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4744 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4747 3. Give more explanation in the error message when the command for a transport
4748 filter fails to execute.
4750 4. There are several places where Exim runs a non-Exim command in a
4751 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4752 was being done only for the command run by the queryprogram router. It is
4753 now done for all such subprocesses. The other cases are: ${run, transport
4754 filters, and the commands run by the lmtp and pipe transports.
4756 5. Some older OS have a limit of 256 on the maximum number of file
4757 descriptors. Exim was using setrlimit() to set 1000 as a large value
4758 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4759 systems. I've change it so that if it can't get 1000, it tries for 256.
4761 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4762 was an oversight, and furthermore, ever since the addition of extra
4763 controls (e.g. 4.43/32), the checks on when to allow different forms of
4764 "control" were broken. There should now be diagnostics for all cases when a
4765 control that does not make sense is encountered.
4767 7. $recipients is now available in the predata ACL (oversight).
4769 8. Tidy the search cache before the fork to do a delivery from a message
4770 received from the command line. Otherwise the child will trigger a lookup
4771 failure and thereby defer the delivery if it tries to use (for example) a
4772 cached ldap connection that the parent has called unbind on.
4774 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4775 of $address_data from the recipient verification was clobbered by the
4776 sender verification.
4778 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4779 was its contents. (It was OK if the option was not defined at all.)
4781 11. A "Completed" log line is now written for messages that are removed from
4782 the spool by the -Mrm option.
4784 12. $host_address is now set to the target address during the checking of
4785 ignore_target_hosts.
4787 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4788 being passed; this would have caused $sender_host_name to have been used if
4789 matching the list had actually called for a host name (not very likely,
4790 since this list is usually IP addresses). A host name is now passed as
4793 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4794 code that specifies a non-restarting handler (typically sigaction() in
4795 modern systems) in an attempt to fix a rare and obscure crash bug.
4797 15. Narrowed the window for a race in the daemon that could cause it to ignore
4798 SIGCHLD signals. This is not a major problem, because they are used only to
4799 wake it up if nothing else does.
4801 16. A malformed maildirsize file could cause Exim to calculate negative values
4802 for the mailbox size or file count. Odd effects could occur as a result.
4803 The maildirsize information is now recalculated if the size or filecount
4806 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4807 support for a long time. Removed HAVE_SYS_VFS_H.
4809 18. Updated exipick to current release from John Jetmore.
4811 19. Allow an empty sender to be matched against a lookup in an address list.
4812 Previously the only cases considered were a regular expression, or an
4815 20. Exim went into a mad DNS lookup loop when doing a callout where the
4816 host was specified on the transport, if the DNS lookup yielded more than
4819 21. The RFC2047 encoding function was originally intended for short strings
4820 such as real names; it was not keeping to the 75-character limit for
4821 encoded words that the RFC imposes. It now respects the limit, and
4822 generates multiple encoded words if necessary. To be on the safe side, I
4823 have increased the buffer size for the ${rfc2047: expansion operator from
4826 22. Failure to deliver a bounce message always caused it to be frozen, even if
4827 there was an errors_to setting on the router. The errors_to setting is now
4830 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4831 canonical form (fully expanded) before being placed in
4832 $sender_host_address.
4834 24. Updated eximstats to version 1.33
4836 25. Include certificate and key file names in error message when GnuTLS fails
4837 to set them up, because the GnuTLS error message doesn't include the name
4838 of the failing file when there is a problem reading it.
4840 26. Expand error message when OpenSSL has problems setting up cert/key files.
4843 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4844 (this can affect the format of dates).
4846 28. exim_tidydb, when checking for the continued existence of a message for
4847 which it has found a message-specific retry record, was not finding
4848 messages that were in split spool directories. Consequently, it was
4849 deleting retry records that should have stayed in existence.
4851 29. eximstats updated to version 1.35
4852 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4853 1.35 - bugfix such that pie charts by volume are generated correctly
4855 30. The SPA authentication driver was not abandoning authentication and moving
4856 on to the next authenticator when an expansion was forced to fail,
4857 contradicting the general specification for all authenticators. Instead it
4858 was generating a temporary error. It now behaves as specified.
4860 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4861 (the order specifies the preference for clients). The order is now AES256,
4862 AES128, 3DES, ARCFOUR128.
4864 31. Small patch to Sieve code - explicitly set From: when generating an
4867 32. Exim crashed if a remote delivery caused a very long error message to be
4868 recorded - for instance if somebody sent an entire SpamAssassin report back
4869 as a large number of 550 error lines. This bug was coincidentally fixed by
4870 increasing the size of one of Exim's internal buffers (big_buffer) that
4871 happened as part of the Exiscan merge. However, to be on the safe side, I
4872 have made the code more robust (and fixed the comments that describe what
4875 33. Some experimental protocols are using DNS PTR records for new purposes. The
4876 keys for these records are domain names, not reversed IP addresses. The
4877 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4878 leaves it alone. Component reversal etc. now happens only for IP addresses.
4881 34. The host_aton() function is supposed to be passed a string that is known
4882 to be a valid IP address. However, in the case of IPv6 addresses, it was
4883 not checking this. This is a hostage to fortune. Exim now panics and dies
4884 if the condition is not met. A case was found where this could be provoked
4885 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4886 components; fortuitously, this particular loophole had already been fixed
4887 by change 4.50/55 or 4.44/33 above.
4889 If there are any other similar loopholes, the new check in host_aton()
4890 itself should stop them being exploited. The report I received stated that
4891 data on the command line could provoke the exploit when Exim was running as
4892 exim, but did not say which command line option was involved. All I could
4893 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4894 running as the user.
4897 35. There was a buffer overflow vulnerability in the SPA authentication code
4898 (which came originally from the Samba project). I have added a test to the
4899 spa_base64_to_bits() function which I hope fixes it.
4902 36. The daemon start-up calls getloadavg() while still root for those OS that
4903 need the first call to be done as root, but it missed one case: when
4904 deliver_queue_load_max is set with deliver_drop_privilege. This is
4905 necessary for the benefit of the queue runner, because there is no re-exec
4906 when deliver_drop_privilege is set.
4908 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4909 was in use, was not putting the data itself into the right store pool;
4910 consequently, it could be overwritten for a subsequent message in the same
4911 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4914 38. Sometimes the final signoff response after QUIT could fail to get
4915 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4916 < 0 before doing a fflush(). This bug looks as though it goes back to the
4917 introduction of TLS in release 3.20, but "sometimes" must have been rare
4918 because the tests only now provoked it.
4924 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4925 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4926 Now it returns an integer. A number of calls were still expecting a T/F
4927 return. Fortuitously, in all cases, the tests worked in OK situations,
4928 which is the norm. However, things would have gone wrong on any write
4929 failures on the smtp file descriptor. This function is used when sending
4930 messages over SMTP and also when doing verify callouts.
4932 2. When Exim is called to do synchronous delivery of a locally submitted
4933 message (the -odf or -odi options), it no longer closes stderr before doing
4936 3. Implemented the mua_wrapper option.
4938 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4940 5. Implemented the functions header_remove(), header_testname(),
4941 header_add_at_position(), and receive_remove_recipient(), and exported them
4944 6. If an ACL "warn" statement specified the addition of headers, Exim already
4945 inserted X-ACL-Warn: at the start if there was no header name. However, it
4946 was not making this test for the second and subsequent header lines if
4947 there were newlines in the string. This meant that an invalid header could
4948 be inserted if Exim was badly configured.
4950 7. Allow an ACL "warn" statement to add header lines at the start or after all
4951 the Received: headers, as well as at the end.
4953 8. Added the rcpt_4xx retry error code.
4955 9. Added postmaster_mailfrom=xxx to callout verification option.
4957 10. Added mailfrom=xxxx to the callout verification option, for verify=
4960 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4961 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4963 12. Inserted some casts to stop certain compilers warning when using pointer
4964 differences as field lengths or precisions in printf-type calls (mostly
4965 affecting debugging statements).
4967 13. Added optional readline() support for -be (dynamically loaded).
4969 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4970 same clock tick as a message's arrival, so that its received time was the
4971 same as the "first fail" time on the retry record, and that message
4972 remained on the queue past the ultimate address timeout, every queue runner
4973 would try a delivery (because it was past the ultimate address timeout) but
4974 after another failure, the ultimate address timeout, which should have then
4975 bounced the address, did not kick in. This was a "< instead of <=" error;
4976 in most cases the first failure would have been in the next clock tick
4977 after the received time, and all would be well.
4979 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4980 being recognized when the domain list was tested by the match_domain
4981 condition in an expansion string.
4983 16. Added the ${str2b64: operator.
4985 17. Exim was always calling setrlimit() to set a large limit for the number of
4986 processes, without checking whether the existing limit was already
4987 adequate. (It did check for the limit on file descriptors.) Furthermore,
4988 errors from getrlimit() and setrlimit() were being ignored. Now they are
4989 logged to the main and panic logs, but Exim does carry on, to try to do its
4990 job under whatever limits there are.
4992 18. Imported PCRE 5.0.
4994 19. Trivial typo in log message " temporarily refused connection" (the leading
4997 20. If the log selector return_path_on_delivery was set and an address was
4998 redirected to /dev/null, the delivery process crashed because it assumed
4999 that a return path would always be set for a "successful" delivery. In this
5000 case, the whole delivery is bypassed as an optimization, and therefore no
5003 21. Internal re-arrangement: the function for sending a challenge and reading
5004 a response while authentication was assuming a zero-terminated challenge
5005 string. It's now changed to take a pointer and a length, to allow for
5006 binary data in such strings.
5008 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5010 23. Exim was not respecting finduser_retries when seeking the login of the
5011 uid under which it was called; it was always trying 10 times. (The default
5012 setting of finduser_retries is zero.) Also, it was sleeping after the final
5013 failure, which is pointless.
5015 24. Implemented tls_on_connect_ports.
5017 25. Implemented acl_smtp_predata.
5019 26. If the domain in control=submission is set empty, Exim assumes that the
5020 authenticated id is a complete email address when it generates From: or
5021 Sender: header lines.
5023 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5024 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5025 chown and chgrp in /bin and hostname in /usr/bin.
5027 28. Exim was keeping the "process log" file open after each use, just as it
5028 does for the main log. This opens the possibility of it remaining open for
5029 long periods when the USR1 signal hits a daemon. Occasional processlog
5030 errors were reported, that could have been caused by this. Anyway, it seems
5031 much more sensible not to leave this file open at all, so that is what now
5034 29. The long-running daemon process does not normally write to the log once it
5035 has entered its main loop, and it closes the log before doing so. This is
5036 so that log files can straightforwardly be renamed and moved. However,
5037 there are a couple of unusual error situations where the daemon does write
5038 log entries, and I had neglected to close the log afterwards.
5040 30. The text of an SMTP error response that was received during a remote
5041 delivery was being truncated at 512 bytes. This is too short for some of
5042 the long messages that one sometimes sees. I've increased the limit to
5045 31. It is now possible to make retry rules that apply only when a message has a
5046 specific sender, in particular, an empty sender.
5048 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5049 it possible to be selective about when SMTP synchronization is enforced.
5051 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5053 32. Implemented hosts_connection_nolog.
5055 33. Added an ACL for QUIT.
5057 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5060 35. Added mailbox_size and mailbox_filecount to appendfile.
5062 36. Added control = no_multiline_responses to ACLs.
5064 37. There was a bug in the logic of the code that waits for the clock to tick
5065 in the case where the clock went backwards by a substantial amount such
5066 that the microsecond fraction of "now" was more than the microsecond
5067 fraction of "then" (but the whole seconds number was less).
5069 38. Added support for the libradius Radius client library this is found on
5070 FreeBSD (previously only the radiusclient library was supported).
5076 1. When certain lookups returned multiple values in the form name=value, the
5077 quoting of the values was not always being done properly. Specifically:
5078 (a) If the value started with a double quote, but contained no whitespace,
5080 (b) If the value contained whitespace other than a space character (i.e.
5081 tabs or newlines or carriage returns) it was not quoted.
5082 This fix has been applied to the mysql and pgsql lookups by writing a
5083 separate quoting function and calling it from the lookup code. The fix
5084 should probably also be applied to nisplus, ibase and oracle lookups, but
5085 since I cannot test any of those, I have not disturbed their existing code.
5087 2. A hit in the callout cache for a specific address caused a log line with no
5088 reason for rejecting RCPT. Now it says "Previous (cached) callout
5089 verification failure".
5091 3. There was an off-by-one bug in the queryprogram router. An over-long
5092 return line was truncated at 256 instead of 255 characters, thereby
5093 overflowing its buffer with the terminating zero. As well as fixing this, I
5094 have increased the buffer size to 1024 (and made a note to document this).
5096 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5097 when Exim is waiting for an SMTP response from a remote server, Exim
5098 restarts its select() call on the socket, thereby resetting its timeout.
5099 This is not a problem when such interrupts are rare. Somebody set up a cron
5100 job to run exiwhat every 2 minutes, which is less than the normal select()
5101 timeout (5 or 10 minutes). This meant that the select() timeout never
5102 kicked in because it was always reset. I have fixed this by comparing the
5103 time when an interrupt arrives with the time at the start of the first call
5104 to select(). If more time than the timeout has elapsed, the interrupt is
5105 treated as a timeout.
5107 5. Some internal re-factoring in preparation for the addition of Sieve
5108 extensions (by MH). In particular, the "personal" test is moved to a
5109 separate function, and given an option for scanning Cc: and Bcc: (which is
5110 not set for Exim filters).
5112 6. When Exim created an email address using the login of the caller as the
5113 local part (e.g. when creating a From: or Sender: header line), it was not
5114 quoting the local part when it contained special characters such as @.
5116 7. Installed new OpenBSD configuration files.
5118 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5119 try to make them clearer.
5121 9. Callout options, other than the timeout value, were being ignored when
5122 verifying sender addresses in header lines. For example, when using
5124 verify = header_sender/callout=no_cache
5126 the cache was (incorrectly) being used.
5128 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5129 only the Cygwin environment.
5131 11. When return_path_on_delivery was set as a log selector, if different remote
5132 addresses in the same message used different return paths and parallel
5133 remote delivery occurred, the wrong values would sometimes be logged.
5134 (Whenever a remote delivery process finished, the return path value from
5135 the most recently started remote delivery process was logged.)
5137 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5138 lines when AUTH and/or TLS are in use. This is the "received protocol"
5139 field. Exim used to use "asmtp" for authenticated SMTP, without any
5140 indication (in the protocol name) for TLS use. Now it follows the RFC and
5141 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5142 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5143 names appear in log lines as well as in Received: header lines.
5145 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5146 extensions, and comparison tests, and to fix some bugs.
5148 14. Changes to the "personal" filter test:
5150 (1) The test was buggy in that it was just doing the equivalent of
5151 "contains" tests on header lines. For example, if a user's address was
5152 anne@some.where, the "personal" test would incorrectly be true for
5154 To: susanne@some.where
5156 This test is now done by extracting each address from the header in turn,
5157 and checking the entire address. Other tests that are part of "personal"
5158 are now done using regular expressions (for example, to check local parts
5159 of addresses in From: header lines).
5161 (2) The list of non-personal local parts in From: addresses has been
5162 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5163 taken from the Sieve specification recommendations.
5165 (3) If the message contains any header line starting with "List-" it is
5166 treated as non-personal.
5168 (4) The test for "circular" in the Subject: header line has been removed
5169 because it now seems ill-conceived.
5171 15. Minor typos in src/EDITME comments corrected.
5173 16. Installed latest exipick from John Jetmore.
5175 17. If headers_add on a router specified a text string that was too long for
5176 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5177 of string_sprintf() is now avoided.
5179 18. $message_body_size was not set (it was always zero) when running the DATA
5180 ACL and the local_scan() function.
5182 19. For the "mail" command in an Exim filter, no default was being set for
5183 the once_repeat time, causing a random time value to be used if "once" was
5184 specified. (If the value happened to be <= 0, no repeat happened.) The
5185 default is now 0s, meaning "never repeat". The "vacation" command was OK
5186 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5187 (I found it when inspecting the code).
5189 20. There is now an overall timeout for performing a callout verification. It
5190 defaults to 4 times the callout timeout, which applies to individual SMTP
5191 commands during the callout. The overall timeout applies when there is more
5192 than one host that can be tried. The timeout is checked before trying the
5193 next host. This prevents very long delays if there are a large number of
5194 hosts and all are timing out (e.g. when the network connections are timing
5195 out). The value of the overall timeout can be changed by specifying an
5196 additional sub-option for "callout", called "maxwait". For example:
5198 verify = sender/callout=5s,maxwait=20s
5200 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5201 to the end before writing, but this should make it even safer).
5203 22. Exim was forgetting that it had advertised PIPELINING for the second and
5204 subsequent messages on an SMTP connection. It was also not resetting its
5205 memory on STARTTLS and an internal HELO.
5207 23. When Exim logs an SMTP synchronization error within a session, it now
5208 records whether PIPELINING has been advertised or not.
5210 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5211 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5212 rather than long int.
5214 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5216 26. Added the never_mail option to autoreply.
5222 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5223 crash if the getsockname() call failed; this can happen if a connection is
5224 closed very soon after it is established. The problem was simply in the
5225 order in which certain operations were done, causing Exim to try to write
5226 to the SMTP stream before it had set up the file descriptor. The bug has
5227 been fixed by making things happen in the correct order.
5233 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5234 before the connection was closed, thus losing the rejection response.
5236 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5237 some early Solaris releases, but causes trouble in current releases where
5238 socklen_t is defined.
5240 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5243 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5246 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5247 that releasing the top of it at the end releases what was used for sub-
5248 expansions (unless the block got too big). However, discard this block if
5249 the first thing is a variable or header, so that we can use its block when
5250 it is dynamic (useful for very large $message_headers, for example).
5252 6. Lookups now cache *every* query, not just the most recent. A new, separate
5253 store pool is used for this. It can be recovered when all lookup caches are
5254 flushed. Lookups now release memory at the end of their result strings.
5255 This has involved some general refactoring of the lookup sources.
5257 7. Some code has been added to the store_xxx() functions to reduce the amount
5258 of flapping under certain conditions.
5260 8. log_incoming_interface used to affect only the <= reception log lines. Now
5261 it causes the local interface and port to be added to several more SMTP log
5262 lines, for example "SMTP connection from", and rejection lines.
5264 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5266 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5268 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5269 could be overwritten at the end of the current message (or the start of a
5270 new message if it was set in a HELO ACL). The value is now preserved for
5271 the duration of the SMTP connection.
5273 12. If a transport had a headers_rewrite setting, and a matching header line
5274 contained an unqualified address, that address was qualified, even if it
5275 did not match any rewriting rules. The underlying bug was that the values
5276 of the flags that permit the existence of unqualified sender and recipient
5277 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5278 non-local messages, and by -bnq for local messages) were not being
5279 preserved with the message after it was received.
5281 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5282 "next input=" as part of the text comprising the host identity instead of
5283 the correct text. The code was using the same buffer for two different
5284 strings. However, depending on which order the printing function evaluated
5285 its arguments, the bug did not always show up. Under Linux, for example, my
5286 test suite worked just fine.
5288 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5289 doesn't work with some older versions of Perl. It has been changed to "my",
5290 which in any case is probably the better facility to use.
5292 15. A really picky compiler found some instances of statements for creating
5293 error messages that either had too many or two few arguments for the format
5296 16. The size of the buffer for calls to the DNS resolver has been increased
5297 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5298 for addresses that have a lot of PTR records. This alleviates a problem; it
5299 does not fully solve it.
5301 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5302 buffer now truncates the list and logs the incident, which is the same
5303 action as happens when Exim is looking up a host name and its aliases.
5304 Previously in this situation something unpredictable would happen;
5305 sometimes it was "internal error: store_reset failed".
5307 18. If a server dropped the connection unexpectedly when an Exim client was
5308 using GnuTLS and trying to read a response, the client delivery process
5309 crashed while trying to generate an error log message.
5311 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5312 string, for example:
5314 warn message = H1: something\nH2: something
5316 the text was added as a single header line from Exim's point of view
5317 though it ended up OK in the delivered message. However, searching for the
5318 second and subsequent header lines using $h_h2: did not work. This has been
5319 fixed. Similarly, if a system filter added multiple headers in this way,
5320 the routers could not see them.
5322 20. Expanded the error message when iplsearch is called with an invalid key to
5323 suggest using net-iplsearch in a host list.
5325 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5326 ACLs are no longer actually imposed (and a message to that effect is
5329 22. If a "gecos" field in a passwd entry contained escaped characters, in
5330 particular, if it contained a \" sequence, Exim got it wrong when building
5331 a From: or a Sender: header from that name. A second bug also caused
5332 incorrect handling when an unquoted " was present following a character
5333 that needed quoting.
5335 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5336 was not being matched caselessly.
5338 24. Arranged for all hyphens in the exim.8 source to be escaped with
5341 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5342 properly. Recipient callout cache records were still being keyed to include
5343 the sender, even when use_sender was set false. This led to far more
5344 callouts that were necessary. The sender is no longer included in the key
5345 when use_sender is false.
5347 26. Added "control = submission" modifier to ACLs.
5349 27. Added the ${base62d: operator to decode base 62 numbers.
5351 28. dnsdb lookups can now access SRV records.
5353 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5354 the configuration file.
5356 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5357 (-v) mode. This makes the output for a verbose queue run more intelligible.
5359 31. Added a use_postmaster feature to recipient callouts.
5361 32. Added the $body_zerocount variable, containing the number of binary zero
5362 bytes in the message body.
5364 33. The time of last modification of the "new" subdirectory is now used as the
5365 "mailbox time last read" when there is a quota error for a maildir
5368 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5370 35. Added +ignore_unknown as a special item in host lists.
5372 36. Code for decoding IPv6 addresses in host lists is now included, even if
5373 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5374 address was recognized as an IP address, but was then not correctly decoded
5375 into binary, causing unexpected and incorrect effects when compared with
5382 1. Very minor rewording of debugging text in manualroute to say "list of
5383 hosts" instead of "hostlist".
5385 2. If verify=header_syntax was set, and a header line with an unqualified
5386 address (no domain) and a large number of spaces between the end of the
5387 name and the colon was received, the reception process suffered a buffer
5388 overflow, and (when I tested it) crashed. This was caused by some obsolete
5389 code that should have been removed. The fix is to remove it!
5391 3. When running in the test harness, delay a bit after writing a bounce
5392 message to get a bit more predictability in the log output.
5394 4. Added a call to search_tidyup() just before forking a reception process. In
5395 theory, someone could use a lookup in the expansion of smtp_accept_max_
5396 per_host which, without the tidyup, could leave open a database connection.
5398 5. Added the variables $recipient_data and $sender_data which get set from a
5399 lookup success in an ACL "recipients" or "senders" condition, or a router
5400 "senders" option, similar to $domain_data and $local_part_data.
5402 6. Moved the writing of debug_print from before to after the "senders" test
5405 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5406 problems for message scanning, either using a data ACL, or using
5407 local_scan() because the Received: header was not generated till after they
5408 were called (in order to set the time as the time of reception completion).
5409 I have revised the way this works. The header is now generated after the
5410 body is received, but before the ACL or local_scan() are called. After they
5411 are run, the timestamp in the header is updated.
5417 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5418 before starting a queue runner without re-exec. This happened only when
5419 deliver_drop_privilege was set or when the Exim user was set to root. The
5420 effect of the bug was that timeouts during subsequent deliveries caused
5421 crashes instead of being properly handled. The handler is now left at its
5422 default (and expected) setting.
5424 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5425 message, again when deliver_drop_privilege is set or Exim is run as root.
5426 The bug described in (1) was not present in this case, but the tidying up
5427 of the other signals was missing. I have made the two cases consistent.
5429 3. The ignore_target_hosts setting on a manualroute router was being ignored
5430 for hosts that were looked up using the /MX notation.
5432 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5435 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5436 operated on the sender address. After changing the $sender_address to <>
5437 for the sender address verify, Exim was re-instated it as the original
5438 (before rewriting) address, but remembering that it had rewritten it, so it
5439 wasn't rewriting it again. This bug also had the effect of breaking the
5440 sender address verification caching when the sender address was rewritten.
5442 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5443 This has been changed so that if the ip literal address matches
5444 ignore_target_hosts, the router declines.
5446 7. Added expansion conditions match_domain, match_address, and match_local_
5447 part (NOT match_host).
5449 8. The placeholder for the Received: header didn't have a length field set.
5451 9. Added code to Exim itself and to exim_lock to test for a specific race
5452 condition that could lead to file corruption when using MBX delivery. The
5453 issue is with the lockfile that is created in /tmp. If this file is removed
5454 after a process has opened it but before that process has acquired a lock,
5455 there is the potential for a second process to recreate the file and also
5456 acquire a lock. This could lead to two Exim processes writing to the file
5457 at the same time. The added code performs the same test as UW imapd; it
5458 checks after acquiring the lock that its file descriptor still refers to
5459 the same named file.
5461 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5462 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5465 11. Added the smtp_active_hostname option. If used, this will typically be made
5466 to depend on the incoming interface address. Because $interface_address is
5467 not set up until the daemon has forked a reception process, error responses
5468 that can happen earlier (such as "too many connections") no longer contain
5471 12. If an expansion in a condition on a "warn" statement fails because a lookup
5472 defers, the "warn" statement is abandoned, and the next ACL statement is
5473 processed. Previously this caused the whole ACL to be aborted.
5475 13. Added the iplsearch lookup type.
5477 14. Added ident_timeout as a log selector.
5479 15. Added tls_certificate_verified as a log selector.
5481 16. Added a global option tls_require_ciphers (compare the smtp transport
5482 option of the same name). This controls incoming TLS connections.
5484 17. I finally figured out how to make tls_require_ciphers do a similar thing
5485 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5486 before starting the TLS session.
5488 18. Tabs are now shown as \t in -bP output.
5490 19. If the log selector return_path_on_delivery was set, Exim crashed when
5491 bouncing a message because it had too many Received: header lines.
5493 20. If two routers both had headers_remove settings, and the first one included
5494 a superfluous trailing colon, the final name in the first list and the
5495 first name in the second list were incorrectly joined into one item (with a
5496 colon in the middle).
5502 1. Added -C and -D options to the exinext utility, mainly to make it easier
5503 to include in the automated testing, but these could be helpful when
5504 multiple configurations are in use.
5506 2. The exinext utility was not formatting the output nicely when there was
5507 an alternate port involved in the retry record key, nor when there was a
5508 message id as well (for retries that were specific to a specific message
5509 and a specific host). It was also confused by IPv6 addresses, because of
5510 the additional colons they contain. I have fixed the IPv4 problem, and
5511 patched it up to do a reasonable job for IPv6.
5513 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5514 delivery, the log line now contains "pipelined" if PIPELINING was used.
5516 4. An SMTP transport process used to panic and die if the bind() call to set
5517 an explicit outgoing interface failed. This has been changed; it is now
5518 treated in the same way as a connect() failure.
5520 5. A reference to $sender_host_name in the part of a conditional expansion
5521 that was being skipped was still causing a DNS lookup. This no longer
5524 6. The def: expansion condition was not recognizing references to header lines
5525 that used bh_ and bheader_.
5527 7. Added the _cache feature to named lists.
5529 8. The code for checking quota_filecount in the appendfile transport was
5530 allowing one more file than it should have been.
5532 9. For compatibility with Sendmail, the command line option
5540 and sets the incoming protocol and host name (for trusted callers). The
5541 host name and its colon can be omitted when only the protocol is to be set.
5542 Note the Exim already has two private options, -pd and -ps, that refer to
5543 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5544 "s", but I don't think that's a major issue.
5546 10. A number of refactoring changes to the code, none of which should affect
5549 (a) The number of logging options was getting close to filling up the
5550 32-bit word that was used as a bit map. I have split them into two classes:
5551 those that are passed in the argument to log_write(), and those that are
5552 only ever tested independently outside of that function. These are now in
5553 separate 32-bit words, so there is plenty of room for expansion again.
5554 There is no change in the user interface or the logging behaviour.
5556 (b) When building, for example, log lines, the code previously used a
5557 macro that called string_cat() twice, in order to add two strings. This is
5558 not really sufficiently general. Furthermore, there was one instance where
5559 it was actually wrong because one of the argument was used twice, and in
5560 one call a function was used. (As it happened, calling the function twice
5561 did not affect the overall behaviour.) The macro has been replaced by a
5562 function that can join an arbitrary number of extra strings onto a growing
5565 (c) The code for expansion conditions now uses a table and a binary chop
5566 instead of a serial search (which was left over from when there were very
5567 few conditions). Also, it now recognizes conditions like "pam" even when
5568 the relevant support is not compiled in: a suitably worded error message is
5569 given if an attempt is made to use such a condition.
5571 11. Added ${time_interval:xxxxx}.
5573 12. A bug was causing one of the ddress fields not to be passed back correctly
5574 from remote delivery subprocesses. The field in question was not being
5575 subsequently used, so this caused to problems in practice.
5577 13. Added new log selectors queue_time and deliver_time.
5579 14. Might have fixed a bug in maildirsizefile handling that threw up
5580 "unexpected character" debug warnings, and recalculated the data
5581 unnecessarily. In any case, I expanded the warning message to give more
5584 15. Added the message "Restricted characters in address" to the statements in
5585 the default ACL that block characters like @ and % in local parts.
5587 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5588 Three changes have been made:
5590 (a) There was a serious bug; a negative response to MAIL caused the whole
5591 recipient domain to be cached as invalid, thereby blocking all messages
5592 to all local parts at the same domain, from all senders. This bug has
5593 been fixed. The domain is no longer cached after a negative response to
5594 MAIL if the sender used is not empty.
5596 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5599 (c) A new callout option, "use_sender" has been added for people who want
5600 the modified behaviour.
5606 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5609 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5610 indeed breaks things for older releases.
5612 3. Added additional logging to the case where there is a problem reading data
5613 from a filter that is running in a subprocess using a pipe, in order to
5614 try to track down a specific problem.
5616 4. Testing facility fudge: when running in the test harness and attempting
5617 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5618 getting "No route to host". Convert this to a timeout.
5620 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5623 6. Some OS don't have socklen_t but use size_t instead. This affects the
5624 fifth argument of getsockopt() amongst other things. This is now
5625 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5626 can be set for individual OS. I have set it for SunOS5, OSF1, and
5627 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5628 some earlier ones do not.
5630 7. Change 4.30/15 was not doing the test caselessly.
5632 8. The standard form for an IPv6 address literal was being rejected by address
5633 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5634 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5635 this, as well as the form without the "IPv6" on the front (but only when
5636 address literals are enabled, of course).
5638 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5640 10. Exim crashed if a message with an empty sender address specified by -f
5641 encountered a router with an errors_to setting. This could be provoked only
5642 by a command such as
5646 where an empty string was supplied; "<>" did not hit this bug.
5648 11. Installed PCRE release 4.5.
5650 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5651 remained set. It is now erased.
5653 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5654 times from message ids (which are base 36 rather than the normal 62).
5656 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5657 were being counted as actual protocol errors, and logged if the log
5658 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5659 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5660 and DATA following a set of rejected RCPTs do not count as protocol errors.
5661 In other words, Exim assumes they were pipelined, though this may not
5662 actually be the case. Of course, in all cases the client gets an
5663 appropriate error code.
5665 15. If a lookup fails in an ACL condition, a message about the failure may
5666 be available; it is used if testing the ACL cannot continue, because most
5667 such messages specify what the cause of the deferral is. However, some
5668 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5669 that caused an old message to be retained and used if a later statement
5670 caused a defer, replacing the real cause of the deferral.
5672 16. If an IP address had so many PTR records that the DNS lookup buffer
5673 was not large enough to hold them, Exim could crash while trying to process
5674 the truncated data. It now detects and logs this case.
5676 17. Further to 4.21/58, another change has been made: if (and only if) the
5677 first line of a message (the first header line) ends with CRLF, a bare LF
5678 in a subsequent header line has a space inserted after it, so as not to
5679 terminate the header.
5681 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5682 unnecessarily, used atoi() instead of strtol(), and didn't check the
5683 termination when getting file sizes from file names by regex.
5685 19. Completely re-implemented the support for maildirsize files, in the light
5686 of a number of problems with the previous contributed implementation
5687 (4.30/29). In particular:
5689 . If the quota is zero, the maildirsize file is maintained, but no quota is
5692 . If the maildir directory does not exist, it is created before any attempt
5693 to write a maildirsize file.
5695 . The quota value in the file is just a cache; if the quota is changed in
5696 the transport, the new value overrides.
5698 . A regular expression is available for excluding directories from the
5701 20. The autoreply transport checks the characters in options that define the
5702 message's headers; it allows continued headers, but it was checking with
5703 isspace() after an embedded newline instead of explicitly looking for a
5706 21. If all the "regular" hosts to which an address was routed had passed their
5707 expiry times, and had not reached their retry times, the address was
5708 bounced, even if fallback hosts were defined. Now Exim should go on to try
5711 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5712 equivalent code in the SMTP transport. Some hosts send humungous responses
5713 to HELO/EHLO, more than 1024 it seems.
5715 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5716 gives compiler warnings in some environments. I've now done it "properly",
5719 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5720 (because of problems with the built-in one) was declared to return uschar *
5721 instead of char *, causing compiler failure.
5723 25. Fixed a file descriptor leak when processing alias/forward files.
5725 26. Fixed a minor format string issue in dbfn.c.
5727 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5729 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5730 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5733 29. When a sender address is verified, it is cached, to save repeating the test
5734 when there is more than one recipient in a message. However, when the
5735 verification involves a callout, it is possible for different callout
5736 options to be set for different recipients. It is too complicated to keep
5737 track of this in the cache, so now Exim always runs a verification when a
5738 callout is required, relying on the callout cache for the optimization.
5739 The overhead is duplication of the address routing, but this should not be
5742 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5743 to be verified with callout=postmaster, and the main callout worked but the
5744 postmaster check failed, the verification correctly failed. However, if a
5745 subsequent RCPT command asked for sender verification *without* the
5746 postmaster check, incorrect caching caused this verification also to fail,
5749 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5750 it was not caching the DNS options (qualify_single, search_parents) that
5751 were used when the lookup failed. A subsequent lookup with different
5752 options therefore always gave the same answer, though there were cases
5753 where it should not have. (Example: a "domains = !$mx_any" option on a
5754 dnslookup router: the "domains" option is always processed without any
5755 widening, but the router might have qualify_single set.) Now Exim uses the
5756 cached value only when the same options are set.
5758 32. Added John Jetmore's "exipick" utility to the distribution.
5760 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5761 than a timeout (e.g. a certificate is required, and is not provided), an
5762 Exim server now closes the connection immediately. Previously it waited for
5763 the client to close - but if the client is SSL, it seems that they each
5764 wait for each other, leading to a delay before one of them times out.
5766 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5767 maintained 0.8.x compatibility because I don't think many are using it, and
5768 it is clearly obsolete.
5770 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5773 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5774 client certificate was expired. A simple patch fixes this, though I don't
5775 understand the full logic of why the verify callback is called multiple
5778 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5779 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5780 which causes problems with some clients (such as the Certicom SSL Plus
5781 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5782 disables the coutermeasure allowing Eudora to connect."
5784 38. Exim was not checking that a write() to a log file succeeded. This could
5785 lead to Bad Things if a log got too big, in particular if it hit a file
5786 size limit. Exim now panics and dies if it cannot write to a log file, just
5787 as it does if it cannot open a log file.
5789 39. Modified OS/Makefile-Linux so that it now contains
5791 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5793 The two -D definitions ensure that Exim is compiled with large file
5794 support, which makes it possible to handle log files that are bigger than
5797 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5798 instance) a domain was checked against a named list that involved a lookup,
5799 causing $domain_data to be set, then another domain was checked against the
5800 same list, then the first domain was re-checked, the value of $domain_data
5801 after the final check could be wrong. In particular, if the second check
5802 failed, it could be set empty. This bug probably also applied to
5805 41. The strip_trailing_dot option was not being applied to the address given
5806 with the -f command-line option.
5808 42. The code for reading a message's header from the spool was incrementing
5809 $received_count, but never initializing it. This meant that the value was
5810 incorrect (doubled) while delivering a message in the same process in which
5811 it was received. In the most common configuration of Exim, this never
5812 happens - a fresh exec is done - but it can happen when
5813 deliver_drop_privilege is set.
5815 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5816 it now includes up to 150 characters of the unexpected data in the log
5819 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5820 and building data strings. The size of both of these buffers was 10 000
5821 bytes - far larger than anybody would *ever* want, thought I. Needless to
5822 say, somebody hit the limit. I have increased the maximum line length to
5823 20 000 and the maximum data length of concatenated lines to 100 000. I have
5824 also fixed two bugs, because there was no checking on these buffers. Tsk,
5825 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5826 buffer is too small.
5828 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5829 lsearch lookups. Now it does.
5831 46. When parsing a route_list item in a manualroute router, a fixed-length
5832 buffer was used for the list of hosts. I made this 1024 bytes long,
5833 thinking that nobody would ever have a list of hosts that long. Wrong.
5834 Somebody had a whole pile of complicated expansion conditions, and the
5835 string was silently truncated, leading to an expansion error. It turns out
5836 that it is easier to change to an unlimited length (owing to other changes
5837 that have happened since this code was originally written) than to build
5838 structure for giving a limitation error. The length of the item that
5839 expands into the list of hosts is now unlimited.
5841 47. The lsearch lookup could not handle data where the length of text line was
5842 more than 4095 characters. Such lines were truncated, leading to shortened
5843 data being returned. It should now handle lines of any length.
5845 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5846 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5849 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5850 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5851 no longer happens in comment lines. A list of replacements is now placed
5852 at the head of all of the source files, except those whose only change is
5853 to replace PERL_COMMAND in the very first #! line.
5855 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5856 messages on the queue, with a bottom-up merge sort, using code contributed
5857 by Michael Haardt. This should make operations like -bp somewhat faster on
5858 large queues. It won't affect queue runners, except when queue_run_in_order
5861 51. Installed eximstats 1.31 in the distribution.
5863 52. Added support for SRV lookups to the dnslookup router.
5865 53. If an ACL referred to $message_body or $message_body_end, the value was not
5866 reset for any messages that followed in the same SMTP session.
5868 54. The store-handling optimization for building very long strings was not
5869 differentiating between the different store pools. I don't think this
5870 actually made any difference in practice, but I've tidied it.
5872 55. While running the routers to verify a sender address, $sender_address
5873 was still set to the sender address. This is wrong, because when routing to
5874 send a bounce to the sender, it would be empty. Therefore, I have changed
5875 it so that, while verifying a sender address, $sender_address is set to <>.
5876 (There is no change to what happens when verifying a recipient address.)
5878 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5879 target A or AAAA records (if not already returned) without resetting the
5880 qualify_single or search_parents options of the DNS resolver. These are
5881 inappropriate in this case because the targets of MX and SRV records must
5882 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5883 target that, when qualified, matched something in the local domain. These
5884 two options are now turned off when doing these lookups.
5886 57. It seems that at least some releases of Reiserfs (which does not have the
5887 concept of a fixed number of inodes) returns zero and not -1 for the
5888 number of available inodes. This interacted badly with check_spool_inodes,
5889 which assumed that -1 was the "no such thing" setting. What I have done is
5890 to check that the total number of inodes is greater than zero before doing
5891 the test of how many are available.
5893 58. When a "warn" ACL statement has a log_message modifier, the message is
5894 remembered, and not repeated. This is to avoid a lot of repetition when a
5895 message has many recipients that cause the same warning to be written.
5896 However, Exim was preserving the list of already written lines for an
5897 entire SMTP session, which doesn't seem right. The memory is now reset if a
5898 new message is started.
5900 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5901 debugging output unless log_rewrite was also set.
5903 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5904 of (int)(handle) when we know that handle contains (void *)(-1).
5906 61. The Exim daemon panic-logs an error return when it closes the incoming
5907 connection. However "connection reset by peer" seems to be common, and
5908 isn't really an error worthy of noting specially, so that particular error
5911 62. When Exim is trying to find all the local interfaces, it used to panic and
5912 die if the ioctl to get the interface flags failed. However, it seems that
5913 on at least one OS (Solaris 9) it is possible to have an interface that is
5914 included in the list of interfaces, but for which you get a failure error
5915 for this call. This happens when the interface is not "plumbed" into a
5916 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5917 failure of the "get flags" call assumes that the interface is down.
5919 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5920 makes life easier for people who are doing arithmetic on fields extracted
5921 from dates, where you often get leading zeros that should not be
5922 interpreted as octal.
5924 64. Added qualify_domain to the redirect router, to override the global
5927 65. If a pathologically long header line contained very many addresses (the
5928 report of this problem mentioned 10 000) and each of them was rewritten,
5929 Exim could use up a very large amount of memory. (It kept on making new
5930 copies of the header line as it rewrote, and never released the old ones.)
5931 At the expense of a bit more processing, the header rewriting function has
5932 been changed so that it no longer eats memory in this way.
5934 66. The generation of the Received: header has been moved from the time that a
5935 message starts to be received, to the time that it finishes. The timestamp
5936 in the Received: header should now be very close to that of the <= log
5937 line. There are two side-effects of this change:
5939 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5940 logged header lines no longer include the local Received: line, because
5941 it has not yet been created. The same applies to a copy of the message
5942 that is returned to a non-SMTP sender when a message is rejected.
5944 (b) When a filter file is tested using -bf, no additional Received: header
5945 is added to the test message. After some thought, I decided that this
5948 This change does not affect the value of $received_for. It is still set
5949 after address rewriting, but before local_scan() is called.
5951 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5953 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5954 gave an unhelpful panic error message, and a defer error. I have managed to
5955 change this behaviour so that it now rejects any supplied certificate,
5956 which seems right, as the list of acceptable certificates is empty.
5958 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5959 gave an unhelpful defer error. I have not managed to make this reject any
5960 supplied certificates, but the error message it gives is "no certificate
5961 supplied", which is not helpful.
5963 70. exigrep's output now also includes lines that are not associated with any
5964 message, but which match the given pattern. Implemented by a patch from
5965 Martin Sluka, which also tidied up the Perl a bit.
5967 71. Recipient callout verification, like sender verification, was using <> in
5968 the MAIL FROM command. This isn't really the right thing, since the actual
5969 sender may affect whether the remote host accepts the recipient or not. I
5970 have changed it to use the actual sender in the callout; this means that
5971 the cache record is now keyed on a recipient/sender pair, not just the
5972 recipient address. There doesn't seem to be a real danger of callout loops,
5973 since a callout by the remote host to check the sender would use <>.
5974 [SEE ABOVE: changed after hitting problems.]
5976 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5977 temporary errors. However, in the case of such a code being given after
5978 the end of a data transmission (i.e. after ".") Exim was failing to write
5979 a retry record for the message. (Yes, there was some broken host that was
5980 actually sending 8xx at this point.)
5982 73. An unknown lookup type in a host list could cause Exim to panic-die when
5983 the list was checked. (An example that provoked this was putting <; in the
5984 middle of a list instead of at the start.) If this happened during a DATA
5985 ACL check, a -D file could be left lying around. This kind of configuration
5986 error no longer causes Exim to die; instead it causes a defer error. The
5987 incident is still logged to the main and panic logs.
5989 74. Buglet left over from Exim 3 conversion. The message "too many messages
5990 in one connection" was written to the rejectlog but not the mainlog, except
5991 when address rewriting (yes!) was being logged.
5993 75. Added write_rejectlog option.
5995 76. When a system filter was run not as root (that is, when system_filter_user
5996 was set), the values of the $n variables were not being returned to the
5997 main process; thus, they were not subsequently available in the $sn
6000 77. Added +return_path_on_delivery log selector.
6002 78. A connection timeout was being treated differently from recipients deferred
6003 when testing hosts_max_try with a message that was older than the host's
6004 retry timeout. (The host should not be counted, thus allowing all hosts to
6005 be tried at least once before bouncing.) This may have been the cause of an
6006 occasionally reported bug whereby a message would remain on the queue
6007 longer than the retry timeout, but would be bounced if a delivery was
6008 forced. I say "may" because I never totally pinned down the problem;
6009 setting up timeout/retry tests is difficult. See also the next item.
6011 79. The ultimate address timeout was not being applied to errors that involved
6012 a combination of host plus message (for example, a timeout on a MAIL
6013 command). When an address resolved to a number of possible hosts, and they
6014 were not all tried for each delivery (e.g. because of hosts_max_try), a
6015 message could remain on the queue longer than the retry timeout.
6017 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6020 81. Fixed an obscure SMTP outgoing bug which required at least the following
6021 conditions: (a) there was another message waiting for the same server;
6022 (b) the server returned 5xx to all RCPT commands in the first message so
6023 that the message was not completed; (c) the server dropped the connection
6024 or gave a negative response to the RSET that Exim sends to abort the
6025 transaction. The observed case was a dropped connection after DATA that had
6026 been sent in pipelining mode. That is, the server had advertised PIPELINING
6027 but was not implementing it correctly. The effect of the bug was incorrect
6028 behaviour, such as trying another host, and this could lead to a crash.
6034 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6035 and daemon.c were passed as pointers to ints; they should have been
6036 pointers to socklen_t variables (which are typically unsigned ints).
6038 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6041 3. Fixed a really odd bug that affected only the testing scheme; patching a
6042 certain fixed string in the binary changed the value of another string that
6043 happened to be identical to the end of the original first string.
6045 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6046 name", it returns that address as the IP address. On some operating
6047 systems (e.g. Solaris), it also passes back the IP address string as the
6048 "host name". However, on others (e.g. Linux), it passes back an empty
6049 string. Exim wasn't checking for this, and was changing the host name to an
6050 empty string, assuming it had been canonicalized.
6052 5. Although rare, it is permitted to have more than one PTR record for a given
6053 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6054 all the names associated with an address, because they do in Solaris.
6055 However, it seems that they do not in Linux for data that comes from the
6056 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6057 I found this out when I moved to a new Linux workstation and tried to run
6058 the Exim test suite.
6060 To get round this problem I have changed the code so that it now does its
6061 own call to the DNS to look up PTR records when searching for a host name.
6062 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6063 addresses that are only in /etc/hosts are still found.
6065 This behaviour is, however, controlled by an option called host_lookup_
6066 order, which defaults to "bydns:byaddr". If people want to use the other
6067 order, or indeed, just use one or the other means of lookup, they can
6068 specify it in this variable.
6070 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6071 some operating systems, this comes back from gethostbyaddr() as an empty
6072 string, and this is what Exim used to test for. However, it seems that in
6073 other systems, "." is yielded. Exim now tests for this case too.
6075 7. The values of check_spool_space and check_log_space are now held internally
6076 as a number of kilobytes instead of an absolute number of bytes. If a
6077 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6078 kilobyte. This means that much larger values can be stored.
6080 8. Exim monitor: an attempt to get the action menu when not actually pointing
6081 at a message produces an empty menu entitled "No message selected". This
6082 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6083 no entries in it ("Shell widget menu has zero width and/or height"). So I
6084 have added a single, blank menu entry in this case.
6086 9. Added ${quote_local_part.
6088 10. MIME decoding is now applied to the contents of Subject: header lines when
6091 11. Now that a reference to $sender_host_address automatically causes a reverse
6092 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6093 host lookup before query-style lookups in lists that might use this
6094 variable. This has therefore been abolished, and the "net-" prefix is no
6095 longer necessary for query-style lookups.
6097 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6098 have been a typo for LFLAGS, so it has been changed.
6100 13. The install script calls Exim with "-C /dev/null" in order to find the
6101 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6102 to be output. However, since Exim outputs its version number before the
6103 error, it didn't break the script. It just looked ugly. I fixed this by
6104 always allowing "-C /dev/null" if the caller is root.
6106 14. Ignore overlarge ACL variable number when reading spool file - insurance
6107 against a later release with more variables having written the file.
6109 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6110 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6111 this, as well as the form without the "IPv6" on the front.
6113 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6114 OS/Makefile-Darwin file.
6116 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6117 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6119 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6120 variable. However, some OS use ints for inodes. I've added cast to long int
6121 to get rid of the compiler warning.
6123 19. I had forgotten to lock out "/../" in configuration file names when
6124 ALT_CONFIG_PREFIX was set.
6126 20. Routers used for verification do not need to specify transports. However,
6127 if such a router generated a host list, and callout was configured, Exim
6128 crashed, because it could not find a port number from the (non-existent)
6129 transport. It now assumes port 25 in this circumstance.
6131 21. Added the -t option to exigrep.
6133 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6134 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6135 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6136 them is set, LOOKUP_LSEARCH is forced.
6138 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6140 24. Added sender and host information to the "rejected by local_scan()" log
6141 line; previously there was no indication of these.
6143 25. Added .include_if_exists.
6145 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6146 Linux. It turns out that not all file systems support this. Apparently some
6147 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6148 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6149 not supported on the file descriptor, is now ignored when Exim is trying to
6150 sync a directory. This applies only to Linux.
6152 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6154 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6155 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6158 29. Added support for maildirsize files from supplied patch (modified a bit).
6160 30. The use of :fail: followed by an empty string could lead Exim to respond to
6161 sender verification failures with (e.g.):
6163 550 Verification failed for <xxx>
6164 550 Sender verify failed
6166 where the first response line was missing the '-' that indicates it is not
6167 the final line of the response.
6169 31. The loop for finding the name of the user that called Exim had a hardwired
6170 limit of 10; it now uses the value of finduser_retries, which is used for
6171 all other user lookups.
6173 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6176 33. Exim was neglecting to zero errno before one call of strtol() when
6177 expanding a string and expecting an integer value. On some systems this
6178 resulted in spurious "integer overflow" errors. Also, it was casting the
6179 result into an int without checking.
6181 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6182 did not work. The code looks as if it has *never* worked, though it appears
6183 to have been documented since at least release 1.62. I have made it work.
6185 35. The "timeout_DNS" error in retry rules, also documented since at least
6186 1.62, also never worked. As it isn't clear exactly what this means, and
6187 clearly it isn't a major issue, I have abolished the feature by treating it
6188 as "timeout", and writing a warning to the main and panic logs.
6190 36. The display of retry rules for -brt wasn't always showing the error code
6193 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6194 timeout_connect_A, timeout_connect_MX.
6196 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6197 to the empty sender.
6199 39. The daemon was not analysing the content of -oX till after it had closed
6200 stderr and disconnected from the controlling terminal. This meant that any
6201 syntax errors were only noted on the panic log, and the return code from
6202 the command was 0. By re-arranging the code a little, I've made the
6203 decoding happen first, so such errors now appear on stderr, and the return
6204 code is 1. However, the actual setting up of the sockets still happens in
6205 the disconnected process, so errors there are still only recorded on the
6208 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6209 connections (as happens on some IP stacks) was logged at start up time as
6210 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6211 it from "IPv6 and IPv4", which means that two separate sockets are being
6214 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6215 says whether AF_INET or AF_INET6 was passed as an argument.
6217 42. Exiwhat output was messed up when time zones were included in log
6220 43. Exiwhat now gives more information about the daemon's listening ports,
6221 and whether -tls-on-connect was used.
6223 44. The "port" option of the smtp transport is now expanded.
6225 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6226 silently ignored. Now an error message is written to the main and panic
6229 46. There's a new ACL modifier called "logwrite" which writes to a log file
6230 as soon as it is encountered.
6232 47. Added $local_user_uid and $local_user_gid at routing time.
6234 48. Exim crashed when trying to verify a sender address that was being
6237 49. Exim was recognizing only a space character after ".include". It now also
6238 recognizes a tab character.
6240 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6241 extracting the relevant information from the specification. The man page no
6242 longer contains scrambled data for the -d option, and I've added a section
6243 at the front about calling Exim under different names.
6245 51. Added "extra_headers" argument to the "mail" command in filter files.
6247 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6250 53. Installed eximstats 1.29.
6252 54. Added transport_filter_timeout as a generic transport option.
6254 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6255 Cc: header lines. This was required by RFC 822, but it not required by RFC
6258 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6259 incoming messages that did not have them. Now it does so only if the
6260 message originates locally, that is, if there is no associated remote host
6261 address. When Resent- header lines are present, this applies to the Resent-
6262 lines rather than the non-Resent- lines.
6264 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6265 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6267 58. Messages for configuration errors now include the name of the main
6268 configuration file - useful now that there may be more than one file in a
6269 list (.included file names were always shown).
6271 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6272 for those rare installations that do not start the daemon as root or run it
6273 setuid root. I've cut out the call to initgroups() if the daemon is not
6276 60. The Exim user and group can now be bound into the binary as text strings
6277 that are looked up at the start of Exim's processing.
6279 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6281 62. Added $mailstore_basename variable.
6283 63. Installed patch to sieve.c from Michael Haardt.
6285 64. When Exim failed to open the panic log after failing to open the main log,
6286 the original message it was trying to log was written to stderr and debug
6287 output, but if they were not available (the usual case in production), it
6288 was lost. Now it is written to syslog before the two lines that record the
6289 failures to open the logs.
6291 65. Users' Exim filters run in subprocesses under the user's uid. It is
6292 possible for a "deliver" command or an alias in a "personal" command to
6293 provoke an address rewrite. If logging of address rewriting is configured,
6294 this fails because the process is not running as root or exim. There may be
6295 a better way of dealing with this, but for the moment (because 4.30 needs
6296 to be released), I have disabled address rewrite logging when running a
6297 filter in a non-root, non-exim process.
6303 1. The buildconfig auxiliary program wasn't quoting the value set for
6304 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6305 not defined. This bug was masked in 4.22 by the effect that was fixed in
6308 2. Some messages that were rejected after a message id was allocated were
6309 shown as "incomplete" by exigrep. It no longer does this for messages that
6310 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6312 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6313 have allow_domain_literals set, the ID did not get logged in the <= line.
6314 Domain literals are now always recognized in Message-ID: header lines.
6316 4. The first argument for a ${extract expansion item is the key name or field
6317 number. Leading and trailing spaces in this item were not being ignored,
6318 causing some misleading effects.
6320 5. When deliver_drop_privilege was set, single queue runner processes started
6321 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6322 same command in the process it spins off) were not dropping privilege.
6324 6. When the daemon running as "exim" started a queue runner, it always
6325 re-executed Exim in the spun-off process. This is a waste of effort when
6326 deliver_drop_privilege is set. The new process now just calls the
6327 queue-runner function directly.
6333 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6336 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6337 ignored. Though the use of -oP was forcing the writing of a pid file, it
6338 was always written to the default place.
6340 3. If the message "no IP address found for host xxxx" is generated during
6341 incoming verification, it is now followed by identification of the incoming
6342 connection (so you can more easily find what provoked it).
6344 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6346 5. Added some features to "harden" Exim a bit more against certain attacks:
6348 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6349 be put in Local/Makefile. This is like the never_users runtime option,
6350 but it cannot be overridden. The default setting is "root".
6352 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6353 prefix string with which any file named in a -C command line option
6356 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6357 is retained for -C and -D only if the caller of Exim is root. Without
6358 it, the exim user may also use -C and -D and retain privilege.
6360 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6361 command line option is disabled.
6363 6. Macro names set by the -D option must start with an upper case letter, just
6364 like macro names defined in the configuration file.
6366 7. Added "dereference=" facility to LDAP.
6368 8. Two instances of the typo "uknown" in the source files are fixed.
6370 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6371 the Configure-Makefile script screwed up while processing it.
6373 10. Incorporated PCRE 4.4.
6375 11. The SMTP synchronization check was not operating right at the start of an
6376 SMTP session. For example, it could not catch a HELO sent before the client
6377 waited for the greeting. There is now a check for outstanding input at the
6378 point when the greeting is written. Because of the duplex, asynchronous
6379 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6380 way, but not yet received, when the check is performed.
6382 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6383 on TCP/IP sockets, because this apparently causes some broken clients to
6386 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6387 unchanged) from the Cygwin maintainer.
6389 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6390 was supported instead of testing for mbx. Effectively a typo.
6392 15. The spa authenticator server code was not checking that the input it
6393 received was valid base64.
6395 16. The debug output line for the "set" modifier in ACLs was not showing the
6396 name of the variable that was being set.
6398 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6400 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6401 lookup on its own. Something else was needed to trigger the lookup. For
6402 example, a match in host_lookup or the need for a host name in a host list.
6403 Now, if $sender_host_name is referenced and the host name has not yet been
6404 looked up, a lookup is performed. If the lookup fails, the variable remains
6405 empty, and $host_lookup_failed is set to "1".
6407 19. Added "eqi" as a case-independent comparison operator.
6409 20. The saslauthd authentication condition could segfault if neither service
6410 nor realm was specified.
6412 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6413 error message that was logged was misleading, and incoming SMTP
6414 connections were dropped. The message is now more accurate, and temporary
6415 errors are given to SMTP connections.
6417 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6418 (except RSET) with a 421 error, until QUIT is received. However, it was
6419 failing to send a response to QUIT.
6421 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6422 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6423 ACL is now run after the helo_try_verify_hosts code.
6425 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6426 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6427 case-independent, so other case variants are also recognized). Apparently
6428 some systems use these upper case variants.
6430 25. If more than two messages were waiting for the same host, and a transport
6431 filter was specified for the transport, Exim sent two messages over the
6432 same TCP/IP connection, and then failed with "socket operation on non-
6433 socket" when it tried to send the third.
6435 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6437 27. The extern definition of crypt16() in expand.c was not being excluded when
6438 the OS had its own crypt16() function.
6440 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6441 as a preferred synonym for return_size_limit, both as an option and as an
6444 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6446 30. Changed the default configuration ACL to relax the local part checking rule
6447 for addresses that are not in any local domains. For these addresses,
6448 slashes and pipe symbols are allowed within local parts, but the sequence
6449 /../ is explicitly forbidden.
6451 31. SPA server authentication was not clearing the challenge buffer before
6454 32. log_message in a "warn" ACL statement was writing to the reject log as
6455 well as to the main log, which contradicts the documentation and doesn't
6456 seem right (because no rejection is happening). So I have stopped it.
6458 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6459 However, I am unable to do any testing of this.
6461 34. Fixed an infelicity in the appendfile transport. When checking directories
6462 for a mailbox, to see if any needed to be created, it was accidentally
6463 using path names with one or more superfluous leading slashes; tracing
6464 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6466 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6467 contents are added to the log line that is written for every discarded
6468 recipient. (Previously a log_message setting was ignored.)
6470 36. The ${quote: operator now quotes the string if it is empty.
6472 37. The install script runs exim in order to find its version number. If for
6473 some reason other than non-existence or emptiness, which it checks, it
6474 could not run './exim', it was installing it with an empty version number,
6475 i.e. as "exim-". This error state is now caught, and the installation is
6478 38. An argument was missing from the function that creates an error message
6479 when Exim fails to connect to the socket for saslauthd authentication.
6480 This could cause Exim to crash, or give a corrupted message.
6482 39. Added isip, isip4, and isip6 to ${if conditions.
6484 40. The ACL variables $acl_xx are now saved with the message, and can be
6485 accessed later in routers, transports, and filters.
6487 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6488 strings in the file are not string-expanded.
6490 42. If a MAIL command specified a SIZE value that was too large to fit into an
6491 int variable, the check against message_size_limit failed. Such values are
6492 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6493 day this will have to be increased, but I don't think I want to be around
6494 when emails are that large.
6501 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6502 iconv() is not standard in FreeBSD.
6504 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6505 IPv6 enabled. The observed symptom was a segmentation fault on return from
6506 the function os_common_find_running_interfaces() in src/os.c.
6508 3. In the check_special_case() function in daemon.c I had used "errno" as an
6509 argument name, which causes warnings on some systems. This was basically a
6510 typo, since it was named "eno" in the comments!
6512 4. The code that waits for the clock to tick (at a resolution of some fraction
6513 of a second) so as to ensure message-id uniqueness was always waiting for
6514 at least one whole tick, when it could have waited for less. [This is
6515 almost certainly not relevant at current processor speeds, where it is
6516 unlikely to ever wait at all. But we try to future-proof.]
6518 5. The function that sleeps for a time interval that includes fractions of a
6519 second contained a race. It did not block SIGALRM between setting the
6520 timer, and suspending (a couple of lines later). If the interval was short
6521 and the sigsuspend() was delayed until after it had expired, the suspension
6522 never ended. On busy systems this could lead to processes getting stuck for
6525 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6526 process, before it forks any delivery processes. The open lookup caching
6527 mechanism meant that the open file or database connection was passed into
6528 the delivery process. The problem was that delivery processes always tidy
6529 up cached lookup data. This could cause a problem for the next delivery
6530 process started by the queue runner, because the external queue runner
6531 process does not know about the closure. So the next delivery process
6532 still has data in the lookup cache. In the case of a file lookup, there was
6533 no problem because closing a file descriptor in a subprocess doesn't affect
6534 the parent. However, if the lookup was caching a connection to a database,
6535 the connection was closed, and the second delivery process was likely to
6536 see errors such as "PGSQL: query failed: server closed the connection
6537 unexpectedly". The problem has been fixed by closing all cached lookups
6538 in a queue runner before running a delivery process.
6540 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6541 seem to have the "const" qualifier which it has on other OS. I've
6544 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6545 *of the same type* (client or server) with the same public name that an
6546 error should be diagnosed.
6548 9. When Exim looked up a host name for an IP address, but failed to find the
6549 original IP address when looking up the host name (a safety check), it
6550 output the message "<ip address> does not match any IP for NULL", which was
6551 confusing, to say the least. The bug was that the host name should have
6552 appeared instead of "NULL".
6554 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6555 user that is built into the binary, and the -C or -D options is used, root
6556 privilege is dropped before the configuration file is read. In addition,
6557 logging is switched to stderr instead of the normal log files. If the
6558 configuration then re-defines the Exim user, the unprivileged environment
6559 is probably not what is expected, so Exim logs a panic warning message (but
6562 However, if deliver_drop_privilege is set, the unprivileged state may well
6563 be exactly what is intended, so the warning has been cut out in that case,
6564 and Exim is allowed to try to write to its normal log files.
6570 1. smtp_return_error_details was not giving details for temporary sender
6571 or receiver verification errors.
6573 2. Diagnose a configuration error if two authenticators have the same public
6576 3. Exim used not to create the message log file for a message until the first
6577 delivery attempt. This could be confusing when incoming messages were held
6578 for policy or load reasons. The message log file is now created at the time
6579 the message is received, and an initial "Received" line is written to it.
6581 4. The automatically generated man page for command line options had a minor
6582 bug that caused no ill effects; however, a more serious problem was that
6583 the procedure for building the man page automatically didn't always
6584 operate. Consequently, release 4.20 contains an out-of-date version. This
6585 shouldn't happen again.
6587 5. When building Exim with embedded Perl support, the script that builds the
6588 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6589 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6591 6. The freeze_tell option was not being used for messages that were frozen on
6592 arrival, either by an ACL or by local_scan().
6594 7. Added the smtp_incomplete_transaction log selector.
6596 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6597 was accepting AUTH without a new EHLO.
6599 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6600 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6603 10. Logging of TCP/IP connections (when configured) now happens in the main
6604 daemon process instead of the child process, so that the TCP/IP connection
6605 count is more accurate (but it can never be perfect).
6607 11. The use of "drop" in a nested ACL was not being handled correctly in the
6608 outer ACL. Now, if condition failure induced by the nested "drop" causes
6609 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6610 or "require"), the connection is dropped.
6612 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6613 that yield "discard" can now be used with an "accept" or a "discard" verb,
6614 but an error is generated for any others (because I can't see a useful way
6615 to define what should happen).
6617 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6618 are now processed in the same way as lines in the Exim configuration file.
6619 In particular, continuation lines are supported.
6621 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6623 15. Added -ti meaning -t -i.
6625 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6626 domains, and warn by logging if others are found.
6628 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6629 returned from GIFCONF when Exim is trying to find the list of interfaces on
6630 a host. The code in os.c has been modified to copy each ifreq to an aligned
6631 structure in all cases.
6633 Also, in some cases, the returned ifreq's were being copied to a 'struct
6634 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6635 means the last couple of bytes of an IPv6 address could be chopped if the
6636 ifreq contained only a normal sockaddr (14 bytes storage).
6638 18. Named domain lists were not supported in the hosts_treat_as_local option.
6639 An entry such as +xxxx was not recognized, and was treated as a literal
6642 19. Ensure that header lines added by a DATA ACL are included in the reject log
6643 if the ACL subsequently rejects the message.
6645 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6646 MD5 (which is deprecated).
6648 21. When testing a filter file using -bf, Exim was writing a message when it
6649 took the sender from a "From " line in the message, but it was not doing so
6650 when it took $return_path from a Return-Path: header line. It now does.
6652 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6653 with a valid header line field name (a series of printing characters
6654 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6656 23. Changed "disc" in the source to "disk" to conform to the documentation and
6657 the book and for uniformity.
6659 24. Ignore Sendmail's -Ooption=value command line item.
6661 25. When execve() failed while trying to run a command in a pipe transport,
6662 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6663 could be confused with a return value of 69 from the command itself. This
6664 has been changed to 127, the value the shell returns if it is asked to run
6665 a non-existent command. The wording for the related log line suggests a
6666 non-existent command as the problem.
6668 26. If received_header_text expands to an empty string, do not add a Received:
6669 header line to the message. (Well, it adds a token one on the spool, but
6670 marks it "old" so that it doesn't get used or transmitted.)
6672 27. Installed eximstats 1.28 (addition of -nt option).
6674 28. There was no check for failure on the call to getsockname() in the daemon
6675 code. This can fail if there is a shortage of resources on the system, with
6676 ENOMEM, for example. A temporary error is now given on failure.
6678 29. Contrary to the C standard, it seems that in some environments, the
6679 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6680 program. Exim now does this explicitly; it affects the formatting of
6681 timestamps using strftime().
6683 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6684 complaints. I've now initialized all the variables, to avoid this.
6686 32. Header lines added by a system filter were not being "seen" during
6687 transport-time rewrites.
6689 33. The info_callback() function passed to OpenSSL is set up with type void
6690 (*)(SSL *, int, int), as described somewhere. However, when calling the
6691 function (actually a macro) that sets it up, the type void(*)() is
6692 expected. I've put in a cast to prevent warnings from picky compilers.
6694 34. If a DNS black list lookup found a CNAME record, but there were no A
6695 records associated with the domain it pointed at, Exim crashed.
6697 35. If a DNS black list lookup returned more than one A record, Exim ignored
6698 all but the first. It now scans all returned addresses if a particular IP
6699 value is being sought. In this situation, the contents of the
6700 $dnslist_value variable are a list of all the addresses, separated by a
6703 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6704 to accept a host name and all its aliases if the forward lookup for any of
6705 them yielded the IP address of the incoming connection. Now it accepts only
6706 those names whose forward lookup yields the correct IP address. Any other
6707 names are discarded. This closes a loophole whereby a rogue DNS
6708 administrator could create reverse DNS records to break through a
6709 wildcarded host restriction in an ACL.
6711 37. If a user filter or a system filter that ran in a subprocess used any of
6712 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6713 the wrong values were passed to the pipe command ($thisaddress had the
6714 value of $0, $0 had the value of $1, etc). This bug was introduced by
6715 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6717 38. Improved the line breaking for long SMTP error messages from ACLs.
6718 Previously, if there was no break point between 40 and 75 characters, Exim
6719 left the rest of the message alone. Two changes have been made: (a) I've
6720 reduced the minimum length to 35 characters; (b) if it can't find a break
6721 point between 35 and 75 characters, it looks ahead and uses the first one
6722 that it finds. This may give the occasional overlong line, but at least the
6723 remaining text gets split now.
6725 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6726 file descriptors might be low, and that setting 1000 would always raise it.
6727 It turns out that in some environments, the limit is already over 1000 and
6728 that lowering it causes trouble. So now Exim takes care not to decrease it.
6730 40. When delivering a message, the value of $return_path is set to $sender_
6731 address at the start of routing (routers may change the value). By an
6732 oversight, this default was not being set up when an address was tested by
6733 -bt or -bv, which affected the outcome if any router or filter referred to
6736 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6737 log only when "message" or "log_message" are set. However, if one of the
6738 conditions was an address verification, or a call to a nested ACL, the
6739 messages generated by the underlying test were being passed through. This
6740 no longer happens. The underlying message is available in $acl_verify_
6741 message for both "message" and "log_message" expansions, so it can be
6742 passed through if needed.
6744 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6745 new expansion $bh_ to give the encoded byte string without charset
6746 translation. Translation happens only if iconv() is available; HAVE_ICONV
6747 indicates this at build time. HEADERS_CHARSET gives the charset to
6748 translate to; headers_charset can change it in the configuration, and
6749 "headers charset" can change it in an individual filter file.
6751 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6752 that creates RFC 2047 encoded "words" labels them as that charset instead
6753 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6754 expansion operator; (ii) when Exim creates a From: line for a local
6755 message; (iii) when a header line is rewritten to include a "phrase" part.
6757 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6758 buggy, causing it to skip the first lines of messages whose message ID
6759 ended in 'D'. This would not have bitten before Exim release 4.14, because
6760 message IDs were unlikely to end in 'D' before then. The effect was to have
6761 incorrect size information for certain domains.
6763 45. #include "config.h" was missing at the start of the crypt16.c module. This
6764 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6767 46. If there was a timeout during a "random" callout check, Exim treated it as
6768 a failure of the random address, and carried on sending RSET and the real
6769 address. If the delay was just some slowness somewhere, the response to the
6770 original RCPT would be taken as a response to RSET and so on, causing
6771 mayhem of various kinds.
6773 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6774 when I implemented it. It didn't allow for the fact that some option values
6775 may legitimately be negative (e.g. size_addition), and it didn't even do
6776 the right test for positive values.
6778 48. Domain names in DNS records are case-independent. Exim always looks them up
6779 in lower case. Some resolvers return domain names in exactly the case they
6780 appear in the zone file, that is, they may contain uppercase letters. Not
6781 all resolvers do this - some return always lower case. Exim was treating a
6782 change of case by a resolver as a change of domain, similar to a widening
6783 of a domain abbreviation. This triggered its re-routing code and so it was
6784 trying to route what was effectively the same domain again. This normally
6785 caused routing to fail (because the router wouldn't handle the domain
6786 twice). Now Exim checks for this case specially, and just changes the
6787 casing of the domain that it ultimately uses when it transmits the message
6790 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6793 50. If a filter generated a file delivery with a non-absolute name (possible if
6794 no home directory exists for the router), the forbid_file option was not
6797 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6798 the existing equality matching.
6800 52. Exim was using ints instead of ino_t variables in some places where it was
6801 dealing with inode numbers.
6803 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6804 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6805 called TMPDIR, and if it finds it is different, it changes its value.
6807 54. The smtp_printf() function is now made available to local_scan() so
6808 additional output lines can be written before returning. There is also an
6809 smtp_fflush() function to enable the detection of a dropped connection.
6810 The variables smtp_input and smtp_batched_input are exported to
6813 55. Changed the default runtime configuration: the message "Unknown user"
6814 has been removed from the ACL, and instead placed on the localuser router,
6815 using the cannot_route_message feature. This means that any verification
6816 failures that generate their own messages won't get overridden. Similarly,
6817 the "Unrouteable address" message that was in the ACL for unverifiable
6818 relay addresses has also been removed.
6820 56. Added hosts_avoid_esmtp to the smtp transport.
6822 57. The exicyclog script was not checking for the esoteric option
6823 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6824 will work only if exicyclog is run under the appropriate euid.
6826 58. Following a discussion on the list, the rules by which Exim recognises line
6827 endings on incoming messages have been changed. The -dropcr and drop_cr
6828 options are now no-ops, retained only for backwards compatibility. The
6829 following line terminators are recognized: LF CRLF CR. However, special
6830 processing applies to CR:
6832 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6833 nor a local message in the state where . is a terminator.
6835 (ii) If a bare CR is encountered in a header line, an extra space is added
6836 after the line terminator so as not to end the header. The reasoning
6837 behind this is that bare CRs in header lines are most likely either
6838 to be mistakes, or people trying to play silly games.
6840 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6841 was being incorrectly given as 18 bytes larger than it should have been.
6842 This is a VOB (very old bug).
6844 60. This may never have affected anything current, but just in case it has:
6845 When the local host is found other than at the start of a list of hosts,
6846 the local host, those with the same MX, and any that follow, are discarded.
6847 When the list in question was part of a longer list of hosts, the following
6848 hosts (not currently being processed) were also being discarded. This no
6849 longer happens. I'm not sure if this situation could ever has previously
6852 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6855 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6856 header. This is something that is recommended in a new Internet Draft, and
6857 is something that is documented as being done by Sendmail. There are two
6858 possible values. For messages generated by the autoreply transport, Exim
6861 Auto-Submitted: auto-replied
6863 whereas for all other generated messages (e.g. bounces) it adds
6865 Auto-Submitted: auto-generated
6867 63. The "personal" condition in filters now includes a test for the
6868 Auto-Submitted: header. If it contains the string "auto-" the message it
6869 not considered personal.
6871 64. Added rcpt_include_affixes as a generic transport option.
6873 65. Added queue_only_override (default true).
6875 66. Added the syslog_duplication option.
6877 67. If what should have been the first header line of a message consisted of
6878 a space followed by a colon, Exim was mis-interpreting it as a header line.
6879 It isn't of course - it is syntactically invalid and should therefore be
6880 treated as the start of the message body. The misbehaviour could have
6881 caused a number of strange effects, including loss of data in subsequent
6882 header lines, and spool format errors.
6884 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6885 client host had authenticated. This control can now be exercised by an ACL
6886 for more flexibility.
6888 69. By default, callouts do not happen when testing with -bh. There is now a
6889 variant, -bhc, which does actually run the callout code, including
6890 consulting and updating the callout cache.
6892 70. Added support for saslauthd authentication, courtesy of Alexander
6895 71. If statvfs() failed on the spool or log directories while checking their
6896 size for availability, Exim confusingly gave the error "space shortage".
6897 Furthermore, in debugging mode it crashed with a floating point exception.
6898 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6899 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6900 serious problem, Exim now writes to the main and panic logs when this
6901 happens, with details of the failure. It then refuses to accept the
6902 incoming message, giving the message "spool directory problem" or "log
6903 directory problem" with a 421 code for SMTP messages.
6905 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6906 0, 1, and 2 exist, because some OS complain for execs without them (see
6907 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6908 descriptors. However, the code omitted to check that the open succeeded,
6909 causing mysterious errors if for some reason the permissions on /dev/null
6910 got screwed. Now Exim writes a message to the main and panic logs, and
6911 bombs out if it can't open /dev/null.
6913 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6914 interact so that it is all more flexible. It is supposed to remain
6915 backwards compatible. Also added extra_local_interfaces.
6917 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6918 to bomb out with an assertion failure - to the client this appears as a
6919 connection drop. This problem occurs in the part of the code that was taken
6920 from the Samba project. Fortunately, the assertion is in a very simple
6921 function, so I have fixed this by reproducing the function inline in the
6922 one place where it is called, and arranging for authentication to fail
6923 instead of killing the process with assert().
6925 75. The SPA client code was not working when the server requested OEM rather
6926 than Unicode encoding.
6928 76. Added code to make require_files with a specific uid setting more usable in
6929 the case where statting the file as root fails - usually a non-root-mounted
6930 NFS file system. When this happens and the failure is EACCES, Exim now
6931 forks a subprocess and does the per-uid checking as the relevant uid.
6933 77. Added process_log_path.
6935 78. If log_file_path was not explicitly set, a setting of check_log_space or
6936 check_log_inodes was ignored.
6938 79. If a space check for the spool or log partitions fails, the incident is now
6939 logged. Of course, in the latter case the data may get lost...
6941 80. Added the %p formatting code to string_format() so that it can be used to
6942 print addresses in debug_print(). Adjusted all the address printing in the
6943 debugging in store.c to use %p rather than %d.
6945 81. There was a concern that a line of code in smtp_in.c could overflow a
6946 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6947 initially expressed, the concern was not well-founded, because trailing
6948 spaces are removed early. However, if the trailing spaces were followed by
6949 a NULL, they did not get removed, so the overflow was possible. Two fixes
6952 (a) I re-wrote the offending code in a cleaner fashion.
6953 (b) If an incoming SMTP command contains a NULL character, it is rejected
6956 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6957 runs initgroups(), so that if the Exim user is in any additional groups,
6958 they will be used during message reception.
6964 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob