1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Rework error string handling in TLS interface so that the caller in
12 more cases is responsible for logging. This permits library-sourced
13 string to be attached to addresses during delivery, and collapses
14 pairs of long lines into single ones.
16 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
17 during configuration. Wildcards are allowed and expanded.
19 JH/02 Rework error string handling in DKIM to pass more info back to callers.
20 This permits better logging.
22 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
23 do not close it down and have the child transport start it up again on
24 the passed-on TCP connection. Instead, proxy the child (and any
25 subsequent ones) for TLS via a unix-domain socket channel. Logging is
26 affected: the continued delivery log lines do not have any DNSSEC, TLS
27 Certificate or OCSP information. TLS cipher information is still logged.
29 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
30 identical IP addresses on different listening ports. Will also affect
33 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
34 add noisy ifdef guards to special-case this sillyness.
35 Patch from Bernd Kuhls.
37 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
38 than 255 are no longer allowed.
40 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
41 Disable the session-cache too, which might reduce our load. Since we
42 currrectly use a new context for every connection, both as server and
43 client, there is no benefit for these.
44 GnuTLS appears to not support tickets server-side by default (we don't
45 call gnutls_session_ticket_enable_server()) but client side is enabled
46 by default on recent versions (3.1.3 +) unless the PFS priority string
49 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
50 <https://reproducible-builds.org/specs/source-date-epoch/>.
52 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
53 the check for any unsuccessful recipients did not notice the limit, and
54 erroneously found still-pending ones.
56 JH/08 Pipeline CHUNKING command and data together, on kernels that support
57 MSG_MORE. Only in-clear (not on TLS connections).
59 JH/09 Avoid using a temporary file during transport using dkim. Unless a
60 transport-filter is involved we can buffer the headers in memory for
61 creating the signature, and read the spool data file once for the
62 signature and again for transmission.
64 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
65 4.77 as the kernel support then wasn't solid, having issues in 64bit
66 mode. Now, it's been long enough. Add support for FreeBSD also.
68 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
69 case where the routing stage had gathered several addresses to send to
70 a host before calling the transport for the first, we previously failed
71 to close down TLS in the old transport process before passing the TCP
72 connection to the new process. The new one sent a STARTTLS command
73 which naturally failed, giving a failed delivery and bloating the retry
74 database. Investigation and fix prototype from Wolfgang Breyha.
76 JH/12 Fix check on SMTP command input synchronisation. Previously there were
77 false-negatives in the check that the sender had not preempted a response
78 or prompt from Exim (running as a server), due to that code's lack of
79 awareness of the SMTP input buffering.
81 PP/04 Add commandline_checks_require_admin option.
82 Exim drops privileges sanely, various checks such as -be aren't a
83 security problem, as long as you trust local users with access to their
84 own account. When invoked by services which pass untrusted data to
85 Exim, this might be an issue. Set this option in main configuration
86 AND make fixes to the calling application, such as using `--` to stop
89 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
90 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
91 responses to those, into a single TLS record each way (this usually means
92 a single packet). As a side issue, smtp_enforce_sync now works on TLS
95 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
96 affects you only if you're dancing at the edge of the param size limits.
97 If you are, and this message makes sense to you, then: raise the
98 configured limit or use OpenSSL 1.1. Nothing we can do for older
101 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
102 cmdline element to get the documented default one. Previously it was
105 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
106 are unsuable for non-message-specific reasons - eg. connection timeout,
107 banner-time rejection.
109 JH/16 Fix logging of delivery remote port, when specified by router, under
112 PP/06 Repair manualroute's ability to take options in any order, even if one
113 is the name of a transport.
116 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
118 JH/17 Change the list-building routines interface to use the expanding-string
119 triplet model, for better allocation and copying behaviour.
121 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
122 Previously it was constructed the first time a possibly-matching string
123 was met in the configuration file input during startup; now it is done
126 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
127 compatible one, to avoid the (poorly documented) possibility of a config
128 file in the working directory redirecting the DB files, possibly correpting
129 some existing file. CVE-2017-10140 assigned for BDB.
131 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
132 cache-hot. Previously, although the result was properly cached, the
133 initial verify call returned a defer.
135 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
136 the main verify for receipient in uncached-mode.
138 JH/22 Retire historical build files to an "unsupported" subdir. These are
139 defined as "ones for which we have no current evidence of testing".
141 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
142 if present. Previously it was ignored.
144 JH/24 Start using specified-initialisers in C structure init coding. This is
145 a C99 feature (it's 2017, so now considered safe).
147 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
148 if was a fixed-sized field and bitmask ops via macros; it is now more
151 PP/07 GitHub PR 56: Apply MariaDB build fix.
152 Patch provided by Jaroslav Škarvada.
154 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
155 during Coverity cleanups [4.87 JH/47]
156 Diagnosis and fix provided by Michael Fischer v. Mollard.
158 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
159 the right size to place the terminating semicolon on its own folded
160 line, the header hash was calculated to an incorrect value thanks to
161 the (relaxed) space the fold became.
163 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
164 and confused the parent.
166 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
167 which could crash as a result. This could lead to undeliverable messages.
173 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
174 than -2003 did; needs libidn2 in addition to libidn.
176 JH/02 The path option on a pipe transport is now expanded before use.
178 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
179 Patch provided by "Björn", documentation fix added too.
181 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
182 missing a wire-to-host endian conversion.
184 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
185 close after a BDAT command line could be taken as a following command,
186 giving a synch failure. Fix by only checking for synch immediately
187 before acknowledging the chunk.
189 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
190 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
191 macro. Patches provided by Josh Soref.
193 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
194 Previously we did not; the RFC seems ambiguous and VRFY is not listed
195 by IANA as a service extension. However, John Klensin suggests that we
198 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
199 the dkim code may be unix-mode line endings rather than smtp wire-format
200 CRLF, so prepend a CR to any bare LF.
202 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
203 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
205 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
206 the first were themselves being wrongly included in the feed into dkim
207 processing; with most chunk sizes in use this resulted in an incorrect
208 body hash calculated value.
210 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
211 DKIM signature block, for verification. Although advised against by
212 standards it is specifically not ruled illegal.
214 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
216 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
217 missing a body hash (the bh= tag).
219 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
220 It seems that HAProxy sends the Proxy Protocol information in clear and
221 only then does a TLS startup, so do the same.
223 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
224 TCP connections (such as for Spamd) unless the daemon successfully set
225 Fast Open mode on its listening sockets. This fixes breakage seen on
226 too-old kernels or those not configured for Fast Open, at the cost of
227 requiring both directions being enabled for TFO, and TFO never being used
228 by non-daemon-related Exim processes.
230 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
231 endings, at least on the first header line. Try to canonify any that get
232 past that check, despite the cost.
234 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
235 now limited to an arbitrary five deep, while parsing addresses with the
236 strip_excess_angle_brackets option enabled.
238 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
239 instead leave the unprompted TLS handshake in socket buffer for the
240 TLS library to consume.
242 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
244 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
246 JH/16 Drop variables when they go out of scope. Memory management drops a whole
247 region in one operation, for speed, and this leaves assigned pointers
248 dangling. Add checks run only under the testsuite which checks all
249 variables at a store-reset and panics on a dangling pointer; add code
250 explicitly nulling out all the variables discovered. Fixes one known
251 bug: a transport crash, where a dangling pointer for $sending_ip_address
252 originally assigned in a verify callout, is re-used.
254 PP/06 Drop '.' from @INC in various Perl scripts.
256 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
258 PP/08 Reduce a number of compilation warnings under clang; building with
259 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
260 should be warning-free.
262 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
264 HS/01 Fix portability problems introduced by PP/08 for platforms where
265 realloc(NULL) is not equivalent to malloc() [SunOS et al].
267 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
268 chunk. This allows us to accept broken chunked messages. We need a more
269 general solution here.
271 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
272 already-broken messages in the queue.
274 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
276 JH/19 Fix reference counting bug in routing-generated-address tracking.
282 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
283 supports it and a size is available (ie. the sending peer gave us one).
285 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
286 years of being deprecated). The replacements are the ACLs
287 acl_smtp_mime and acl_not_smtp_mime.
289 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
290 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
291 or even an in-clear connection were permitted. Now, if the host lookup
292 was dnssec and dane was requested then the host is only used if the
293 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
294 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
295 if one fails this test.
296 This means that a poorly-configured remote DNS will make it incommunicado;
297 but it protects against a DNS-interception attack on it.
299 JH/04 Bug 1810: make continued-use of an open smtp transport connection
300 non-noisy when a race steals the message being considered.
302 JH/05 If main configuration option tls_certificate is unset, generate a
303 self-signed certificate for inbound TLS connections.
305 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
306 in rewrites and routers.
308 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
309 and logged a warning sing 4.83; now they are a configuration file error.
311 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
312 (lacking @domain). Apply the same qualification processing as RCPT.
314 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
316 JH/10 Support ${sha256:} applied to a string (as well as the previous
319 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
320 a cutthrough deliver is pending, as we always want to make a connection.
321 This also avoids re-routing the message when later placing the cutthrough
322 connection after a verify cache hit.
323 Do not update it with the verify result either.
325 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
326 when routing results in more than one destination address.
328 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
329 signing (which inhibits the cutthrough capability). Previously only
330 the presence of an option was tested; now an expansion evaluating as
331 empty is permissible (obviously it should depend only on data available
332 when the cutthrough connection is made).
334 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
335 the relevant preceding SMTP command did not note the pipelining mode.
337 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
338 Previously they were not counted.
340 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
341 as one having no matching records. Previously we deferred the message
342 that needed the lookup.
344 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
345 distinguished as "(=".
347 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
348 for missing MX records. Previously it only worked for missing A records.
350 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
352 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
353 after the data-go-ahead and data-ack. Patch from Jason Betts.
355 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
356 even for a "none" policy. Patch from Tony Meyer.
358 JH/22 Fix continued use of a connection for further deliveries. If a port was
359 specified by a router, it must also match for the delivery to be
362 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
363 When one of the recipients of a message was unsuitable for the connection
364 (has no matching addresses), we lost track of needing to mark it
365 deferred. As a result mail would be lost.
367 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
369 JH/25 Decoding ACL controls is now done using a binary search; the source code
370 takes up less space and should be simpler to maintain. Merge the ACL
371 condition decode tables also, with similar effect.
373 JH/26 Fix problem with one_time used on a redirect router which returned the
374 parent address unchanged. A retry would see the parent address marked as
375 delivered, so not attempt the (identical) child. As a result mail would
378 JH/27 Fix a possible security hole, wherein a process operating with the Exim
379 UID can gain a root shell. Credit to http://www.halfdog.net/ for
380 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
383 JH/28 Enable {spool,log} filesystem space and inode checks as default.
384 Main config options check_{log,spool}_{inodes,space} are now
385 100 inodes, 10MB unless set otherwise in the configuration.
387 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
388 Previously it only applied to the main-section connection policy
391 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
393 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
394 by me. Added RFC7919 DH primes as an alternative.
396 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
397 are not in the system include path.
399 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
400 GnuTLS, when a session startup failed (eg because the client disconnected)
401 Exim did stdio operations after fclose. This was exposed by a recent
402 change which nulled out the file handle after the fclose.
404 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
405 signed directly by the cert-signing cert, rather than an intermediate
406 OCSP-signing cert. This is the model used by LetsEncrypt.
408 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
410 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
411 an incoming connection.
413 HS/02 Bug 1802: Do not half-close the connection after sending a request
416 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
417 fallback to "prime256v1".
419 JH/34 SECURITY: Use proper copy of DATA command in error message.
420 Could leak key material. Remotely exploitable. CVE-2016-9963.
426 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
427 and 3.4.4 - once the server is enabled to respond to an OCSP request
428 it does even when not requested, resulting in a stapling non-aware
429 client dropping the TLS connection.
431 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
432 support variable-length bit vectors. No functional change.
434 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
435 The I= interface field on outgoing lines is now after the H= remote
436 host field, same as incoming lines. There is a separate
437 outgoing_interface log selector which allows you to disable the
440 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
441 If not running log_selector +smtp_connection the mainlog would be held
442 open indefinitely after a "too many connections" event, including to a
443 deleted file after a log rotate. Leave the per net connection logging
444 leaving it open for efficiency as that will be quickly detected by the
445 check on the next write.
447 HS/01 Bug 1671: Fix post transport crash.
448 Processing the wait-<transport> messages could crash the delivery
449 process if the message IDs didn't exist for some reason. When
450 using 'split_spool_directory=yes' the construction of the spool
451 file name failed already, exposing the same netto behaviour.
453 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
454 mime_regex ACL conditions.
456 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
457 to DSN fail messages (bounces): remote IP, remote greeting, remote response
458 to HELO, local diagnostic string.
460 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
461 log line to debug. Even when configured with a tls authenticator many
462 client connections are expected to not authenticate in this way, so
463 an authenticate fail is not an error.
465 HS/02 Add the Exim version string to the process info. This way exiwhat
466 gives some more detail about the running daemon.
468 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
469 matter for fast-change records such as DNSBLs.
471 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
472 retry record, even if constant. There may be multiple transports with
473 different interface settings and the retry behaviour needs to be kept
476 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
478 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
480 JH/10 Bug 840: fix log_defer_output option of pipe transport
482 JH/11 Bug 830: use same host for all RCPTS of a message, even under
483 hosts_randomize. This matters a lot when combined with mua_wrapper.
485 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
486 ${quote_pgsql:<string>} operator.
488 JH/13 Bug 1708: avoid misaligned access in cached lookup.
490 JH/14 Change header file name for freeradius-client. Relevant if compiling
491 with Radius support; from the Gentoo tree and checked under Fedora.
493 JH/15 Bug 1712: Introduce $prdr_requested flag variable
495 JH/16 Bug 1714: Permit an empty string as expansion result for transport
496 option transport_filter, meaning no filtering.
498 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
500 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
501 defaults to "*" (all hosts). The variable is now available when not built
502 with TLS, default unset, mainly to enable keeping the testsuite sane.
503 If a server certificate is not supplied (via tls_certificate) an error is
504 logged, and clients will find TLS connections fail on startup. Presumably
505 they will retry in-clear.
506 Packagers of Exim are strongly encouraged to create a server certificate
507 at installation time.
509 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
510 with the $config_file variable.
512 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
513 in transport context, after the attempt, and per-recipient. The latter type
514 is per host attempted. The event data is the error message, and the errno
515 information encodes the lookup type (A vs. MX) used for the (first) host,
516 and the trailing two digits of the smtp 4xx response.
518 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
519 to write to mainlog (or rejectlog, paniclog) in the window between file
520 creation and permissions/ownership being changed. Particularly affects
521 installations where exicyclog is run as root, rather than exim user;
522 result is that the running daemon panics and dies.
524 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
526 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
527 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
528 "pri" and "weight". Note that the previous implicit priority given by the
529 list order is no longer honoured.
531 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
534 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
535 by defining SUPPORT_SOCKS.
537 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
538 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
539 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
540 variables are renamed to proxy_{local,external}_{address,port}.
542 JH/27 Move Internationalisation support from Experimental to mainline, enabled
543 for a build by defining SUPPORT_I18N
545 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
546 of the query string, and make ${quote_redis:} do that quoting.
548 JH/29 Move Events support from Experimental to mainline, enabled by default
549 and removable for a build by defining DISABLE_EVENT.
551 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
553 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
554 cached by the daemon.
556 JH/32 Move Redis support from Experimental to mainline, enabled for a build
557 by defining LOOKUP_REDIS. The libhiredis library is required.
559 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
560 keys are given for lookup.
562 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
563 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
564 only supported when built with TLS support. The PolarSSL SHA routines
565 are still used when the TLS library is too old for convenient support.
567 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
568 openssl_options), for security. OpenSSL forces this from version 1.1.0
569 server-side so match that on older versions.
571 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
572 allocation for $value could be released as the expansion processing
573 concluded, but leaving the global pointer active for it.
575 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
576 and to use the domains and local_parts ACL conditions.
578 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
579 incorrectly not doubled on cutthrough transmission, hence seen as a
580 body-termination at the receiving system - resulting in truncated mails.
581 Commonly the sender saw a TCP-level error, and retransmitted the message
582 via the normal store-and-forward channel. This could result in duplicates
583 received - but deduplicating mailstores were liable to retain only the
584 initial truncated version.
586 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
588 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
590 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
591 we're in there, support oversigning also; bug 1309.
593 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
595 HS/04 Add support for keep_environment and add_environment options.
597 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
598 either intentional arithmetic overflow during PRNG, or testing config-
601 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
602 delivery resulted in actual delivery. Cancel cutthrough before DATA
605 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
606 rejects a recipient: pass the reject to the originator.
608 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
609 Many were false-positives and ignorable, but it's worth fixing the
612 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
613 for the new environment-manipulation done at startup. Move the routines
614 from being local to tls.c to being global via the os.c file.
616 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
617 an extract embedded as result-arg for a map, the first arg for extract
618 is unavailable so we cannot tell if this is a numbered or keyed
619 extraction. Accept either.
625 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
628 JH/02 The smtp transport option "multi_domain" is now expanded.
630 JH/03 The smtp transport now requests PRDR by default, if the server offers
633 JH/04 Certificate name checking on server certificates, when exim is a client,
634 is now done by default. The transport option tls_verify_cert_hostnames
635 can be used to disable this per-host. The build option
636 EXPERIMENTAL_CERTNAMES is withdrawn.
638 JH/05 The value of the tls_verify_certificates smtp transport and main options
639 default to the word "system" to access the system default CA bundle.
640 For GnuTLS, only version 3.0.20 or later.
642 JH/06 Verification of the server certificate for a TLS connection is now tried
643 (but not required) by default. The verification status is now logged by
644 default, for both outbound TLS and client-certificate supplying inbound
647 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
650 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
651 Status Notification (bounce) messages are now MIME format per RFC 3464.
652 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
653 under the control of the dsn_advertise_hosts option, and routers may
654 have a dsn_lasthop option.
656 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
657 default, modifiable by a malware= option. The list separator for
658 the options can now be changed in the usual way. Bug 68.
660 JH/10 The smtp_receive_timeout main option is now expanded before use.
662 JH/11 The incoming_interface log option now also enables logging of the
663 local interface on delivery outgoing connections.
665 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
666 if the interface and destination host and port all match.
668 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
671 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
672 Patch from Andrew Lewis.
674 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
675 now supports optional time-restrictions, weighting, and priority
676 modifiers per server. Patch originally by <rommer@active.by>.
678 JH/16 The spamd_address main option now supports a mixed list of local
679 and remote servers. Remote servers can be IPv6 addresses, and
680 specify a port-range.
682 JH/17 Bug 68: The spamd_address main option now supports an optional
683 timeout value per server.
685 JH/18 Bug 1581: Router and transport options headers_add/remove can
686 now have the list separator specified.
688 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
691 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
694 JH/21 Support for the A6 type of dns record is withdrawn.
696 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
697 rather than the verbs used.
699 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
700 from 255 to 1024 chars.
702 JH/24 Verification callouts now attempt to use TLS by default.
704 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
705 are generic router options now. The defaults didn't change.
707 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
708 Original patch from Alexander Shikoff, worked over by JH.
710 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
711 files if ZCAT_COMMAND is not executable.
713 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
715 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
717 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
718 Normally benign, it bites when the pair was led to by a CNAME;
719 modern usage is to not canonicalize the domain to a CNAME target
720 (and we were inconsistent anyway for A-only vs AAAA+A).
722 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
724 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
725 when evaluating $sender_host_dnssec.
727 JH/31 Check the HELO verification lookup for DNSSEC, adding new
728 $sender_helo_dnssec variable.
730 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
732 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
734 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
736 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
737 documented as working, but never had. Support all but $spam_report.
739 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
740 added for tls authenticator.
742 HS/03 Add perl_taintmode main config option
748 TL/01 When running the test suite, the README says that variables such as
749 no_msglog_check are global and can be placed anywhere in a specific
750 test's script, however it was observed that placement needed to be near
751 the beginning for it to behave that way. Changed the runtest perl
752 script to read through the entire script once to detect and set these
753 variables, reset to the beginning of the script, and then run through
754 the script parsing/test process like normal.
756 TL/02 The BSD's have an arc4random API. One of the functions to induce
757 adding randomness was arc4random_stir(), but it has been removed in
758 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
759 function when detected.
761 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
762 cause callback expansion.
764 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
765 syntax errors in an expansion can be treated as a string instead of
766 logging or causing an error, due to the internal use of bool_lax
767 instead of bool when processing it.
769 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
770 server certificates when making smtp deliveries.
772 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
774 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
776 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
778 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
779 Merged patch from Sebastian Wiedenroth.
781 JH/05 Fix results-pipe from transport process. Several recipients, combined
782 with certificate use, exposed issues where response data items split
783 over buffer boundaries were not parsed properly. This eventually
784 resulted in duplicates being sent. This issue only became common enough
785 to notice due to the introduction of connection certificate information,
786 the item size being so much larger. Found and fixed by Wolfgang Breyha.
788 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
789 size buffer was used, resulting in syntax errors when an expansion
792 JH/07 Add support for directories of certificates when compiled with a GnuTLS
793 version 3.3.6 or later.
795 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
796 is EXPERIMENTAL_EVENT, the main-configuration and transport options
797 both become "event_action", the variables become $event_name, $event_data
798 and $event_defer_errno. There is a new variable $verify_mode, usable in
799 routers, transports and related events. The tls:cert event is now also
800 raised for inbound connections, if the main configuration event_action
803 TL/06 In test suite, disable OCSP for old versions of openssl which contained
804 early OCSP support, but no stapling (appears to be less than 1.0.0).
806 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
807 server certificate names available under the smtp transport option
808 "tls_verify_cert_hostname" now do not permit multi-component wildcard
811 JH/10 Time-related extraction expansions from certificates now use the main
812 option "timezone" setting for output formatting, and are consistent
813 between OpenSSL and GnuTLS compilations. Bug 1541.
815 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
816 encoded parameter in the incoming message. Bug 1558.
818 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
819 include certificate info, eximon was claiming there were spoolfile
822 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
824 JH/14 Log delivery-related information more consistently, using the sequence
825 "H=<name> [<ip>]" wherever possible.
827 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
828 are problematic for Debian distribution, omit them from the release
831 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
833 JH/16 Fix string representation of time values on 64bit time_t architectures.
836 JH/17 Fix a null-indirection in certextract expansions when a nondefault
837 output list separator was used.
842 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
843 checkers that were complaining about end of non-void function with no
846 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
847 This was a regression introduced in 4.83 by another bugfix.
849 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
851 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
852 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
858 TF/01 Correctly close the server side of TLS when forking for delivery.
860 When a message was received over SMTP with TLS, Exim failed to clear up
861 the incoming connection properly after forking off the child process to
862 deliver the message. In some situations the subsequent outgoing
863 delivery connection happened to have the same fd number as the incoming
864 connection previously had. Exim would try to use TLS and fail, logging
865 a "Bad file descriptor" error.
867 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
868 utilities have not been installed.
870 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
871 temporary space as the ACL may create new global variables.
873 TL/01 LDAP support uses per connection or global context settings, depending
874 upon the detected version of the libraries at build time.
876 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
877 to extract and use the src ip:port in logging and expansions as if it
878 were a direct connection from the outside internet. PPv2 support was
879 updated based on HAProxy spec change in May 2014.
881 JH/02 Add ${listextract {number}{list}{success}{fail}}.
883 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
884 Properly escape header and check for NULL return.
886 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
889 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
891 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
892 characters in header names, implemented as a verify condition.
893 Contributed by Michael Fischer v. Mollard.
895 TL/05 Rename SPF condition results err_perm and err_temp to standardized
896 results permerror and temperror. Previous values are deprecated but
897 still accepted. In a future release, err_perm and err_temp will be
898 completely removed, which will be a backward incompatibility if the
899 ACL tests for either of these two old results. Patch contributed by
900 user bes-internal on the mailing list.
902 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
904 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
905 selectors, in both main and reject logs.
907 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
910 JH/07 Add malware type "sock" for talking to simple daemon.
912 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
914 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
915 routers/transports under cutthrough routing.
917 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
918 numbers. Touch up "bool" conditional to keep the same definition.
920 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
922 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
924 JH/12 Expand items in router/transport headers_add or headers_remove lists
925 individually rather than the list as a whole. Bug 1452.
927 Required for reasonable handling of multiple headers_ options when
928 they may be empty; requires that headers_remove items with embedded
929 colons must have them doubled (or the list-separator changed).
931 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
932 view the policy declared in the DMARC record. Currently, $dmarc_status
933 is a combined value of both the record presence and the result of the
936 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
938 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
939 dnslookup router and the smtp transport (applying to the forward
942 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
943 of ldap servers used for a specific lookup. Patch provided by Heiko
946 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
947 New variable $lookup_dnssec_authenticated for observability.
949 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
950 Patch submitted by Lars Timman.
952 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
954 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
955 Requires trusted mode and valid format message id, aborts otherwise.
956 Patch contributed by Heiko Schlichting.
958 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
959 certextract with support for various fields. Bug 1358.
961 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
962 is requested by default, modifiable by smtp transport option
965 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
966 operate on certificate variables to give certificate fingerprints
967 Also new ${sha256:cert_variable}.
969 JH/23 The PRDR feature is moved from being Experimental into the mainline.
971 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
972 Christian Aistleitner.
974 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
976 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
977 file. Patch from Wolfgang Breyha.
979 JH/25 Expand the coverage of the delivery $host and $host_address to
980 client authenticators run in verify callout. Bug 1476.
982 JH/26 Port service names are now accepted for tls_on_connect_ports, to
983 align with daemon_smtp_ports. Bug 72.
985 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
986 support and error reporting did not work properly.
988 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
989 and is readable. Patch from Andrew Colin Kissa.
991 TL/14 Enhance documentation of ${run expansion and how it parses the
992 commandline after expansion, particularly in the case when an
993 unquoted variable expansion results in an empty value.
995 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
997 PP/02 Fix internal collision of T_APL on systems which support RFC3123
998 by renaming away from it. Addresses GH issue 15, reported by
1001 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1003 TL/15 SECURITY: prevent double expansion in math comparison functions
1004 (can expand unsanitized data). Not remotely exploitable.
1011 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1013 PP/02 Make -n do something, by making it not do something.
1014 When combined with -bP, the name of an option is not output.
1016 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1019 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1020 $sender_host_name and config options to manage this, and basic check
1023 PP/05 DSCP support for outbound connections and control modifier for inbound.
1025 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1026 (Only plugin which currently uses this is kerberos4, which nobody should
1027 be using, but we should make it available and other future plugins might
1028 conceivably use it, even though it would break NAT; stuff *should* be
1029 using channel bindings instead).
1031 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1032 name; added for Sendmail compatibility; requires admin caller.
1033 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1034 just ignore it); requires trusted caller.
1035 Also parse but ignore: -Ac -Am -X<logfile>
1038 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1040 TL/02 Add +smtp_confirmation as a default logging option.
1042 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1043 Patch by Magnus Holmgren from 2007-02-20.
1045 TL/04 Bugzilla 1281 - Spec typo.
1046 Bugzilla 1283 - Spec typo.
1047 Bugzilla 1290 - Spec grammar fixes.
1049 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1051 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1053 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1054 mailing list by by Dmitry Isaikin.
1056 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1058 JH/02 Support "G" suffix to numbers in ${if comparisons.
1060 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1062 NM/01 Bugzilla 1197 - Spec typo
1063 Bugzilla 1196 - Spec examples corrections
1065 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1067 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1068 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1071 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1072 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1074 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1075 "acl {{name}{arg}...}", and optional args on acl condition
1078 JH/05 Permit multiple router/transport headers_add/remove lines.
1080 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1082 JH/07 Avoid using a waiting database for a single-message-only transport.
1083 Performance patch from Paul Fisher. Bugzilla 1262.
1085 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1088 JH/09 Add $headers_added variable, with content from use of ACL modifier
1089 add_header (but not yet added to the message). Bugzilla 199.
1091 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1092 Pulled from Bugzilla 817 by Wolfgang Breyha.
1094 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1096 (nb: this is the same fix as in Exim 4.80.1)
1098 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1101 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1104 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1106 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1107 advertise SMTP AUTH mechanism to us, instead of a generic
1108 protocol violation error. Also, make Exim more robust to bad
1109 data from the Dovecot auth socket.
1111 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1113 When a queue runner is handling a message, Exim first routes the
1114 recipient addresses, during which it prunes them based on the retry
1115 hints database. After that it attempts to deliver the message to
1116 any remaining recipients. It then updates the hints database using
1119 So if a recipient address works intermittently, it can get repeatedly
1120 deferred at routing time. The retry hints record remains fresh so the
1121 address never reaches the final cutoff time.
1123 This is a fairly common occurrence when a user is bumping up against
1124 their storage quota. Exim had some logic in its local delivery code
1125 to deal with this. However it did not apply to per-recipient defers
1126 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1128 This change adds a proper retry rule check during routing so that the
1129 final cutoff time is checked against the message's age. We only do
1130 this check if there is an address retry record and there is not a
1131 domain retry record; this implies that previous attempts to handle
1132 the address had the retry_use_local_parts option turned on. We use
1133 this as an approximation for the destination being like a local
1134 delivery, as in LMTP.
1136 I suspect this new check makes the old local delivery cutoff check
1137 redundant, but I have not verified this so I left the code in place.
1139 TF/02 Correct gecos expansion when From: is a prefix of the username.
1141 Test 0254 submits a message to Exim with the header
1145 When I ran the test suite under the user fanf2, Exim expanded
1146 the header to contain my full name, whereas it should have added
1147 a Resent-Sender: header. It erroneously treats any prefix of the
1148 username as equal to the username.
1150 This change corrects that bug.
1152 GF/01 DCC debug and logging tidyup
1153 Error conditions log to paniclog rather than rejectlog.
1154 Debug lines prefixed by "DCC: " to remove any ambiguity.
1156 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1158 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1159 Bug spotted by Jeremy Harris; was flawed since initial commit.
1160 Would have resulted in OCSP responses post-SNI triggering an Exim
1161 NULL dereference and crash.
1163 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1165 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1166 Bug detection, analysis and fix by Samuel Thibault.
1167 Bugzilla 1331, Debian bug #698092.
1169 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1171 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1172 Server implementation by Todd Lyons, client by JH.
1173 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1174 config variable "prdr_enable" controls whether the server
1175 advertises the facility. If the client requests PRDR a new
1176 acl_data_smtp_prdr ACL is called once for each recipient, after
1177 the body content is received and before the acl_smtp_data ACL.
1178 The client is controlled by both of: a hosts_try_prdr option
1179 on the smtp transport, and the server advertisement.
1180 Default client logging of deliveries and rejections involving
1181 PRDR are flagged with the string "PRDR".
1183 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1184 fclose(). Diagnosis by Todd Lyons.
1186 PP/17 Update configure.default to handle IPv6 localhost better.
1187 Patch by Alain Williams (plus minor tweaks).
1190 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1191 This is now consistent with GnuTLS, and is now documented: the
1192 previous undocumented portable approach to treating the option as
1193 unset was to force an expansion failure. That still works, and
1194 an empty string is now equivalent.
1196 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1197 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1198 not performing validation itself.
1200 PP/20 Added force_command boolean option to pipe transport.
1201 Patch from Nick Koston, of cPanel Inc.
1203 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1206 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1208 PP/21 Fix eximon continuous updating with timestamped log-files.
1209 Broken in a format-string cleanup in 4.80, missed when I repaired the
1210 other false fix of the same issue.
1211 Report and fix from Heiko Schlichting.
1214 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1215 Report from Prashanth Katuri.
1217 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1218 It's SecureTransport, so affects any MacOS clients which use the
1219 system-integrated TLS libraries, including email clients.
1221 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1222 using a MIME ACL for non-SMTP local injection.
1223 Report and assistance in diagnosis by Warren Baker.
1225 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1227 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1229 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1230 last id that failed so it may be referenced in subsequent ACL's.
1232 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1235 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1236 ldap library initialization, allowing self-signed CA's to be
1237 used. Also properly sets require_cert option later in code by
1238 using NULL (global ldap config) instead of ldap handle (per
1239 session). Bug diagnosis and testing by alxgomz.
1241 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1242 the src/util/ subdirectory.
1244 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1245 renamed to Transport Post Delivery Action by Jeremy Harris, as
1248 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1249 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1250 redis_servers = needs to be configured which will be used by the redis
1251 lookup. Patch from Warren Baker, of The Packet Hub.
1253 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1255 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1256 hostname or reverse DNS when processing a host list. Used suggestions
1257 from multiple comments on this bug.
1259 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1261 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1262 Missed a few lines, added it to make the runtest require no keyboard
1265 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1266 contains upper case chars. Make router use caseful_local_part.
1268 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1269 support when GnuTLS has been built with p11-kit.
1275 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1277 This, or similar/improved, will also be change PP/11 of 4.82.
1283 PP/01 Handle short writes when writing local log-files.
1284 In practice, only affects FreeBSD (8 onwards).
1285 Bugzilla 1053, with thanks to Dmitry Isaikin.
1287 NM/01 Bugzilla 949 - Documentation tweak
1289 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1292 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1294 PP/02 Implemented gsasl authenticator.
1296 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1298 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1299 `pkg-config foo` for cflags/libs.
1301 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1302 with rest of GSASL and with heimdal_gssapi.
1304 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1305 `pkg-config foo` for cflags/libs for the TLS implementation.
1307 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1308 properties get this fed in as external SSF. A number of robustness
1309 and debugging improvements to the cyrus_sasl authenticator.
1311 PP/08 cyrus_sasl server now expands the server_realm option.
1313 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1314 Patch by Jeremy Harris.
1316 PP/10 Added dbmjz lookup type.
1318 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1320 PP/12 MAIL args handles TAB as well as SP, for better interop with
1321 non-compliant senders.
1322 Analysis and variant patch by Todd Lyons.
1324 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1325 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1326 Patch from Dirk Mueller <dmueller@suse.com>
1328 PP/13 tls_peerdn now print-escaped for spool files.
1329 Observed some $tls_peerdn in wild which contained \n, which resulted
1330 in spool file corruption.
1332 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1333 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1334 or write after TLS renegotiation, which otherwise led to messages
1337 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1338 as a tracking header (ie: a signed header comes before the signature).
1339 Patch from Wolfgang Breyha.
1341 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1342 comma-sep list; embedded commas doubled.
1344 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1346 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1348 Report and patch from Dmitry Banschikov.
1350 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1351 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1352 failures. We appear to now support TLS1.1+ with Exim.
1354 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1355 lets Exim select keys and certificates based upon TLS SNI from client.
1356 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1357 before an outbound SMTP session. New log_selector, +tls_sni.
1359 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1360 NULL dereference. Report and patch from Alun Jones.
1362 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1363 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1364 Not seeing resolver debug output on NetBSD, but suspect this is a
1365 resolver implementation change.
1367 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1368 Left warnings. Added "eximon gdb" invocation mode.
1370 PP/21 Defaulting "accept_8bitmime" to true, not false.
1372 PP/22 Added -bw for inetd wait mode support.
1374 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1375 locate the relevant includes and libraries. Made this the default.
1377 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1378 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1380 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1381 This may cause build issues on older platforms.
1383 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1384 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1385 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1386 Added SNI support via GnuTLS too.
1387 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1389 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1391 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1392 Applied second patch from Janne, implementing suggestion to default
1393 multiple-strings-in-record handling to match SPF spec.
1395 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1397 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1398 read-only, out of scope).
1399 Patch from Wolfgang Breyha, report from Stuart Northfield.
1401 PP/29 Fix three issues highlighted by clang analyser static analysis.
1402 Only crash-plausible issue would require the Cambridge-specific
1403 iplookup router and a misconfiguration.
1404 Report from Marcin Mirosław.
1406 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1408 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1409 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1410 As part of this, removing so much warning spew let me fix some minor
1411 real issues in debug logging.
1413 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1414 assignment on my part. Fixed.
1416 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1417 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1418 Janne Snabb (who went above and beyond: thank you).
1420 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1421 string otherwise requires a connection and a bunch more work and it's
1422 relatively easy to get wrong. Should also expose TLS library linkage
1425 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1426 64-bit ${eval} (JH/03).
1428 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1429 GNU libc to support some of the 64-bit stuff, should not lead to
1430 conflicts. Defined before os.h is pulled in, so if a given platform
1431 needs to override this, it can.
1433 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1434 protection layer was required, which is not implemented.
1435 Bugzilla 1254, patch from Wolfgang Breyha.
1437 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1438 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1439 tls_dhparam take prime identifiers. Also unbreak combination of
1440 OpenSSL+DH_params+TLSSNI.
1442 PP/39 Disable SSLv2 by default in OpenSSL support.
1448 PP/01 Solaris build fix for Oracle's LDAP libraries.
1449 Bugzilla 1109, patch from Stephen Usher.
1451 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1453 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1456 TF/02 Fix a couple more cases where we did not log the error message
1457 when unlink() failed. See also change 4.74-TF/03.
1459 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1460 lock up or crash if it happened to be inside a call to libc when it
1461 got a SIGUSR1 from exiwhat.
1463 The SIGUSR1 handler appends the current process status to the process
1464 log which is later printed by exiwhat. It used to use the general
1465 purpose logging code to do this, but several functions it calls are
1466 not safe for signals.
1468 The new output code in the SIGUSR1 handler is specific to the process
1469 log, and simple enough that it's easy to inspect for signal safety.
1470 Removing some special cases also simplifies the general logging code.
1471 Removing the spurious timestamps from the process log simplifies
1474 TF/04 Improved ratelimit ACL condition.
1476 The /noupdate option has been deprecated in favour of /readonly which
1477 has clearer semantics. The /leaky, /strict, and /readonly update modes
1478 are mutually exclusive. The update mode is no longer included in the
1479 database key; it just determines when the database is updated. (This
1480 means that when you upgrade Exim will forget old rate measurements.)
1482 Exim now checks that the per_* options are used with an update mode that
1483 makes sense for the current ACL. For example, when Exim is processing a
1484 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1485 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1486 must specify per_mail/readonly. If you omit the update mode it defaults to
1487 /leaky where that makes sense (as before) or /readonly where required.
1489 The /noupdate option is now undocumented but still supported for
1490 backwards compatibility. It is equivalent to /readonly except that in
1491 ACLs where /readonly is required you may specify /leaky/noupdate or
1492 /strict/noupdate which are treated the same as /readonly.
1494 A useful new feature is the /count= option. This is a generalization
1495 of the per_byte option, so that you can measure the throughput of other
1496 aggregate values. For example, the per_byte option is now equivalent
1497 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1499 The per_rcpt option has been generalized using the /count= mechanism
1500 (though it's more complicated than the per_byte equivalence). When it is
1501 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1502 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1503 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1504 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1505 Note that using per_rcpt with a non-readonly update mode in more than
1506 one ACL will cause the recipients to be double-counted. (The per_mail
1507 and per_byte options don't have this problem.)
1509 The handling of very low rates has changed slightly. If the computed rate
1510 is less than the event's count (usually one) then this event is the first
1511 after a long gap. In this case the rate is set to the same as this event's
1512 count, so that the first message of a spam run is counted properly.
1514 The major new feature is a mechanism for counting the rate of unique
1515 events. The new per_addr option counts the number of different
1516 recipients that someone has sent messages to in the last time period. It
1517 behaves like per_rcpt if all the recipient addresses are different, but
1518 duplicate recipient addresses do not increase the measured rate. Like
1519 the /count= option this is a general mechanism, so the per_addr option
1520 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1521 example, measure the rate that a client uses different sender addresses
1522 with the options per_mail/unique=$sender_address. There are further
1523 details in the main documentation.
1525 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1527 TF/06 Removed a few PCRE remnants.
1529 TF/07 Automatically extract Exim's version number from tags in the git
1530 repository when doing development or release builds.
1532 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1533 Bugzilla 879. Patch from Paul Fisher.
1535 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1536 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1539 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1541 PP/05 Variable $av_failed, true if the AV scanner deferred.
1542 Bugzilla 1078. Patch from John Horne.
1544 PP/06 Stop make process more reliably on build failure.
1545 Bugzilla 1087. Patch from Heiko Schlittermann.
1547 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1548 Bugzilla 1089. Patch from Heiko Schlittermann.
1550 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1551 Bugzilla 1131. Patch from Holger Weiß.
1553 PP/09 Handle IPv6 addresses with SPF.
1554 Bugzilla 860. Patch from Wolfgang Breyha.
1556 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1558 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1561 PP/11 match_* no longer expand right-hand-side by default.
1562 New compile-time build option, EXPAND_LISTMATCH_RHS.
1563 New expansion conditions, "inlist", "inlisti".
1565 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1567 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1569 PP/14 fix log_write() format string regression from TF/03.
1570 Bugzilla 1152. Patch from Dmitry Isaikin.
1576 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1578 PP/02 Harmonised TLS library version reporting; only show if debugging.
1579 Layout now matches that introduced for other libraries in 4.74 PP/03.
1581 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1583 PP/04 New "dns_use_edns0" global option.
1585 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1588 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1589 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1591 TK/01 Updated PolarSSL code to 0.14.2.
1592 Bugzilla 1097. Patch from Andreas Metzler.
1594 PP/07 Catch divide-by-zero in ${eval:...}.
1595 Fixes bugzilla 1102.
1597 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1600 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1601 format-string attack -- SECURITY: remote arbitrary code execution.
1603 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1604 time unintentionally subject to list matching rules, letting the header
1605 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1606 arbitrary string expansion). This allowed for information disclosure.
1608 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1609 INT_MIN/-1 -- value coerced to INT_MAX.
1615 NM/01 Workaround for PCRE version dependency in version reporting
1618 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1619 This fixes portability to compilers other than gcc, notably
1620 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1622 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1623 makefiles for portability to HP-UX and POSIX correctness.
1625 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1626 Also via indented variable definition in the Makefile.
1627 (Debugging by Oliver Heesakkers).
1629 PP/02 Restore caching of spamd results with expanded spamd_address.
1630 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1632 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1633 Improves build reliability. Fix from: Frank Elsner
1635 NM/02 Fix wide character breakage in the rfc2047 coding
1636 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1638 NM/03 Allow underscore in dnslist lookups
1639 Fixes bug 1026. Patch from Graeme Fowler
1641 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1642 Code patches from Adam Ciarcinski of NetBSD.
1644 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1647 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1648 is logged, to avoid truncation. Patch from John Horne.
1650 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1651 Patch from Jakob Hirsch.
1653 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1654 SQL string expansion failure details.
1655 Patch from Andrey Oktyabrski.
1657 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1658 Patch from Simon Arlott.
1660 PP/09 New lookups functionality failed to compile on old gcc which rejects
1661 extern declarations in function scope.
1662 Patch from Oliver Fleischmann
1664 PP/10 Use sig_atomic_t for flags set from signal handlers.
1665 Check getgroups() return and improve debugging.
1666 Fixed developed for diagnosis in bug 927 (which turned out to be
1669 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1670 Patch from Mark Zealey.
1672 PP/12 Bugzilla 1056: Improved spamd server selection.
1673 Patch from Mark Zealey.
1675 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1676 Based on patch from Heiko Schlittermann.
1678 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1679 Patch from Uwe Doering, sign-off by Michael Haardt.
1681 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1682 variable declaration deep within a block. Bug and patch from
1685 PP/15 lookups-Makefile IRIX compatibility coercion.
1687 PP/16 Make DISABLE_DKIM build knob functional.
1689 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1690 Patch by Simon Arlott
1692 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1693 variable argument macros. Our copy now differs from upstream.
1699 TF/01 Failure to get a lock on a hints database can have serious
1700 consequences so log it to the panic log.
1702 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1703 controlled using the smtp_confirmation log selector.
1705 TF/03 Include the error message when we fail to unlink a spool file.
1707 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1708 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1709 for maintaining out-of-tree patches for some time.
1711 PP/01 Bugzilla 139: Documentation and portability issues.
1712 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1713 Handle per-OS dynamic-module compilation flags.
1715 PP/02 Let /dev/null have normal permissions.
1716 The 4.73 fixes were a little too stringent and complained about the
1717 permissions on /dev/null. Exempt it from some checks.
1718 Reported by Andreas M. Kirchwitz.
1720 PP/03 Report version information for many libraries, including
1721 Exim version information for dynamically loaded libraries. Created
1722 version.h, now support a version extension string for distributors
1723 who patch heavily. Dynamic module ABI change.
1725 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1726 privilege escalation vulnerability whereby the Exim run-time user
1727 can cause root to append content of the attacker's choosing to
1730 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1733 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1734 If dropping privileges for untrusted macros, we disabled normal logging
1735 on the basis that it would fail; for the Exim run-time user, this is not
1736 the case, and it resulted in successful deliveries going unlogged.
1737 Fixed. Reported by Andreas Metzler.
1743 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1744 only prepend for the Resent-* case. Fixes regression introduced in
1745 Exim 4.70 by NM/22 for Bugzilla 607.
1747 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1748 increasing numbers of administrators be bitten by this.
1750 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1752 PP/03 Bugzilla 994: added openssl_options main configuration option.
1754 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1756 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1758 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1760 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1761 without a peer certificate, leading to a segfault because of an
1762 assumption that peers always have certificates. Be a little more
1763 paranoid. Problem reported by Martin Tscholak.
1765 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1766 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1767 NB: ClamAV planning to remove STREAM in "middle of 2010".
1768 CL also introduces -bmalware, various -d+acl logging additions and
1769 more caution in buffer sizes.
1771 PP/09 Implemented reverse_ip expansion operator.
1773 PP/10 Bugzilla 937: provide a "debug" ACL control.
1775 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1777 PP/12 Bugzilla 973: Implement --version.
1779 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1781 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1783 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1785 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1786 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1787 ignore trailing whitespace.
1789 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1791 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1794 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1795 Notification from Dr Andrew Aitchison.
1797 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1798 ExtendedDetectionInfo response format.
1799 Notification from John Horne.
1801 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1804 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1805 XSL and documented dependency on system catalogs, with examples of how
1808 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1811 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1812 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1813 configuration file which is writeable by the Exim user or group.
1815 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1816 of configuration files to cover files specified with the -C option if
1817 they are going to be used with root privileges, not just the default
1820 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1821 option (effectively making it always true).
1823 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1824 files to be used while preserving root privileges.
1826 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1827 that rogue child processes cannot use them.
1829 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1830 run-time user, instead of root.
1832 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1833 Exim run-time user without dropping privileges.
1835 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1836 result string, instead of calling string_vformat() twice with the same
1839 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1840 for other users. Others should always drop root privileges if they use
1841 -C on the command line, even for a whitelisted configure file.
1843 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1845 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1851 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1852 $data_path, and $header_path variables; fixed documentation bugs and
1855 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1856 exipick to access non-standard spools, including the "frozen" queue
1859 NM/01 Bugzilla 965: Support mysql stored procedures.
1860 Patch from Alain Williams
1862 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1864 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1865 Patch from Andreas Metzler
1867 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1868 Patch from Kirill Miazine
1870 NM/05 Bugzilla 671: Added umask to procmail example.
1872 JJ/03 installed exipick 20100323.0, fixing doc bug
1874 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1875 directory. Notification and patch from Dan Rosenberg.
1877 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1879 TK/02 Improve log output when DKIM signing operation fails.
1881 MH/01 Treat the transport option dkim_domain as a colon separated
1882 list, not as a single string, and sign the message with each element,
1883 omitting multiple occurences of the same signer.
1885 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1886 Bugzilla 985, 986. Patch by Simon Arlott
1888 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1889 Patch by Simon Arlott
1891 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1892 MBX locking. Notification from Dan Rosenberg.
1898 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1900 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1902 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1904 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1906 NM/04 Bugzilla 915: Flag broken perl installation during build.
1912 TK/01 Added patch by Johannes Berg that expands the main option
1913 "spamd_address" if it starts with a dollar sign.
1915 TK/02 Write list of recipients to X-Envelope-Sender header when building
1916 the mbox-format spool file for content scanning (suggested by Jakob
1919 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1920 (http://www.dcc-servers.net/) support via dccifd. Activated by
1921 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1923 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1924 by Mark Daniel Reidel <mr@df.eu>.
1926 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1927 When building exim an external PCRE library is now needed -
1928 PCRE is a system library on the majority of modern systems.
1929 See entry on PCRE_LIBS in EDITME file.
1931 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1932 conversation. Added nologin parameter to request.
1933 Patch contributed by Kirill Miazine.
1935 TF/01 Do not log submission mode rewrites if they do not change the address.
1937 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1939 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1940 log files in place. Contributed by Roberto Lima.
1942 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1944 TF/03 Bugzilla 615: When checking the local_parts router precondition
1945 after a local_part_suffix or local_part_prefix option, Exim now
1946 does not use the address's named list lookup cache, since this
1947 contains cached lookups for the whole local part.
1949 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1950 Robert Millan. Documentation is in experimental-spec.txt.
1952 TF/04 Bugzilla 668: Fix parallel build (make -j).
1954 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1956 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1957 Patch provided by Jan Srzednicki.
1959 TF/05 Leading white space used to be stripped from $spam_report which
1960 wrecked the formatting. Now it is preserved.
1962 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1963 that they are available at delivery time.
1965 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1967 TF/08 TLS error reporting now respects the incoming_interface and
1968 incoming_port log selectors.
1970 TF/09 Produce a more useful error message if an SMTP transport's hosts
1971 setting expands to an empty string.
1973 NM/06 Bugzilla 744: EXPN did not work under TLS.
1974 Patch provided by Phil Pennock.
1976 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1977 Patch provided by Richard Godbee.
1979 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1980 acl_smtp_notquit, added index entry.
1982 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1983 Patch provided by Eugene Bujak.
1985 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1986 accept(). Patch provided by Maxim Dounin.
1988 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1989 Patch provided by Phil Pennock.
1991 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1993 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1994 Patch provided by Brad "anomie" Jorsch.
1996 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1997 Patch provided by Dean Brooks.
1999 TK/05 Add native DKIM support (does not depend on external libraries).
2001 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2002 Patch provided by Graeme Fowler.
2004 NM/16 Bugzilla 851: Documentation example syntax fix.
2006 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2008 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2011 NM/19 Bugzilla 745: TLS version reporting.
2012 Patch provided by Phil Pennock.
2014 NM/20 Bugzilla 167: bool: condition support.
2015 Patch provided by Phil Pennock.
2017 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2018 clients. Patch provided by Phil Pennock.
2020 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2021 Patch provided by Brad "anomie" Jorsch.
2023 NM/23 Bugzilla 687: Fix misparses in eximstats.
2024 Patch provided by Heiko Schlittermann.
2026 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2027 Patch provided by Heiko Schlittermann.
2029 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2030 plus update to original patch.
2032 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2034 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2035 Patch provided by David Brownlee.
2037 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2039 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2041 NM/30 Bugzilla 888: TLS documentation bugfixes.
2043 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2045 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2046 Unlike the original bugzilla I have changed all shell scripts in src tree.
2048 NM/33 Bugzilla 898: Transport filter timeout fix.
2049 Patch by Todd Rinaldo.
2051 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2052 Patch by Serge Demonchaux.
2054 NM/35 Bugzilla 39: Base64 decode bug fixes.
2055 Patch by Jakob Hirsch.
2057 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2059 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2061 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2063 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2069 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2070 ALT-N's libdkim that I have put here:
2071 http://duncanthrax.net/exim-experimental/
2073 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2074 were called 'true' and it seems that C99 defines that as a reserved
2075 keyword to be used with 'bool' variable types. That means you could
2076 not include C99-style headers which use bools without triggering
2077 build errors in sieve.c.
2079 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2080 as mailq or other aliases. Changed the --help handling significantly
2081 to do whats expected. exim_usage() emits usage/help information.
2083 SC/01 Added the -bylocaldomain option to eximstats.
2085 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2087 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2089 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2095 PH/01 Another patch from the Sieve maintainer.
2097 PH/02 When an IPv6 address is converted to a string for single-key lookup
2098 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2099 dots are used instead of colons so that keys in lsearch files need not
2100 contain colons. This was done some time before quoting was made available
2101 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2102 (notated using the quote facility) so as to distinguish them from IPv4
2103 keys. This meant that lookups for IP addresses in host lists did not work
2104 for iplsearch lookups.
2106 This has been fixed by arranging for IPv6 addresses to be expressed with
2107 colons if the lookup type is iplsearch. This is not incompatible, because
2108 previously such lookups could never work.
2110 The situation is now rather anomalous, since one *can* have colons in
2111 ordinary lsearch keys. However, making the change in all cases is
2112 incompatible and would probably break a number of configurations.
2114 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2117 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2118 conversion specification without a maximum field width, thereby enabling
2119 a rogue spamd server to cause a buffer overflow. While nobody in their
2120 right mind would setup Exim to query an untrusted spamd server, an
2121 attacker that gains access to a server running spamd could potentially
2122 exploit this vulnerability to run arbitrary code as the Exim user.
2124 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2125 $primary_hostname instead of what libspf2 thinks the hosts name is.
2127 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2128 a directory entry by the name of the lookup key. Previously, if a
2129 symlink pointed to a non-existing file or a file in a directory that
2130 Exim lacked permissions to read, a lookup for a key matching that
2131 symlink would fail. Now it is enough that a matching directory entry
2132 exists, symlink or not. (Bugzilla 503.)
2134 PH/03 The body_linecount and body_zerocount variables are now exported in the
2137 PH/04 Added the $dnslist_matched variable.
2139 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2140 This means they are set thereafter only if the connection becomes
2143 PH/06 Added the client_condition to authenticators so that some can be skipped
2144 by clients under certain conditions.
2146 PH/07 The error message for a badly-placed control=no_multiline_responses left
2147 "_responses" off the end of the name.
2149 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2151 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2152 (without spaces) instead of just copying the configuration text.
2154 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2156 PH/11 Added $max_received_linelength.
2158 PH/12 Added +ignore_defer and +include_defer to host lists.
2160 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2161 way in which PCRE > 7.0 is built.
2163 PH/14 Implemented queue_only_load_latch.
2165 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2166 MAIL command. The effect was to mangle the value on 64-bit systems.
2168 PH/16 Another patch from the Sieve maintainer.
2170 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2172 PH/18 If a system quota error occurred while trying to create the file for
2173 a maildir delivery, the message "Mailbox is full" was not appended to the
2174 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2175 only to a quota excession during the actual writing of the file.
2177 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2178 characters?) which causes problems in log lines. The DN values are now
2179 passed through string_printing() before being added to log lines.
2181 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2182 and InterBase are left for another time.)
2184 PH/21 Added message_body_newlines option.
2186 PH/22 Guard against possible overflow in moan_check_errorcopy().
2188 PH/23 POSIX allows open() to be a macro; guard against that.
2190 PH/24 If the recipient of an error message contained an @ in the local part
2191 (suitably quoted, of course), incorrect values were put in $domain and
2192 $local_part during the evaluation of errors_copy.
2198 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2199 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2202 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2203 issue a MAIL command.
2205 PH/02 In an ACL statement such as
2207 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2209 if a client was not listed at all, or was listed with a value other than
2210 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2211 the condition was not true (as it should be), so access was not denied.
2212 The bug was that the ! inversion was incorrectly passed on to the second
2213 item. This has been fixed.
2215 PH/03 Added additional dnslists conditions == and =& which are different from
2216 = and & when the dns lookup returns more than one IP address.
2218 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2219 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2221 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2222 FSYNC, which compiles an option called disable_fsync that allows for
2223 bypassing fsync(). The documentation is heavily laced with warnings.
2225 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2227 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2228 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2229 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2230 including adding "make clean"; (3) Added -fPIC when compiling the test
2231 dynamically loaded module, to get rid of a warning.
2233 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2234 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2235 = 0s. The bug is otherwise harmless.
2237 PH/07 There was a bug in the dovecot authenticator such that the value of
2238 $auth1 could be overwritten, and so not correctly preserved, after a
2239 successful authentication. This usually meant that the value preserved by
2240 the server_setid option was incorrect.
2242 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2244 PH/09 Installed PCRE release 7.0.
2246 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2247 run for batched SMTP input. It is now run at the start of every message
2248 in the batch. While fixing this I discovered that the process information
2249 (output by running exiwhat) was not always getting set for -bs and -bS
2250 input. This is fixed, and it now also says "batched" for BSMTP.
2252 PH/11 Added control=no_pipelining.
2254 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2255 patch, slightly modified), and move the expansion of helo_data till after
2256 the connection is made in the smtp transport (so it can use these
2259 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2261 PH/14 Added log_selector = +pid.
2263 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2265 PH/16 Add ${if forany and ${if forall.
2267 PH/17 Added dsn_from option to vary the From: line in DSNs.
2269 PH/18 Flush SMTP output before performing a callout, unless control =
2270 no_callout_flush is set.
2272 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2273 was true (the default) a successful delivery failed to delete the retry
2274 item, thus causing premature timeout of the address. The bug is now
2277 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2279 PH/21 Long custom messages for fakedefer and fakereject are now split up
2280 into multiline responses in the same way that messages for "deny" and
2281 other ACL rejections are.
2283 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2284 with slight modification.
2286 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2287 draft, changing the syntax and factoring some duplicate code".
2289 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2290 for deliveries of the second and subsequent messages over the same SMTP
2293 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2294 ${reduce, with only minor "tidies".
2296 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2298 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2299 expansion side effects.
2301 PH/27 When a message times out after an over-quota error from an Exim-imposed
2302 quota, the bounce message says "mailbox is full". This message was not
2303 being given when it was a system quota that was exceeded. It now should
2306 MH/03 Made $recipients available in local_scan(). local_scan() already has
2307 better access to the recipient list through recipients_list[], but
2308 $recipients can be useful in postmaster-provided expansion strings.
2310 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2311 in the case of a MAIL command with additional options following the
2312 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2313 were accidentally chopped off.
2315 PH/29 SMTP synchronization checks are implemented when a command is read -
2316 there is a check that no more input is waiting when there shouldn't be
2317 any. However, for some commands, a delay in an ACL can mean that it is
2318 some time before the response is written. In this time, more input might
2319 arrive, invalidly. So now there are extra checks after an ACL has run for
2320 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2321 pipelining has not been advertised.
2323 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2325 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2326 returned in the SMTP response when smtp_return_error_details was false.
2327 This has been fixed.
2329 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2330 instead of the C I/O that was originally supplied, because problems were
2331 reported on Solaris.
2333 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2334 Exim which did not show up earlier: it was assuming that a call to
2335 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2336 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2337 was a macro that became an assignment, so it seemed to work. This has
2338 changed to a proper function call with a void return, hence the compile
2339 error. Exim's code has been fixed.
2341 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2344 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2346 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2348 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2349 because it turns out to be too big or there is a timeout) there is no
2350 "Completed" line in the log. When some messages of this type were
2351 selected by exigrep, they were listed as "not completed". Others were
2352 picked up by some special patterns. I have improved the selection
2353 criteria to be more general.
2355 PH/38 The host_find_failed option in the manualroute router can now be set
2356 to "ignore", to completely ignore a host whose IP address cannot be
2357 found. If all hosts are ignored, the behaviour is controlled by the new
2358 host_all_ignored option.
2360 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2361 homing or because of multiple MX records with /mx) generated more than
2362 one IP address, and the following item turned out to be the local host,
2363 all the secondary addresses of the first item were incorrectly removed
2364 from the list, along with the local host and any following hosts (which
2365 is what is supposed to happen).
2367 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2368 whoever called Exim into the -H file. In the case of the daemon it was
2369 behaving confusingly. When first started, it used values for whoever
2370 started the daemon, but after a SIGHUP it used the Exim user (because it
2371 calls itself on a restart). I have changed the code so that it now always
2374 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2375 message are rejected with the same error (e.g. no authentication or bad
2376 sender address), and a DATA command is nevertheless sent (as can happen
2377 with PIPELINING or a stupid MUA), the error message that was given to the
2378 RCPT commands is included in the rejection of the DATA command. This is
2379 intended to be helpful for MUAs that show only the final error to their
2382 PH/42 Another patch from the Sieve maintainer.
2384 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2385 Eximstats - Fixed some broken HTML links and added missing column headers
2387 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2388 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2390 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2392 PH/43 Yet another patch from the Sieve maintainer.
2394 PH/44 I found a way to check for a TCP/IP connection going away before sending
2395 the response to the final '.' that terminates a message, but only in the
2396 case where the client has not sent further data following the '.'
2397 (unfortunately, this is allowed). However, in many cases there won't be
2398 any further data because there won't be any more messages to send. A call
2399 to select() can be used: if it shows that the input is "ready", there is
2400 either input waiting, or the socket has been closed. An attempt to read
2401 the next input character can distinguish the two cases. Previously, Exim
2402 would have sent an OK response which the client would never have see.
2403 This could lead to message repetition. This fix should cure that, at
2404 least in a lot of common cases.
2406 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2407 advertised in response to EHLO.
2413 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2414 fixed by 4.65/MH/01 (is this a record?) are fixed:
2416 (i) An empty string was always treated as zero by the numeric comparison
2417 operators. This behaviour has been restored.
2419 (ii) It is documented that the numeric comparison operators always treat
2420 their arguments as decimal numbers. This was broken in that numbers
2421 starting with 0 were being interpreted as octal.
2423 While fixing these problems I realized that there was another issue that
2424 hadn't been noticed. Values of message_size_limit (both the global option
2425 and the transport option) were treated as octal if they started with 0.
2426 The documentation was vague. These values are now always treated as
2427 decimal, and I will make that clear in the documentation.
2433 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2434 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2437 MH/01 Don't check that the operands of numeric comparison operators are
2438 integers when their expansion is in "skipping" mode (fixes bug
2439 introduced by 4.64-PH/07).
2441 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2442 child addresses, Exim now panics and dies. Previously, because the count
2443 is held in a short int, deliveries were likely to be lost. As such a
2444 large number of recipients for a single message is ridiculous
2445 (performance will be very, very poor), I have chosen to impose a limit
2446 rather than extend the field.
2452 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2453 leftover -K file (the existence of which was triggered by #402).
2454 While we were at it, introduced process PID as part of the -K
2455 filename. This should rule out race conditions when creating
2458 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2459 processing considerably. Previous code took too long for large mails,
2460 triggering a timeout which in turn triggers #401.
2462 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2463 in the DK code in transports.c. sendfile() is not really portable,
2464 hence the _LINUX specificness.
2466 TF/01 In the add_headers option to the mail command in an Exim filter,
2467 there was a bug that Exim would claim a syntax error in any
2468 header after the first one which had an odd number of characters
2471 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2472 callout verification, Exim cached a "reject" for the entire domain. This
2473 is correct for most verifications, but it is not correct for a recipient
2474 verification with use_sender or use_postmaster set, because in that case
2475 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2476 case of MAIL FROM:<> rejection from other early rejections (e.g.
2477 rejection of HELO). When verifying a recipient using a non-null MAIL
2478 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2479 Whatever the result of the callout, the value of the domain cache is
2480 left unchanged (for any other kind of callout, getting as far as trying
2481 RCPT means that the domain itself is ok).
2483 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2486 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2487 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2488 session, the daemon ignores ECONNECT errors and logs others; it now
2489 ignores EPIPE as well.
2491 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2492 (quoted-printable decoding).
2494 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2495 later the small subsequent patch to fix an introduced bug.
2497 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2499 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2501 PH/08 An error is now given if message_size_limit is specified negative.
2503 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2504 to be given (somewhat) arbitrary names.
2506 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2509 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2510 miscellaneous code fixes
2512 PH/10 Added the log_reject_target ACL modifier to specify where to log
2515 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2516 hostname. This is wrong, because it relates to the incoming message (and
2517 probably the interface on which it is arriving) and not to the outgoing
2518 callout (which could be using a different interface). This has been
2519 changed to use the value of the helo_data option from the smtp transport
2520 instead - this is what is used when a message is actually being sent. If
2521 there is no remote transport (possible with a router that sets up host
2522 addresses), $smtp_active_hostname is used.
2524 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2525 tweaks were necessary in order to get it to work (see also 21 below):
2526 (a) The code assumed that strncpy() returns a negative number on buffer
2527 overflow, which isn't the case. Replaced with Exim's string_format()
2529 (b) There were several signed/unsigned issues. I just did the minimum
2530 hacking in of casts. There is scope for a larger refactoring.
2531 (c) The code used strcasecmp() which is not a standard C function.
2532 Replaced with Exim's strcmpic() function.
2533 (d) The code set only $1; it now sets $auth1 as well.
2534 (e) A simple test gave the error "authentication client didn't specify
2535 service in request". It would seem that Dovecot has changed its
2536 interface. Fortunately there's a specification; I followed it and
2537 changed what the client sends and it appears to be working now.
2539 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2542 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2543 address A is aliased to B and C, where B exists and C does not. Without
2544 -v the output is "A verified" because verification stops after a
2545 successful redirection if more than one address is generated. However,
2546 with -v the child addresses are also verified. Exim was outputting "A
2547 failed to verify" and then showing the successful verification for C,
2548 with its parentage. It now outputs "B failed to verify", showing B's
2549 parentage before showing the successful verification of C.
2551 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2552 look up a TXT record in a specific list after matching in a combined
2555 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2556 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2557 they consult the DNS. I had assumed they would set it the way they
2558 wanted; and indeed my experiments on Linux seem to show that in some
2559 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2560 To be on the safe side, however, I have now made the interface to
2561 host_find_byname() similar to host_find_bydns(), with an argument
2562 containing the DNS resolver options. The host_find_byname() function now
2563 sets these options at its start, just as host_find_bydns() does. The smtp
2564 transport options dns_qualify_single and dns_search_parents are passed to
2565 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2566 of host_find_byname() use the default settings of RES_DEFNAMES
2567 (qualify_single) but not RES_DNSRCH (search_parents).
2569 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2570 spool_read_header() do less string testing, by means of a preliminary
2571 switch on the second character of optional "-foo" lines. (This is
2572 overdue, caused by the large number of possibilities that now exist.
2573 Originally there were few.) While I was there, I also converted the
2574 str(n)cmp tests so they don't re-test the leading "-" and the first
2575 character, in the hope this might squeeze out yet more improvement.
2577 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2578 flag allowing group syntax was set by the header_syntax check but not
2579 turned off, possible causing trouble later; (2) The flag was not being
2580 set at all for the header_verify test, causing "group"-style headers to
2581 be rejected. I have now set it in this case, and also caused header_
2582 verify to ignore an empty address taken from a group. While doing this, I
2583 came across some other cases where the code for allowing group syntax
2584 while scanning a header line wasn't quite right (mostly, not resetting
2585 the flag correctly in the right place). These bugs could have caused
2586 trouble for malformed header lines. I hope it is now all correct.
2588 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2589 with the "reply" argument non-NULL. The code, however (which originally
2590 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2591 but it didn't always do it. This confused somebody who was copying the
2592 code for some other use. I have removed all the tests.
2594 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2595 feature that was used to support insecure browsers during the U.S. crypto
2596 embargo. It requires special client support, and Exim is probably the
2597 only MTA that supported it -- and would never use it because real RSA is
2598 always available. This code has been removed, because it had the bad
2599 effect of slowing Exim down by computing (never used) parameters for the
2600 RSA_EXPORT functionality.
2602 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2603 authenticator to fail if there's a tab character in the incoming data
2604 (there should never be unless someone is messing about, as it's supposed
2605 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2606 if the connection is using TLS or if the remote IP is the same as the
2607 local IP, and the "valid-client-cert option" if a client certificate has
2610 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2611 authenticators. This can be used for authorization after authentication
2612 succeeds. (In the case of plaintext, it servers for both authentication
2615 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2616 if any retry times were supplied.
2618 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2619 connection, where there is no client IP address to check. In this
2620 situation, the verify now always succeeds.
2622 PH/25 Applied John Jetmore's -Mset patch.
2624 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2626 PH/27 In a string expansion for a processed (not raw) header when multiple
2627 headers of the same name were present, leading whitespace was being
2628 removed from all of them, but trailing whitespace was being removed only
2629 from the last one. Now trailing whitespace is removed from each header
2630 before concatenation. Completely empty headers in a concatenation (as
2631 before) are ignored.
2633 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2634 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2636 PH/29 [Removed. This was a change that I later backed out, and forgot to
2637 correct the ChangeLog entry (that I had efficiently created) before
2638 committing the later change.]
2640 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2641 address errors (4xx response to RCPT) over the same connection as other
2642 messages routed to the same hosts. Such deliveries are always "forced",
2643 so retry times are not inspected. This resulted in far too many retries
2644 for the affected addresses. The effect occurred only when there were more
2645 hosts than the hosts_max_try setting in the smtp transport when it had
2646 the 4xx errors. Those hosts that it had tried were not added to the list
2647 of hosts for which the message was waiting, so if all were tried, there
2648 was no problem. Two fixes have been applied:
2650 (i) If there are any address or message errors in an SMTP delivery, none
2651 of the hosts (tried or untried) are now added to the list of hosts
2652 for which the message is waiting, so the message should not be a
2653 candidate for sending over the same connection that was used for a
2654 successful delivery of some other message. This seems entirely
2655 reasonable: after all the message is NOT "waiting for some host".
2656 This is so "obvious" that I'm not sure why it wasn't done
2657 previously. Hope I haven't missed anything, but it can't do any
2658 harm, as the worst effect is to miss an optimization.
2660 (ii) If, despite (i), such a delivery is accidentally attempted, the
2661 routing retry time is respected, so at least it doesn't keep
2662 hammering the server.
2664 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2665 in ${readsocket because some servers need this prod.
2667 PH/32 Added some extra debug output when updating a wait-xxx database.
2669 PH/33 The hint "could be header name not terminated by colon", which has been
2670 given for certain expansion errors for a long time, was not being given
2671 for the ${if def:h_colon_omitted{... case.
2673 PH/34 The spec says: "With one important exception, whenever a domain list is
2674 being scanned, $domain contains the subject domain." There was at least
2675 one case where this was not true.
2677 PH/35 The error "getsockname() failed: connection reset by peer" was being
2678 written to the panic log as well as the main log, but it isn't really
2679 panic-worthy as it just means the connection died rather early on. I have
2680 removed the panic log writing for the ECONNRESET error when getsockname()
2683 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2684 runs only) independently of the message's sender address. This meant
2685 that, if the 4xx error was in fact related to the sender, a different
2686 message to the same recipient with a different sender could confuse
2687 things. In particular, this can happen when sending to a greylisting
2688 server, but other circumstances could also provoke similar problems.
2689 I have changed the default so that the retry time for these errors is now
2690 based a combination of the sender and recipient addresses. This change
2691 can be overridden by setting address_retry_include_sender=false in the
2694 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2695 remote server are returned as part of bounce messages. This was not
2696 happening for LMTP over a pipe (the lmtp transport), but now it is the
2697 same for both kinds of LMTP.
2699 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2700 in header lines that were in fact CNAMEs. This is no longer the case.
2702 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2703 and queue runs started by the daemon processed all messages. This has
2704 been fixed so that -R and -S can now usefully be given with -q<time>.
2706 PH/40 Import PCRE release 6.7 (fixes some bugs).
2708 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2710 PH/42 Give an error if -q is specified more than once.
2712 PH/43 Renamed the variables $interface_address and $interface_port as
2713 $received_ip_address and $received_port, to make it clear that these
2714 values apply to message reception, and not to the outgoing interface when
2715 a message is delivered. (The old names remain recognized, of course.)
2717 PH/44 There was no timeout on the connect() call when using a Unix domain
2718 socket in the ${readsocket expansion. There now is.
2720 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2721 be meaningful with "accept".
2723 SC/01 Eximstats V1.43
2724 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2726 SC/02 Eximstats V1.44
2727 Use a glob alias rather than an array ref in the generated
2728 parser. This improves both readability and performance.
2730 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2731 Collect SpamAssassin and rejection statistics.
2732 Don't display local sender or destination tables unless
2733 there is data to show.
2734 Added average volumes into the top table text output.
2736 SC/04 Eximstats V1.46
2737 Collect data on the number of addresses (recipients)
2738 as well as the number of messages.
2740 SC/05 Eximstats V1.47
2741 Added 'Message too big' to the list of mail rejection
2742 reasons (thanks to Marco Gaiarin).
2744 SC/06 Eximstats V1.48
2745 Mainlog lines which have GMT offsets and are too short to
2746 have a flag are now skipped.
2748 SC/07 Eximstats V1.49 (Alain Williams)
2749 Added the -emptyok flag.
2751 SC/08 Eximstats V1.50
2752 Fixes for obtaining the IP address from reject messages.
2754 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2755 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2756 whitespace changes from 4.64-PH/27
2758 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2761 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2762 are found, allow negative numbers in numeric criteria)
2764 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2766 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2767 to match changes made in 4.64-PH/43
2769 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2771 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2772 "domain" is actually an IP address, and force a failure. This locks out
2773 those revolvers/nameservers that support "A-for-A" lookups, in
2774 contravention of the specifications.
2776 PH/48 When a host name was looked up from an IP address, and the subsequent
2777 forward lookup of the name timed out, the host name was left in
2778 $sender_host_name, contrary to the specification.
2780 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2781 restricted to single-key lookups, Exim was not diagnosing an error if
2782 * or *@ was used with a query-style lookup.
2784 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2786 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2787 long ago, but noone interested enough thought of it. Let's just say that
2788 the "1.1" means that there are some new functions that weren't there at
2789 some point in the past.
2791 PH/51 Error processing for expansion failure of helo_data from an smtp
2792 transport during callout processing was broken.
2794 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2795 tested/used via the -bh/-bhc/-bs options.
2797 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2798 bug, fixed in subsequent PCRE releases).
2800 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2801 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2803 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2809 SC/01 Use a glob alias rather than an array ref in eximstats generated
2810 parser. This improves both readability and performance.
2812 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2813 Don't display local sender or destination tables in eximstats unless
2814 there is data to show.
2815 Added average volumes into the eximstats top table text output.
2817 SC/03 Collect data on the number of addresses (recipients) as well
2818 as the number of messages in eximstats.
2820 TF/01 Correct an error in the documentation for the redirect router. Exim
2821 does (usually) call initgroups() when daemonizing.
2823 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2824 with consistent privilege compared to when running as a daemon.
2826 TF/03 Note in the spec that $authenticated_id is not set for local
2827 submissions from trusted users.
2829 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2830 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2832 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2833 by adding some example configuration directives to the default
2834 configuration file. A little bit of work is required to uncomment the
2835 directives and define how usernames and passwords are checked, but
2836 there is now a framework to start from.
2838 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2839 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2840 without this. I don't know how relevant this is to other LDAP libraries.
2842 PH/02 Add the verb name to the "unknown ACL verb" error.
2844 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2846 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2848 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2849 directory not expanded when it should be if an expanded home directory
2850 was set for the address (which is overridden by the transport).
2852 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2855 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2856 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2857 because it is too late at that time, and has no effect.
2859 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2860 security issue with \' (bugzilla #107). I could not use the
2861 PQescapeStringConn() function, because it needs a PGconn value as one of
2864 PH/08 When testing addresses using -bt, indicate those final addresses that
2865 are duplicates that would not cause an additional delivery. At least one
2866 person was confused, thinking that -bt output corresponded to deliveries.
2867 (Suppressing duplicates isn't a good idea as you lose the information
2868 about possibly different redirections that led to the duplicates.)
2870 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2871 systems where poll() doesn't work, in particular OS X.
2873 PH/10 Added more information to debugging output for retry time not reached.
2875 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2876 operations in malware.c.
2878 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2881 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2882 syslog_duplication set false, log lines that would normally be written
2883 both the the main log and to the reject log were not written to syslog at
2886 PH/14 In the default configuration, change the use of "message" in ACL warn
2887 statements to "add_header".
2889 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2890 not followed by a command (e.g. "seen endif").
2892 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2893 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2896 PH/17 Added extra conditions to the default value of delay_warning_condition
2900 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2901 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2902 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2905 The Auto-Submitted: and various List- headers are standardised, whereas I
2906 don't think Precedence: ever was.
2908 PH/18 Refactored debugging code in route_finduser() to show more information,
2909 in particular, the error code if getpwnam() issues one.
2911 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2912 This is apparently needed in addition to the PH/07 change above to avoid
2913 any possible encoding problems.
2915 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2916 but not after initializing Perl.
2918 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2919 output them only if debugging. By default they are written stderr,
2920 apparently, which is not desirable.
2922 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2925 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2928 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2930 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2931 authenticated or an ident call has been made. Suppress the default
2932 values for $authenticated_id and $authenticated_sender (but permit -oMai
2933 and -oMas) when testing with -bh.
2935 PH/24 Re-jigged the order of the tests in the default configuration so that the
2936 tests for valid domains and recipients precede the DNS black list and CSA
2937 tests, on the grounds that those ones are more expensive.
2939 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2940 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2941 This bug exists in every version of Exim that I still have, right back to
2944 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2945 However, an attempt to turn on case-sensitivity in a regex key by
2946 including (?-i) didn't work because the subject string was already
2947 lowercased, and the effects were non-intuitive. It turns out that a
2948 one-line patch can be used to allow (?-i) to work as expected.
2954 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2955 other effects) broke the use of negated acl sub-conditions.
2957 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2960 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2961 "Deny" causes Exim to reject the incoming connection with a 554 error.
2962 Unfortunately, if there is a major crisis, such as a disk failure,
2963 tcp-wrappers gives "deny", whereas what one would like would be some
2964 kind of temporary error. A kludge has been added to help with this.
2965 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2966 554 error is used if errno is still zero or contains ENOENT (which occurs
2967 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2970 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2972 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2973 errors. Otherwise a message that provokes a temporary error (when other
2974 messages do not) can cause a whole host to time out.
2976 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2977 addresses were routed directly to files or pipes from a redirect router.
2978 File deliveries just didn't batch; pipe deliveries might have suffered
2981 PH/06 A failure to get a lock for a hints database would erroneously always say
2982 "Failed to get write lock", even when it was really a read lock.
2984 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2985 of 0600. This has been changed to use the value of the lockfile_mode
2986 option (which defaults to 0600).
2988 PH/08 Applied small patch from the Sieve maintainer.
2990 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2991 folder from quota calculations, a direct delivery into this folder messed
2992 up the contents of the maildirsize file. This was because the regex was
2993 used only to exclude .Trash (or whatever) when the size of the mailbox
2994 was calculated. There was no check that a delivery was happening into an
2995 excluded directory. This bug has been fixed by ignoring all quota
2996 processing for deliveries into excluded directories.
2998 PH/10 Added the maildirfolder_create_regex option to appendfile.
3004 PH/01 The code for finding all the local interface addresses on a FreeBSD
3005 system running IPv6 was broken. This may well have applied to all BSD
3006 systems, as well as to others that have similar system calls. The broken
3007 code found IPv4 interfaces correctly, but gave incorrect values for the
3008 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3009 that it would not match correctly against @[] and not recognize the IPv6
3012 PH/02 The ipliteral router was not recognizing addresses of the form user@
3013 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3015 PH/03 Added disable_ipv6.
3017 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3018 decoded form, because it is most often used to construct To: headers
3019 lines in autoreplies, and the decoded form may well be syntactically
3020 invalid. However, $reply_address has leading white space removed, and all
3021 newlines turned into spaces so that the autoreply transport does not
3024 PH/05 If group was specified without a user on a router, and no group or user
3025 was specified on a transport, the group from the router was ignored.
3027 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3028 for visible compile-time settings that can be used to change these
3029 numbers, for those that want even more. Backwards compatibility with old
3030 spool files has been maintained. However, going back to a previous Exim
3031 release will lost any variables that are in spool files.
3033 PH/07 Two small changes when running in the test harness: increase delay when
3034 passing a TCP/IP connection to a new process, in case the original
3035 process has to generate a bounce, and remove special handling of
3036 127.0.0.2 (sic), which is no longer necessary.
3038 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3039 be the same on different OS.
3041 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3044 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3045 whether --show-vars was specified or not
3047 JJ/02 exipick: Added support for new ACL variable spool format introduced
3050 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3051 syntactically invalid From: or Reply-to: line, and a filter used this to
3052 generate an autoreply, and therefore failed to obtain an address for the
3053 autoreply, Exim could try to deliver to a non-existent relative file
3054 name, causing unrelated and misleading errors. What now happens is that
3055 it logs this as a hard delivery error, but does not attempt to create a
3058 PH/11 The exinext utility has a -C option for testing purposes, but although
3059 the given file was scanned by exinext itself; it wasn't being passed on
3060 when Exim was called.
3062 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3063 an end-of-file indication when reading a command response.
3065 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3066 compiled. In many other places in Exim, IPv6 addresses are always
3067 recognized, so I have changed this. It also means that IPv4 domain
3068 literals of the form [IPV4:n.n.n.n] are now always recognized.
3070 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3071 used if the router is not running as root, for example, when verifying at
3072 ACL time, or when using -bh. The debugging output from this situation was
3073 non-existent - all you got was a failure to exec. I have made two
3076 (a) Failures to set uid/gid, the current directory, or a process leader
3077 in a subprocess such as that created by queryprogram now generate
3078 suitable debugging output when -d is set.
3080 (b) The queryprogram router detects when it is not running as root,
3081 outputs suitable debugging information if -d is set, and then runs
3082 the subprocess without attempting to change uid/gid.
3084 PH/15 Minor change to Makefile for building test_host (undocumented testing
3087 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3088 additional section of a DNS packet that returns MX or SRV records.
3089 Instead, it always explicitly searches for A/AAAA records. This avoids
3090 major problems that occur when a DNS server includes only records of one
3091 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3092 fixed another bug: if SRV records were looked up and the corresponding
3093 address records were *not* found in the additional section, the port
3094 values from the SRV records were lost.
3096 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3097 using the correct key (the original address) when searching the retry
3098 rules in order to find which one to use for generating the retry hint.
3100 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3101 adding the default one. Similarly, if it contains a Reply-To: header, the
3102 errors_reply_to option, if set, is not used.
3104 PH/19 When calculating a retry time, Exim used to measure the "time since
3105 failure" by looking at the "first failed" field in the retry record. Now
3106 it does not use this if it is later than than the arrival time of the
3107 message. Instead it uses the arrival time. This makes for better
3108 behaviour in cases where some deliveries succeed, thus re-setting the
3109 "first failed" field. An example is a quota failure for a huge message
3110 when small messages continue to be delivered. Without this change, the
3111 "time since failure" will always be short, possible causing more frequent
3112 delivery attempts for the huge message than are intended.
3113 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3115 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3116 $1, $2, $3) because the numerical variables can be reset during some
3117 expansion items (e.g. "match"), thereby losing the authentication data.
3119 PH/21 Make -bV show the size of off_t variables so that the test suite can
3120 decide whether to run tests for quotas > 2G.
3122 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3123 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3124 filecount value is greater than 2G or if a quota value is greater than 2G
3125 on a system where the size of off_t is not greater than 4, a panic error
3128 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3129 never match. The debug and -bh output now contains an explicit error
3130 message indicating a malformed IPv4 address or mask.
3132 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3133 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3134 PH/23 above applies.
3136 PH/25 Do not write to syslog when running in the test harness. The only
3137 occasion when this arises is a failure to open the main or panic logs
3138 (for which there is an explicit test).
3140 PH/26 Added the /no_tell option to "control=freeze".
3142 PH/27 If a host name lookup failed very early in a connection, for example, if
3143 the IP address matched host_lookup and the reverse lookup yielded a name
3144 that did not have a forward lookup, an error message of the form "no IP
3145 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3146 could be logged. Now it outputs the IP address instead of "NULL".
3148 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3149 allows the sender and the authenticated sender to be set when
3150 submitting a message from within Exim. Since child_open_exim() is
3151 documented for local_scan(), the new function should be too.
3153 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3154 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3155 results in an empty string is now treated as unset.
3157 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3159 PH/31 Added sender_verify_fail logging option.
3161 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3162 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3163 tidied the source and removed it altogether.
3165 PH/33 When a queue run was abandoned because the load average was too high, a
3166 log line was always written; now it is written only if the queue_run log
3167 selector is set. In addition, the log line for abandonment now contains
3168 information about the queue run such as the pid. This is always present
3169 in "start" and "stop" lines but was omitted from the "abandon" line.
3171 PH/34 Omit spaces between a header name and the colon in the error message that
3172 is given when verify = headers_syntax fails (if there are lots of them,
3173 the message gets confusing).
3175 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3176 names, as there are now some PTR records that contain slashes. This check
3177 is only to protect against broken name servers that fall over on strange
3178 characters, so the fact that it applies to all lookups doesn't matter.
3180 PH/36 Now that the new test suite is complete, we can remove some of the
3181 special code in Exim that was needed for the old test suite. For example,
3182 sorting DNS records because real resolvers return them in an arbitrary
3183 order. The new test suite's fake resolver always returns records in the
3186 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3187 bounces) except when queue_only is set, to avoid logging races between
3188 the different processes.
3190 PH/38 Panic-die if .include specifies a non-absolute path.
3192 PH/39 A tweak to the "H" retry rule from its user.
3194 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3195 a label. They prevented compilation on older perls.
3197 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3198 a warning to be raised on newish perls.
3200 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3201 on queue. Changes to match documented behaviour of showing count of
3202 messages matching specified criteria.
3204 PH/40 Changed the default ident timeout from 30s to 5s.
3206 PH/41 Added support for the use of login_cap features, on those BSD systems
3207 that have them, for controlling the resources used by pipe deliveries.
3209 PH/42 The content-scanning code uses fopen() to create files in which to put
3210 message data. Previously it was not paying any attention to the mode of
3211 the files. Exim runs with umask(0) because the rest of the code creates
3212 files with open(), and sets the required mode explicitly. Thus, these
3213 files were ending up world-writeable. This was not a big issue, because,
3214 being within the spool directory, they were not world-accessible. I have
3215 created a function called modefopen, which takes an additional mode
3216 argument. It sets umask(777), creates the file, chmods it to the required
3217 mode, then resets the umask. All the relevant calls to fopen() in the
3218 content scanning code have been changed to use this function.
3220 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3221 to 24 hours. This avoids potential overflow problems when processing G
3222 and H retry rules. I suspect nobody ever tinkers with this value.
3224 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3226 PH/45 When the plaintext authenticator is running as a client, the server's
3227 challenges are checked to ensure they are valid base64 strings. By
3228 default, the authentication attempt is cancelled if an invalid string is
3229 received. Setting client_ignore_invalid_base64 true ignores these errors.
3230 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3231 they are received. Thus, the responses can be made to depend on the
3232 challenges. If an invalid string is ignored, an empty string is placed in
3235 PH/46 Messages that are created by the autoreply transport now contains a
3236 References: header, in accordance with RFCs 2822 and 3834.
3238 PH/47 Added authenticated_sender_force to the smtp transport.
3240 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3242 PH/49 Installed latest patch from the Sieve maintainer.
3244 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3245 was also set, the appendfile transport was unnecessarily scanning a
3246 directory of message files (e.g. for maildir delivery) to find the count
3247 of files (along with the size), even though it did not need this
3248 information. It now does the scan only if it needs to find either the
3249 size of the count of files.
3251 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3253 PH/52 Two bugs concerned with error handling when the smtp transport is
3256 (i) Exim was not creating retry information for temporary errors given
3257 for individual recipients after the DATA command when the smtp transport
3258 was used in LMTP mode. This meant that they could be retried too
3259 frequently, and not timed out correctly.
3261 (ii) Exim was setting the flag that allows error details to be returned
3262 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3263 recipients that were returned after the DATA command.
3265 PH/53 This is related to PH/52, but is more general: for any failing address,
3266 when detailed error information was permitted to be returned to the
3267 sender, but the error was temporary, then after the final timeout, only
3268 "retry timeout exceeded" was returned. Now it returns the full error as
3269 well as "retry timeout exceeded".
3271 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3272 do this, and (what is worse) MTAs that accept it.
3274 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3275 will now be deprecated.
3277 PH/56 New os.c-cygwin from the Cygwin maintainer.
3279 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3280 formats (previously only available in exim formats via -bpr, -bpru,
3281 and -bpra. Now also available in native and exiqgrep formats)
3283 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3284 with very large, slow to parse queues
3286 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3288 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3290 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3291 responses to authentication challenges, though it was showing the
3292 challenges; (ii) I've removed the CR characters from the debug output for
3295 PH/58 Allow for the insertion of a newline as well as a space when a string
3296 is turned into more than one encoded-word during RFC 2047 encoding. The
3297 Sieve code now uses this.
3299 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3300 data_4xx, lost_connection, tls_required.
3302 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3303 message was being sent as an SMTP response.
3305 PH/61 Add -l and -k options to exicyclog.
3307 PH/62 When verifying, if an address was redirected to one new address, so that
3308 verification continued, and the new address failed or deferred after
3309 having set something in $address_data, the value of $address_data was not
3310 passed back to the ACL. This was different to the case when no
3311 redirection occurred. The value is now passed back in both cases.
3313 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3314 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3315 use login_cap.h, so on its own it isn't the distinguishing feature. The
3316 new name refers directly to the setclassresources() function.
3318 PH/65 Added configuration files for NetBSD3.
3320 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3322 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3323 is preferred over IPv4.
3325 PH/68 The bounce_return_message and bounce_return_body options were not being
3326 honoured for bounces generated during the reception of non-SMTP messages.
3327 In particular, this applied to messages rejected by the ACL. This bug has
3328 been fixed. However, if bounce_return_message is true and bounce_return_
3329 body is false, the headers that are returned for a non-SMTP message
3330 include only those that have been read before the error was detected.
3331 (In the case of an ACL rejection, they have all been read.)
3333 PH/69 The HTML version of the specification is now built in a directory called
3334 spec_html instead of spec.html, because the latter looks like a path with
3335 a MIME-type, and this confuses some software.
3337 PH/70 Catch two compiler warnings in sieve.c.
3339 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3340 function verify_get_ident() calls ip_connect() to connect a socket, but
3341 if the "connect()" function timed out, ip_connect() used to close the
3342 socket. However, verify_get_ident() also closes the socket later, and in
3343 between Exim writes to the log, which may get opened at this point. When
3344 the socket was closed in ip_connect(), the log could get the same file
3345 descriptor number as the socket. This naturally causes chaos. The fix is
3346 not to close the socket in ip_connect(); the socket should be closed by
3347 the function that creates it. There was only one place in the code where
3348 this was missing, in the iplookup router, which I don't think anybody now
3349 uses, but I've fixed it anyway.
3351 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3352 well as to direct DNS lookups. Otherwise the handling of names in host
3353 lists is inconsistent and therefore confusing.
3359 PH/01 Two changes to the default runtime configuration:
3361 (1) Move the checks for relay_from_hosts and authenticated clients from
3362 after to before the (commented out) DNS black list checks.
3364 (2) Add control=submission to the relay_from_hosts and authenticated
3365 clients checks, on the grounds that messages accepted by these
3366 statements are most likely to be submissions.
3368 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3370 (1) Generate an error if the third argument for the ${prvs expansion is
3373 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3376 (3) Reset the variables that are obtained from the first argument of
3377 ${prvscheck and used in the second argument before leaving the code,
3378 because their memory is reclaimed, so using them afterwards may do
3381 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3382 one (it's much easier than Tom thought :-).
3384 (5) Because of (4), we can now allow for the use of $prvscheck_result
3385 inside the third argument.
3387 PH/03 For some reason, the default setting of PATH when running a command from
3388 a pipe transport was just "/usr/bin". I have changed it to
3391 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3392 anything to be listed in the output from -bV.
3394 PH/05 When a filter generated an autoreply, the entire To: header line was
3395 quoted in the delivery log line, like this:
3397 => >A.N.Other <ano@some.domain> <original@ddress> ...
3399 This has been changed so that it extracts the operative address. There
3400 may be more than one such address. If so, they are comma-separated, like
3403 => >ano@some.domain,ona@other.domain <original@ddress> ...
3405 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3406 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3407 not being looked up in the rDNS to get a host name, Exim was showing the
3408 IP address twice in Received: lines, even though the IP addresses were
3409 identical. For example:
3411 Received: from [1.2.3.4] (helo=[1.2.3.4])
3413 However, if the real host name was known, it was omitting the HELO data
3414 if it matched the actual IP address. This has been tidied up so that it
3415 doesn't show the same IP address twice.
3417 PH/07 When both +timestamp and +memory debugging was on, the value given by
3418 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3419 called by the debug printing, thereby overwriting the timestamp buffer.
3420 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3422 PH/08 When the original message was included in an autoreply transport, it
3423 always said "this is a copy of the message, including all the headers",
3424 even if body_only or headers_only was set. It now gives an appropriate
3427 PH/09 Applied a patch from the Sieve maintainer which:
3429 o fixes some comments
3430 o adds the (disabled) notify extension core
3431 o adds some debug output for the result of if/elsif tests
3432 o points to the current vacation draft in the documentation
3433 and documents the missing references header update
3437 o fixes a bug in processing the envelope test (when testing
3438 multiple envelope elements, the last element determined the
3441 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3442 Electronic Mail") by including:
3444 Auto-submitted: auto-generated
3446 in the messages that it generates (bounce messages and others, such as
3447 warnings). In the case of bounce messages for non-SMTP messages, there was
3448 also a typo: it was using "Auto_submitted" (underscore instead of
3449 hyphen). Since every message generated by Exim is necessarily in response
3450 to another message, thes have all been changed to:
3452 Auto-Submitted: auto-replied
3454 in accordance with these statements in the RFC:
3456 The auto-replied keyword:
3458 - SHOULD be used on messages sent in direct response to another
3459 message by an automatic process,
3461 - MUST NOT be used on manually-generated messages,
3463 - MAY be used on Delivery Status Notifications (DSNs) and Message
3464 Disposition Notifications (MDNs),
3466 - MUST NOT be used on messages generated by automatic or periodic
3467 processes, except for messages which are automatic responses to
3470 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3471 to the default Received: header definition.
3473 PH/12 Added log selector acl_warn_skipped (default on).
3475 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3476 variables because (a) they are in the wrong storage pool and (b) even if
3477 they were copied, it wouldn't work properly because of the caching.
3479 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3480 checking when decoding. Apparently there are clients that generate
3481 overlong encoded strings. Why am I not surprised?
3483 PH/15 If the first argument of "${if match_address" was not empty, but did not
3484 contain an "@" character, Exim crashed. Now it writes a panic log message
3485 and treats the condition as false.
3487 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3489 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3490 extension "envelope-auth". The code is finished and in agreement with
3491 other implementations, but there is no documentation so far and in fact,
3492 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3493 not changing the active code.
3495 Print executed "if" and "elsif" statements when debugging is used. This
3496 helps a great deal to understand what a filter does.
3498 Document more things not specified clearly in RFC3028. I had all this
3499 sorted out, when out of a sudden new issues came to my mind. Oops."
3501 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3504 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3505 canonical form (as documented). However, after a host name lookup from
3506 the IP address, check_host() was doing a simple string comparison with
3507 addresses acquired from the DNS when checking that the found name did
3508 have the original IP as one of its addresses. Since any found IPv6
3509 addresses are likely to be in abbreviated form, the comparison could
3510 fail. Luckily, there already exists a function for doing the comparison
3511 by converting both addresses to binary, so now that is used instead of
3512 the text comparison.
3514 PH/20 There was another similar case to PH/19, when a complete host name was
3515 given in a host list; looking up its IP address could give an abbreviated
3516 form, whereas the current host's name might or might not be abbreviated.
3517 The same fix has been applied.
3523 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3524 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3527 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3528 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3530 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3532 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3533 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3534 Also, for "Auto-Submitted", treat anything other than "no" as
3535 non-personal, in accordance with RFC 3834. (Previously it treated
3536 anything starting "auto-" as non-personal.)
3538 TF/01 The control=submission/name=... option had a problem with syntax
3539 errors if the name included a slash character. The /name= option
3540 now slurps the rest of the string, so it can include any characters
3541 but it must come last in the list of options (after /sender_retain
3544 PH/05 Some modifications to the interface to the fake nameserver for the new
3552 TK/01 Added the "success_on_redirect" address verification option. See
3553 NewStuff for rationale and an example.
3555 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3557 PH/02 Patch to exigrep to allow it to work on syslog lines.
3559 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3560 fread() to skip over the body file's header line, because in Cygwin the
3561 header line is locked and is inaccessible.
3563 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3564 co-exist for some time) to make it clear that it is the Exim ID that is
3565 referenced, not the Message-ID: header line.
3567 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3568 string_format() function, because snprintf() does not exist on all
3571 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3572 ${stat: expansion item.
3574 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3575 protocol synchronization error", to keep the pedants happy.
3577 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3578 well as for IRIX systems, when gcc is being used. See the host.c source
3581 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3583 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3586 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3587 transport and to the smtp transport in LMTP mode.
3589 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3591 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3592 run a filter in a subprocess. This could lead to confusion in subsequent
3593 lookups in the parent process. There should also be a search_tidyup() at
3594 the end of the subprocess.
3596 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3597 only if the host matched helo_try_verify_hosts, which caused the
3598 verification to occur when the EHLO/HELO command was issued. The ACL just
3599 tested the remembered result. Now, if a previous verification attempt has
3600 not happened, "verify = helo" does it there and then.
3602 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3604 TK/03 Fix log output including CR from clamd.
3606 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3607 exist provoked a memory error which could cause a segfault.
3609 PH/15 Installed PCRE 6.2
3611 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3613 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3614 of the problem. Specifically, suggested +O2 rather than +O1 for the
3617 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3619 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3620 record, the port was not correctly logged when the outgoing_port log
3621 selector was set (it logged the transort's default port).
3623 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3624 fallback_hosts, and "hosts" in the smtp transport.
3626 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3627 host errors such as "Connection refused".
3629 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3630 authentication with radiusclient 0.4.9:
3632 - Error returned from rc_read_config was caught wrongly
3633 - Username/password not passed on to radius server due to wrong length.
3635 The presumption is that some radiusclient API changes for 4.51/PH/17
3636 were not taken care of correctly. The code is still untested by me (my
3637 Linux distribution still has 0.3.2 of radiusclient), but it was
3638 contributed by a Radius user.
3640 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3641 expanding the "port" option of the smtp transport.
3643 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3644 while reading a MIME header. Thanks to Tom Hughes for a patch.
3646 PH/24 Include config.h inside local_scan.h so that configuration settings are
3649 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3650 that in an ACL for RCPT (for example), you can examine exactly what was
3653 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3654 commands, but it was not correctly comparing the address with the actual
3655 client host address. Thus, it would show the EHLO address in Received:
3656 header lines when this was not necessary.
3658 PH/27 Added the % operator to ${eval:}.
3660 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3661 it should be ignoring failures (because with -C, for example, it has lost
3662 privilege). It wasn't ignoring creation failures other than "already
3665 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3668 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3669 because an input line was too long, either on its own, or by virtue of
3670 too many continuations, the temporary file was not being removed, and the
3671 return code was incorrect.
3673 PH/31 Missing "BOOL" in function definition in filtertest.c.
3675 PH/32 Applied Sieve patches from the maintainer.
3677 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3679 PH/33 Added "verify = not_blind".
3681 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3682 Local/Makefile (with some defaults set). These are used in built scripts
3683 such as exicyclog, but they have never been used in the exim_install
3684 script (though there are many overriding facilities there). I have
3685 arranged that the exim_install script now takes note of these two
3688 PH/35 Installed configuration files for Dragonfly.
3690 PH/36 When a locally submitted message by a trusted user did not contain a
3691 From: header, and the sender address was obtained from -f or from an SMTP
3692 MAIL command, and the trusted user did not use -F to supply a sender
3693 name, $originator_name was incorrectly used when constructing a From:
3694 header. Furthermore, $originator_name was used for submission mode
3695 messages from external hosts without From: headers in a similar way,
3696 which is clearly wrong.
3698 PH/37 Added control=suppress_local_fixups.
3700 PH/38 When log_selector = +received_sender was set, and the addition of the
3701 sender made the log line's construction buffer exactly full, or one byte
3702 less than full, an overflow happened when the terminating "\n" was
3705 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3706 when the result of a list match is failure because a DNS lookup failed.
3708 PH/40 RM_COMMAND is now used in the building process.
3710 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3711 the "build-* directories that it finds.
3713 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3714 address in a domain literal was a prefix of an interface address.
3716 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3717 when verifying a sender address, unless rewrite_headers is false.
3719 PH/44 Wrote a long comment about why errors_to addresses are verified as
3720 recipients, not senders.
3722 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3723 the ratelimit ACL was added.
3725 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3727 PH/46 Added extra information about PostgreSQL errors to the error string.
3729 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3730 avoiding the need to install special zones in a real server. This is
3731 backwards compatible; if it can't find the fake resolver, it drops back.
3732 Thus, both old and new test suites can be run.
3734 TF/02 Added util/ratelimit.pl
3736 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3737 clock is set back in time.
3739 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3740 Candler <B.Candler@pobox.com>.
3742 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3743 OK for addresses that are the result of redirections.
3745 PH/48 A number of further additions for the benefit of the new test suite,
3746 including a fake gethostbyname() that interfaces to the fake DNS resolver
3749 TF/06 The fix for widen_domains has also been applied to qualify_single and
3750 search_parents which are the other dnslookup options that can cause
3753 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3756 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3758 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3759 service" if it encountered an MX record with an empty target hostname.
3760 The message is now "an MX or SRV record indicated no SMTP service".
3762 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3763 if the DNS of the sending site is misconfigured. This is quite a
3764 common situation. This change restores the behaviour of treating a
3765 helo verification defer as a failure.
3767 PH/51 If self=fail was set on a router, the bounce message did not include the
3768 actual error message.
3774 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3776 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3777 command itself ended in error, the underlying message about the transport
3778 filter timeout was being overwritten with the pipe command error. Now the
3779 underlying error message should be appended to the second error message.
3781 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3783 PH/02 Reduce the amount of output that "make" produces by default. Full output
3784 can still be requested.
3786 PH/03 The warning log line about a condition test deferring for a "warn" verb
3787 was being output only once per connection, rather than after each
3788 occurrence (because it was using the same function as for successful
3789 "warn" verbs). This seems wrong, so I have changed it.
3791 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3792 it should not have, which might have caused a crash in the right
3793 circumstances, but probably never did.
3795 PH/04 Installed a modified version of Tony Finch's patch to make submission
3796 mode fix the return path as well as the Sender: header line, and to
3797 add a /name= option so that you can make the user's friendly name appear
3800 TF/03 Added the control = fakedefer ACL modifier.
3802 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3803 Mark Lowes for thorough testing.
3805 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3807 TK/03 Merged latest SRS patch from Miles Wilton.
3809 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3810 with the definition in sysexits.h (which is #included earlier).
3811 Fortunately, Exim does not actually use EX_OK. The code used to try to
3812 preserve the sysexits.h value, by assuming that macro definitions were
3813 scanned for macro replacements. I have been disabused of this notion,
3814 so now the code just undefines EX_OK before #including unistd.h.
3816 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3817 in the smtp transport. When a block could not be written in a single
3818 write() function, the timeout was being re-applied to each part-write.
3819 This seems wrong - if the receiver was accepting one byte at a time it
3820 would take for ever. The timeout is now adjusted when this happens. It
3821 doesn't have to be particularly precise.
3823 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3824 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3826 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3827 without a domain if the check to <postmaster@domain> fails.
3829 SC/01 Eximstats: added -xls and the ability to specify output files
3830 (patch written by Frank Heydlauf).
3832 SC/02 Eximstats: use FileHandles for outputting results.
3834 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3836 SC/04 Eximstats: fixed display of large numbers with -nvr option
3838 SC/05 Eximstats: fixed merging of reports with empty tables.
3840 SC/06 Eximstats: added the -include_original_destination flag
3842 SC/07 Eximstats: removed tabs and trailing whitespace.
3844 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3846 TK/06 MBOX spool code: Add real "From " MBOX separator line
3847 so the .eml file is really in mbox format (even though
3848 most programs do not really care). Patch from Alex Miller.
3850 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3851 The latter is generated from $received_to and is only set if the
3852 message has one envelope recipient. SA can use these headers,
3853 obviously out-of-the-box. Patch from Alex Miller.
3855 PH/08 The ${def test on a variable was returning false if the variable's
3856 value was "0", contrary to what the specification has always said!
3857 The result should be true unless the variable is empty.
3859 PH/09 The syntax error of a character other than { following "${if
3860 def:variable_name" (after optional whitespace) was not being diagnosed.
3861 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3862 accidental colon was present, for example, could give incorrect results.
3864 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3865 result is used (not including appendfile, where other changes are about
3868 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3869 This involved changing a lot of size variables from int to off_t. It
3870 should work with maildirs and everything.
3872 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3873 spamd dying while we are connected to it.
3875 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3878 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3879 function for BDB 4.3.
3881 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3883 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3884 the sources for PCRE have been re-arranged and more files are now
3887 PH/15 The code I had for printing potentially long long variables in PH/11
3888 above was not the best (it lost precision). The length of off_t variables
3889 is now inspected at build time, and an appropriate printing format (%ld
3890 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3891 to be "long long int" or "long int". This is needed for the internal
3892 formatting function string_vformat().
3894 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3895 the configuration file to be ":syslog", then the script "guesses" where
3896 the logs files are, rather than using the compiled in default. In our
3897 case the guess is not the same as the compiled default, so the script
3898 suddenly stopped working when I started to use syslog. The patch checks
3899 to see if log_file_path is "". If so, it attempts to read it from exim
3900 with no configuration file to get the compiled in version, before it
3901 falls back to the previous guessing code."
3903 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3904 implementing BATV in an Exim configuration. See NewStuff for the gory
3907 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3908 Makefile that are specific to HP-UX.
3910 PH/18 If the "use_postmaster" option was set for a recipient callout together
3911 with the "random" option, the postmaster address was used as the MAIL
3912 FROM address for the random test, but not for the subsequent recipient
3913 test. It is now used for both.
3915 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3916 patch removes a few documentation additions to RFC 3028, because the
3917 latest draft now contains them. It adds the new en;ascii-case comparator
3918 and a new error check for 8bit text in MIME parts. Comparator and
3919 require names are now matched exactly. I enabled the subaddress
3920 extension, but it is not well tested yet (read: it works for me)."
3922 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3923 rework some of the code of TK/09 above to avoid the hardwired use of
3924 "%lld" and "long long". Replaced the call to snprintf() with a call to
3927 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3928 records point to non-existent hosts", "retry timeout exceeded", and
3929 "retry time not reached for any host after a long failure period".
3931 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3932 experimental DomainKeys support:
3934 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3935 (2) On an error such as an illegally used "control", the wrong name for
3936 the control was given.
3938 These problems did NOT occur unless DomainKeys support was compiled.
3940 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3942 PH/24 Added ${if match_ip condition.
3944 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3945 databases so that it will be absolutely obvious if a crash occurs in the
3946 DB library. This is a regular occurrence (often caused by mis-matched
3949 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3950 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3951 picked up on a user's system that detects such things. There doesn't seem
3952 to be a gcc warning option for this - only an attribute that has to be
3953 put on the function's prototype. It seems that in Fedora Core 4 they have
3954 set this on a number of new functions. No doubt there will be more in due
3957 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3958 specify a transport. However, if an address that was verified by such a
3959 router was the subject of a callout, Exim crashed because it tried to
3960 read the rcpt_include_affixes from the non-existent transport. Now it
3961 just assumes that the setting of that option is false. This bug was
3962 introduced by 4.51/PH/31.
3964 PH/28 Changed -d+all to exclude +memory, because that information is very
3965 rarely of interest, but it makes the output a lot bigger. People tend to
3966 do -d+all out of habit.
3968 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3969 code in os-type was giving problems when libc.so lives in lib64, like on
3972 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3973 aren't the modern standard, and it seems that some systems' include files
3974 don't always have them. Exim was already checking for some of the newer
3975 ones like T_AAAA, and defining it itself. I've added checks for all the
3976 record types that Exim uses.
3978 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3979 not automatically generating a new one, as it is supposed to. This
3980 prevented TLS from working. If the file did exist, but contained invalid
3981 data, a new version was generated, as expected. It was only the case of a
3982 non-existent file that was broken.
3984 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3985 with a change in libdomainkeys > 0.64.
3987 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3988 from DNS. If the selector record carries the flag, it now has
3989 precedence over the domain-wide flag.
3991 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3993 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3994 the use of a port name that isn't defined in /etc/services) occurred, the
3995 message was deferred as in a normal delivery, and thus remained on the
3996 spool, instead of being failed because of the mua_wrapper setting. This
3997 is now fixed, and I tidied up some of the mua_wrapper messages at the
4000 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4001 the messages in a hash of arrays rather than using individual hashes.
4002 This is a bit cleaner and results in dramatic memory savings, albeit
4003 at a slight CPU cost.
4005 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4006 as requested by Marc Sherman.
4008 SC/10 Eximstats: added histograms for user specified patterns as requested
4011 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4013 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4014 fopen() in the content-scanning modules that did not already have it.
4020 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4021 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4023 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4025 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4027 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4028 to test Sieve filters that use "vacation".
4030 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4031 that changes the way the GnuTLS parameters are stored in the cache file.
4032 The new format can be generated externally. For backward compatibility,
4033 if the data in the cache doesn't make sense, Exim assumes it has read an
4034 old-format file, and it generates new data and writes a new file. This
4035 means that you can't go back to an older release without removing the
4038 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4039 work if there are any delivery delays because "one_time" forces the
4040 parent to be marked "delivered", so its unseen clone is never tried
4041 again. For this reason, Exim now forbids the simultaneous setting of
4044 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4045 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4046 entry if you want to know the details. The fix, however, neglected to
4047 consider the case when local delivery batching is involved. The test for
4048 "previously delivered" was not happening when checking to see if an
4049 address could be batched with a previous (undelivered) one; under
4050 certain circumstances this could lead to multiple deliveries to the same
4053 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4054 in its include files, and this causes problems building Exim.
4056 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4057 header_syntax) but Exim was just ignoring anything given after a slash.
4058 In particular, this caused confusion with an attempt to use "verify =
4059 reverse_host_lookup/defer_ok". An error is now given when options are
4060 supplied for verify items that do not have them. (Maybe reverse_host_
4061 lookup should have a defer_ok option, but that's a different point.)
4063 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4064 defined by RFC 821) to 2048, because there were problems with some AUTH
4065 commands, and RFC 1869 says the size should be increased for extended
4066 SMTP commands that take arguments.
4068 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4071 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4072 "unknown" error; now it says that the functionality isn't in the binary.
4074 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4075 an address' error message when a string expansion fails (syntax or
4076 whatever). Otherwise the password may appear in the log. Following change
4077 PH/42 below, there is no longer a chance of it appearing in a bounce
4080 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4082 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4083 of its addresses was ever tried. (Bugzilla bug #2.)
4085 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4086 the result incorrectly in the debug output. (It correctly added a newline
4087 to what was transported.)
4089 TF/01 Added $received_time.
4091 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4092 commented out examples of how to interface to a virus scanner and to
4093 SpamAssassin. Also added commented examples of av_scanner and
4094 spamd_address settings.
4096 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4097 and controls are allowed in which ACLs. There were a couple of minor
4098 errors. Some of the entries in the conditions table (which is a table of
4099 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4100 negation of where the condition IS allowed.
4102 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4104 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4105 header file does not have a version number, so I've had to invent a new
4106 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4107 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4108 radiusclient), but it was contributed by a Radius user.
4110 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4111 files or directories, for OpenSSL.
4113 PH/19 When an Exim process that is running as root has to create an Exim log
4114 file, it does so in a subprocess that runs as exim:exim so as to get the
4115 ownership right at creation (otherwise, other Exim processes might see
4116 the file with the wrong ownership). There was no test for failure of this
4117 fork() call, which would lead to the process getting stuck as it waited
4118 for a non-existent subprocess. Forks do occasionally fail when resources
4119 run out. I reviewed all the other calls to fork(); they all seem to check
4122 PH/20 When checking for unexpected SMTP input at connect time (before writing
4123 the banner), Exim was not dealing correctly with a non-positive return
4124 from the read() function. If the client had disconnected by this time,
4125 the result was a log entry for a synchronization error with an empty
4126 string after "input=" when read() returned zero. If read() returned -1
4127 (an event I could not check), uninitialized data bytes were printed.
4128 There were reports of junk text (parts of files, etc) appearing after
4131 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4133 PH/22 Added support for macro redefinition, and (re)definition in between
4134 driver and ACL definitions.
4136 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4137 forgetting to use the resulting value; it was using the unexpanded value.
4139 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4140 hadn't been configured. The fix is from Juergen Kreileder, who
4141 understands it better than I do:
4143 "Here's what I see happening with three configured cyrus_sasl
4144 authenticators configured (plain, login, cram-md5):
4146 On startup auth_cyrus_sasl_init() gets called for each of these.
4147 This means three calls to sasl_listmech() without a specified mech_list.
4148 => SASL tests which mechs of all available mechs actually work
4149 => three warnings about OTP not working
4150 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4152 With the patch, sasl_listmech() also gets called three times. But now
4153 SASL's mech_list option is set to the server_mech specified in the the
4154 authenticator. Or in other words, the answer from sasl_listmech()
4155 gets limited to just the mech you're testing for (which is different
4157 => the return list contains just 'plain' or 'login', 'cram-md5' or
4158 nothing depending on the value of ob->server_mech.
4160 I've just tested the patch: Authentication still works fine,
4161 unavailable mechs specified in the exim configuration are still
4162 caught, and the auth.log warnings about OTP are gone."
4164 PH/25 When debugging is enabled, the contents of the command line are added
4165 to the debugging output, even when log_selector=+arguments is not
4168 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4169 answer is "GNU", and only if the return is "GNU/something" is the answer
4172 PH/27 $acl_verify_message is now set immediately after the failure of a
4173 verification in an ACL, and so is available in subsequent modifiers. In
4174 particular, the message can be preserved by coding like this:
4176 warn !verify = sender
4177 set acl_m0 = $acl_verify_message
4179 Previously, $acl_verify_message was set only while expanding "message"
4180 and "log_message" when a very denied access.
4182 PH/28 Modified OS/os.c-Linux with
4184 -#ifndef OS_LOAD_AVERAGE
4185 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4187 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4188 nomenclature these days.)
4190 PH/29 Installed patch from the Sieve maintainer that adds the options
4191 sieve_useraddress and sieve_subaddress to the redirect router.
4193 PH/30 In these circumstances:
4194 . Two addresses routed to the same list of hosts;
4195 . First host does not offer TLS;
4196 . First host accepts first address;
4197 . First host gives temporary error to second address;
4198 . Second host offers TLS and a TLS session is established;
4199 . Second host accepts second address.
4200 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4201 and peerdn, if requested) that were in fact used only for the second
4204 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4205 attention to any local part prefix or suffix that was matched by the
4206 router that accepted the address. It now behaves in the same way as it
4207 does for delivery: the affixes are removed from the local part unless
4208 rcpt_include_affixes is set on the transport.
4210 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4211 timeout during the DATA phase of an incoming message.
4213 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4214 applied a patch sanctioned by the Sieve maintainer.
4216 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4217 the uid or gid is negative. A case of a negative gid caused this to be
4218 noticed. The fix allows for either to be negative.
4220 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4221 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4224 PH/36 The change PH/12 above was broken. Fixed it.
4226 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4227 the grounds that routing the same address twice would always produce the
4228 same answer. This might have been true once, but it is certainly no
4229 longer true now. Routing a child address may depend on the previous
4230 routing that produced that child. Some complicated redirection strategies
4231 went wrong when messages had multiple recipients, and made Exim's
4232 behaviour dependent on the order in which the addresses were given.
4234 I have moved the duplicate checking until after the routing is complete.
4235 Exim scans the addresses that are assigned to local and remote
4236 transports, and removes any duplicates. This means that more work will be
4237 done, as duplicates will always all be routed, but duplicates are
4238 presumably rare, so I don't expect this is of any significance.
4240 For deliveries to pipes, files, and autoreplies, the duplicate checking
4241 still happens during the routing process, since they are not going to be
4244 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4245 It corrects a timeout issue with spamd. This is Ian's comment: "The
4246 background is that sometimes spamd either never reads data from a
4247 connection it has accepted, or it never writes response data. The exiscan
4248 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4249 blindly assumes that writes won't block so it may never time out."
4251 PH/39 Allow G after quota size as well as K and M.
4253 PH/40 The value set for $authenticated_id in an authenticator may not contain
4254 binary zeroes or newlines because the value is written to log lines and
4255 to spool files. There was no check on this. Now the value is run through
4256 the string_printing() function so that such characters are converted to
4257 printable escape sequences.
4259 PH/41 $message_linecount is a new variable that contains the total number of
4260 lines in the message. Compare $body_linecount, which is the count for the
4263 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4264 bounce and delay warning messages, except in certain special cases, which
4267 (a) An SMTP error message from a remote host;
4268 (b) A message specified in a :fail: redirection;
4269 (c) A message specified in a "fail" command in a system filter;
4270 (d) A message specified in a FAIL return from the queryprogram router;
4271 (e) A message specified by the cannot_route_message router option.
4273 In these cases only, Exim does include the error details in bounce and
4274 warning messages. There are also a few cases where bland messages such
4275 as "unrouteable address" or "local delivery error" are given.
4277 PH/43 $value is now also set for the "else" part of a ${run expansion.
4279 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4280 being worked on, but at least Exim now implements the latest version to
4283 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4284 process to complete was treated as a delivery failure, a timeout while
4285 writing the message to the pipe was logged, but erroneously treated as a
4286 successful delivery. Such timeouts include transport filter timeouts. For
4287 consistency with the overall process timeout, these timeouts are now
4288 treated as errors, giving rise to delivery failures by default. However,
4289 there is now a new Boolean option for the pipe transport called
4290 timeout_defer, which, if set TRUE, converts the failures into defers for
4291 both kinds of timeout. A transport filter timeout is now identified in
4294 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4295 systems where "make" and "gmake" are different, calling "gmake" at top
4296 level broke things. I've arranged for the value of $(MAKE) to be passed
4297 from the Makefile to this script so that it can call the same version of
4301 A note about Exim versions 4.44 and 4.50
4302 ----------------------------------------
4304 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4305 changes of various kinds. As a consequence, a big documentation update was
4306 needed. This delayed the release for rather longer than seemed good, especially
4307 in the light of a couple of (minor) security issues. Therefore, the changes
4308 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4309 release. So 4.44 and 4.50 are in effect two different branches that both start
4312 I have left the 4.50 change log unchanged; it contains all the changes since
4313 4.43. The change log for 4.44 is below; many of its items are identical to
4314 those for 4.50. This seems to be the most sensible way to preserve the
4315 historical information.
4321 1. Minor wording change to the doc/README.SIEVE file.
4323 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4324 computation of the current number of files was incorrect.
4326 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4327 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4330 4. Give more explanation in the error message when the command for a transport
4331 filter fails to execute.
4333 5. There are several places where Exim runs a non-Exim command in a
4334 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4335 was being done only for the command run by the queryprogram router. It is
4336 now done for all such subprocesses. The other cases are: ${run, transport
4337 filters, and the commands run by the lmtp and pipe transports.
4339 6. Added CONFIGURE_GROUP build-time option.
4341 7. Some older OS have a limit of 256 on the maximum number of file
4342 descriptors. Exim was using setrlimit() to set 1000 as a large value
4343 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4344 systems. I've change it so that if it can't get 1000, it tries for 256.
4346 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4347 was an oversight, and furthermore, ever since the addition of extra
4348 controls (e.g. 4.43/32), the checks on when to allow different forms of
4349 "control" were broken. There should now be diagnostics for all cases when a
4350 control that does not make sense is encountered.
4352 9. Added the /retain_sender option to "control=submission".
4354 10. $recipients is now available in the predata ACL (oversight).
4356 11. Tidy the search cache before the fork to do a delivery from a message
4357 received from the command line. Otherwise the child will trigger a lookup
4358 failure and thereby defer the delivery if it tries to use (for example) a
4359 cached ldap connection that the parent has called unbind on.
4361 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4362 of $address_data from the recipient verification was clobbered by the
4363 sender verification.
4365 13. The value of address_data from a sender verification is now available in
4366 $sender_address_data in subsequent conditions in the ACL statement.
4368 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4370 15. Added a new option "connect=<time>" to callout options, to set a different
4373 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4374 was its contents. (It was OK if the option was not defined at all.)
4376 17. A "Completed" log line is now written for messages that are removed from
4377 the spool by the -Mrm option.
4379 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4380 information about exactly what failed.
4382 19. Added -dd to debug only the daemon process.
4384 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4385 handles timeouts, both on the server side and network timeouts. Renamed the
4386 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4388 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4389 It is now set to "smtps".
4391 22. $host_address is now set to the target address during the checking of
4392 ignore_target_hosts.
4394 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4395 being passed; this would have caused $sender_host_name to have been used if
4396 matching the list had actually called for a host name (not very likely,
4397 since this list is usually IP addresses). A host name is now passed as
4400 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4401 code that specifies a non-restarting handler (typically sigaction() in
4402 modern systems) in an attempt to fix a rare and obscure crash bug.
4404 25. Narrowed the window for a race in the daemon that could cause it to ignore
4405 SIGCHLD signals. This is not a major problem, because they are used only to
4406 wake it up if nothing else does.
4408 26. A malformed maildirsize file could cause Exim to calculate negative values
4409 for the mailbox size or file count. Odd effects could occur as a result.
4410 The maildirsize information is now recalculated if the size or filecount
4413 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4414 support for a long time. Removed HAVE_SYS_VFS_H.
4416 28. Installed the latest version of exipick from John Jetmore.
4418 29. In an address list, if the pattern was not a regular expression, an empty
4419 subject address (from a bounce message) matched only if the pattern was an
4420 empty string. Non-empty patterns were not even tested. This was the wrong
4421 because it is perfectly reasonable to use an empty address as part of a
4422 database query. An empty address is now tested by patterns that are
4423 lookups. However, all the other forms of pattern expect the subject to
4424 contain a local part and a domain, and therefore, for them, an empty
4425 address still always fails if the pattern is not itself empty.
4427 30. Exim went into a mad DNS loop when attempting to do a callout where the
4428 host was specified on an smtp transport, and looking it up yielded more
4429 than one IP address.
4431 31. Re-factored the code for checking spool and log partition space into a
4432 function that finds that data and another that does the check. The former
4433 is then used to implement four new variables: $spool_space, $log_space,
4434 $spool_inodes, and $log_inodes.
4436 32. The RFC2047 encoding function was originally intended for short strings
4437 such as real names; it was not keeping to the 75-character limit for
4438 encoded words that the RFC imposes. It now respects the limit, and
4439 generates multiple encoded words if necessary. To be on the safe side, I
4440 have increased the buffer size for the ${rfc2047: expansion operator from
4443 33. It is now permitted to omit both strings after an "if" condition; if the
4444 condition is true, the result is "true". As before, when the second string
4445 is omitted, a false condition yields an empty string. This makes it less
4446 cumbersome to write custom ACL and router conditions.
4448 34. Failure to deliver a bounce message always caused it to be frozen, even if
4449 there was an errors_to setting on the router. The errors_to setting is now
4452 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4453 canonical form (fully expanded) before being placed in
4454 $sender_host_address.
4456 36. The table in the code that translates DNS record types into text (T_A to
4457 "A" for instance) was missing entries for NS and CNAME. It is just possible
4458 that this could have caused confusion if both these types were looked up
4459 for the same domain, because the text type is used as part of Exim's
4460 per-process caching. But the chance of anyone hitting this buglet seems
4463 37. The dnsdb lookup has been extended in a number of ways.
4465 (1) There is a new type, "zns", which walks up the domain tree until it
4466 finds some nameserver records. It should be used with care.
4468 (2) There is a new type, "mxh", which is like "mx" except that it returns
4469 just the host names, not the priorities.
4471 (3) It is now possible to give a list of domains (or IP addresses) to be
4472 looked up. The behaviour when one of the lookups defers can be
4473 controlled by a keyword.
4475 (4) It is now possible to specify the separator character for use when
4476 multiple records are returned.
4478 38. The dnslists ACL condition has been extended: it is now possible to supply
4479 a list of IP addresses and/or domains to be looked up in a particular DNS
4482 39. Added log_selector=+queue_time_overall.
4484 40. When running the queue in the test harness, wait just a tad after forking a
4485 delivery process, to get repeatability of debugging output.
4487 41. Include certificate and key file names in error message when GnuTLS fails
4488 to set them up, because the GnuTLS error message doesn't include the name
4489 of the failing file when there is a problem reading it.
4491 42. Allow both -bf and -bF in the same test run.
4493 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4495 44. The "Exiscan patch" is now merged into the mainline Exim source.
4497 45. Sometimes the final signoff response after QUIT could fail to get
4498 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4499 < 0 before doing a fflush(). This bug looks as though it goes back to the
4500 introduction of TLS in release 3.20, but "sometimes" must have been rare
4501 because the tests only now provoked it.
4503 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4504 (this can affect the format of dates).
4506 47. exim_tidydb, when checking for the continued existence of a message for
4507 which it has found a message-specific retry record, was not finding
4508 messages that were in split spool directories. Consequently, it was
4509 deleting retry records that should have stayed in existence.
4511 48. Steve fixed some bugs in eximstats.
4513 49. The SPA authentication driver was not abandoning authentication and moving
4514 on to the next authenticator when an expansion was forced to fail,
4515 contradicting the general specification for all authenticators. Instead it
4516 was generating a temporary error. It now behaves as specified.
4518 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4519 (the order specifies the preference for clients). The order is now AES256,
4520 AES128, 3DES, ARCFOUR128.
4522 51. Small patch to Sieve code - explicitly set From: when generating an
4525 52. Exim crashed if a remote delivery caused a very long error message to be
4526 recorded - for instance if somebody sent an entire SpamAssassin report back
4527 as a large number of 550 error lines. This bug was coincidentally fixed by
4528 increasing the size of one of Exim's internal buffers (big_buffer) that
4529 happened as part of the Exiscan merge. However, to be on the safe side, I
4530 have made the code more robust (and fixed the comments that describe what
4533 53. Now that there can be additional text after "Completed" in log lines (if
4534 the queue_time_overall log selector is set), a one-byte patch to exigrep
4535 was needed to allow it to recognize "Completed" as not the last thing in
4538 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4539 patch that reportedly fixes this has been added. I am not expert enough to
4540 create a test for it. This is what the patch creator wrote:
4542 "I found a little strange behaviour of ldap code when working with
4543 Windows 2003 AD Domain, where users was placed in more than one
4544 Organization Units. When I tried to give exim partial DN, the exit code
4545 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4546 But simultaneously result of request was absolutely normal ldap result,
4547 so I produce this patch..."
4549 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4550 REFERENCE, so I have modified the code to exclude the patch when that macro
4553 55. Some experimental protocols are using DNS PTR records for new purposes. The
4554 keys for these records are domain names, not reversed IP addresses. The
4555 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4556 leaves it alone. Component reversal etc. now happens only for IP addresses.
4559 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4561 57. Double the size of the debug message buffer (to 2048) so that more of very
4562 long debug lines gets shown.
4564 58. The exicyclog utility now does better if the number of log files to keep
4565 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4567 59. Two changes related to the smtp_active_hostname option:
4569 (1) $smtp_active_hostname is now available as a variable.
4570 (2) The default for smtp_banner uses $smtp_active_hostname instead
4571 of $primary_hostname.
4573 60. The host_aton() function is supposed to be passed a string that is known
4574 to be a valid IP address. However, in the case of IPv6 addresses, it was
4575 not checking this. This is a hostage to fortune. Exim now panics and dies
4576 if the condition is not met. A case was found where this could be provoked
4577 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4578 components; fortuitously, this particular loophole had already been fixed
4579 by change 4.50/55 above.
4581 If there are any other similar loopholes, the new check in host_aton()
4582 itself should stop them being exploited. The report I received stated that
4583 data on the command line could provoke the exploit when Exim was running as
4584 exim, but did not say which command line option was involved. All I could
4585 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4586 running as the user.
4589 61. There was a buffer overflow vulnerability in the SPA authentication code
4590 (which came originally from the Samba project). I have added a test to the
4591 spa_base64_to_bits() function which I hope fixes it.
4594 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4595 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4597 63. The daemon start-up calls getloadavg() while still root for those OS that
4598 need the first call to be done as root, but it missed one case: when
4599 deliver_queue_load_max is set with deliver_drop_privilege. This is
4600 necessary for the benefit of the queue runner, because there is no re-exec
4601 when deliver_drop_privilege is set.
4603 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4604 This has been fixed.
4606 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4607 was in use, was not putting the data itself into the right store pool;
4608 consequently, it could be overwritten for a subsequent message in the same
4609 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4612 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4614 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4615 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4616 the function were treating the return as a boolean value, which happened to
4617 work because 0=false and not-0=true, but is not correct code.
4619 68. The host_aton() function was not handling scoped IPv6 addresses (those
4620 with, for example, "%eth0" on the end) correctly.
4622 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4623 negated items (that is, ~something) in unsigned ints. Some compilers
4624 apparently mutter when there is no cast.
4626 70. If an address verification called from an ACL failed, and did not produce a
4627 user-specific message (i.e. there was only a "system" message), nothing was
4628 put in $acl_verify_message. In this situation, it now puts the system
4631 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4632 session; change 4.31/43 added the unwanted input to the log line - except
4633 that it did not do this in the start of session case. It now does.
4635 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4636 This is wrong and can cause the other end to generate a synchronization
4637 error if it is another Exim or anything else that does the synchronization
4638 check. A QUIT command is no longer sent after a timeout.
4640 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4641 during host lookups.
4643 74. The defer_ok option of callout verification was not working if it was used
4644 when verifying addresses in header lines, that is, for this case:
4646 verify = header_sender/callout=defer_ok
4648 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4649 those file descriptors could be used for SMTP connections. If anything
4650 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4651 could be sent to the SMTP client, causing chaos. The daemon now opens
4652 stdin, stdout, and stderr to /dev/null when it puts itself into the
4655 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4656 log by default. The user can override this with suitable Perl magic.
4658 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4659 the log message when discard triggers, was not working for the DATA ACL or
4660 for the non-SMTP ACL.
4662 78. Error message wording change in sieve.c.
4664 79. If smtp_accept_max_per_host was set, the number of connections could be
4665 restricted to fewer than expected, because the daemon was trying to set up
4666 a new connection before checking whether the processes handling previous
4667 connections had finished. The check for completed processes is now done
4668 earlier. On busy systems, this bug wouldn't be noticed because something
4669 else would have woken the daemon, and it would have reaped the completed
4672 80. If a message was submitted locally by a user whose login name contained one
4673 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4674 It caused a spool format error. I have fixed the spool reading code. A
4675 related problem was that the "from" clause in the Received: line became
4676 illegal because of the space(s). It is now covered by ${quote_local_part.
4678 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4681 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4683 83. Patch from Sieve maintainer to fix unterminated string problem in
4684 "vacation" handling.
4686 84. Some minor changes to the Linux configuration files to help with other
4687 OS variants using glibc.
4689 85. One more patch for Sieve to update vacation handling to latest spec.
4692 ----------------------------------------------------
4693 See the note above about the 4.44 and 4.50 releases.
4694 ----------------------------------------------------
4700 1. Change 4.43/35 introduced a bug that caused file counts to be
4701 incorrectly computed when quota_filecount was set in an appendfile
4704 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4705 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4708 3. Give more explanation in the error message when the command for a transport
4709 filter fails to execute.
4711 4. There are several places where Exim runs a non-Exim command in a
4712 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4713 was being done only for the command run by the queryprogram router. It is
4714 now done for all such subprocesses. The other cases are: ${run, transport
4715 filters, and the commands run by the lmtp and pipe transports.
4717 5. Some older OS have a limit of 256 on the maximum number of file
4718 descriptors. Exim was using setrlimit() to set 1000 as a large value
4719 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4720 systems. I've change it so that if it can't get 1000, it tries for 256.
4722 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4723 was an oversight, and furthermore, ever since the addition of extra
4724 controls (e.g. 4.43/32), the checks on when to allow different forms of
4725 "control" were broken. There should now be diagnostics for all cases when a
4726 control that does not make sense is encountered.
4728 7. $recipients is now available in the predata ACL (oversight).
4730 8. Tidy the search cache before the fork to do a delivery from a message
4731 received from the command line. Otherwise the child will trigger a lookup
4732 failure and thereby defer the delivery if it tries to use (for example) a
4733 cached ldap connection that the parent has called unbind on.
4735 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4736 of $address_data from the recipient verification was clobbered by the
4737 sender verification.
4739 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4740 was its contents. (It was OK if the option was not defined at all.)
4742 11. A "Completed" log line is now written for messages that are removed from
4743 the spool by the -Mrm option.
4745 12. $host_address is now set to the target address during the checking of
4746 ignore_target_hosts.
4748 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4749 being passed; this would have caused $sender_host_name to have been used if
4750 matching the list had actually called for a host name (not very likely,
4751 since this list is usually IP addresses). A host name is now passed as
4754 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4755 code that specifies a non-restarting handler (typically sigaction() in
4756 modern systems) in an attempt to fix a rare and obscure crash bug.
4758 15. Narrowed the window for a race in the daemon that could cause it to ignore
4759 SIGCHLD signals. This is not a major problem, because they are used only to
4760 wake it up if nothing else does.
4762 16. A malformed maildirsize file could cause Exim to calculate negative values
4763 for the mailbox size or file count. Odd effects could occur as a result.
4764 The maildirsize information is now recalculated if the size or filecount
4767 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4768 support for a long time. Removed HAVE_SYS_VFS_H.
4770 18. Updated exipick to current release from John Jetmore.
4772 19. Allow an empty sender to be matched against a lookup in an address list.
4773 Previously the only cases considered were a regular expression, or an
4776 20. Exim went into a mad DNS lookup loop when doing a callout where the
4777 host was specified on the transport, if the DNS lookup yielded more than
4780 21. The RFC2047 encoding function was originally intended for short strings
4781 such as real names; it was not keeping to the 75-character limit for
4782 encoded words that the RFC imposes. It now respects the limit, and
4783 generates multiple encoded words if necessary. To be on the safe side, I
4784 have increased the buffer size for the ${rfc2047: expansion operator from
4787 22. Failure to deliver a bounce message always caused it to be frozen, even if
4788 there was an errors_to setting on the router. The errors_to setting is now
4791 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4792 canonical form (fully expanded) before being placed in
4793 $sender_host_address.
4795 24. Updated eximstats to version 1.33
4797 25. Include certificate and key file names in error message when GnuTLS fails
4798 to set them up, because the GnuTLS error message doesn't include the name
4799 of the failing file when there is a problem reading it.
4801 26. Expand error message when OpenSSL has problems setting up cert/key files.
4804 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4805 (this can affect the format of dates).
4807 28. exim_tidydb, when checking for the continued existence of a message for
4808 which it has found a message-specific retry record, was not finding
4809 messages that were in split spool directories. Consequently, it was
4810 deleting retry records that should have stayed in existence.
4812 29. eximstats updated to version 1.35
4813 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4814 1.35 - bugfix such that pie charts by volume are generated correctly
4816 30. The SPA authentication driver was not abandoning authentication and moving
4817 on to the next authenticator when an expansion was forced to fail,
4818 contradicting the general specification for all authenticators. Instead it
4819 was generating a temporary error. It now behaves as specified.
4821 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4822 (the order specifies the preference for clients). The order is now AES256,
4823 AES128, 3DES, ARCFOUR128.
4825 31. Small patch to Sieve code - explicitly set From: when generating an
4828 32. Exim crashed if a remote delivery caused a very long error message to be
4829 recorded - for instance if somebody sent an entire SpamAssassin report back
4830 as a large number of 550 error lines. This bug was coincidentally fixed by
4831 increasing the size of one of Exim's internal buffers (big_buffer) that
4832 happened as part of the Exiscan merge. However, to be on the safe side, I
4833 have made the code more robust (and fixed the comments that describe what
4836 33. Some experimental protocols are using DNS PTR records for new purposes. The
4837 keys for these records are domain names, not reversed IP addresses. The
4838 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4839 leaves it alone. Component reversal etc. now happens only for IP addresses.
4842 34. The host_aton() function is supposed to be passed a string that is known
4843 to be a valid IP address. However, in the case of IPv6 addresses, it was
4844 not checking this. This is a hostage to fortune. Exim now panics and dies
4845 if the condition is not met. A case was found where this could be provoked
4846 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4847 components; fortuitously, this particular loophole had already been fixed
4848 by change 4.50/55 or 4.44/33 above.
4850 If there are any other similar loopholes, the new check in host_aton()
4851 itself should stop them being exploited. The report I received stated that
4852 data on the command line could provoke the exploit when Exim was running as
4853 exim, but did not say which command line option was involved. All I could
4854 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4855 running as the user.
4858 35. There was a buffer overflow vulnerability in the SPA authentication code
4859 (which came originally from the Samba project). I have added a test to the
4860 spa_base64_to_bits() function which I hope fixes it.
4863 36. The daemon start-up calls getloadavg() while still root for those OS that
4864 need the first call to be done as root, but it missed one case: when
4865 deliver_queue_load_max is set with deliver_drop_privilege. This is
4866 necessary for the benefit of the queue runner, because there is no re-exec
4867 when deliver_drop_privilege is set.
4869 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4870 was in use, was not putting the data itself into the right store pool;
4871 consequently, it could be overwritten for a subsequent message in the same
4872 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4875 38. Sometimes the final signoff response after QUIT could fail to get
4876 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4877 < 0 before doing a fflush(). This bug looks as though it goes back to the
4878 introduction of TLS in release 3.20, but "sometimes" must have been rare
4879 because the tests only now provoked it.
4885 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4886 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4887 Now it returns an integer. A number of calls were still expecting a T/F
4888 return. Fortuitously, in all cases, the tests worked in OK situations,
4889 which is the norm. However, things would have gone wrong on any write
4890 failures on the smtp file descriptor. This function is used when sending
4891 messages over SMTP and also when doing verify callouts.
4893 2. When Exim is called to do synchronous delivery of a locally submitted
4894 message (the -odf or -odi options), it no longer closes stderr before doing
4897 3. Implemented the mua_wrapper option.
4899 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4901 5. Implemented the functions header_remove(), header_testname(),
4902 header_add_at_position(), and receive_remove_recipient(), and exported them
4905 6. If an ACL "warn" statement specified the addition of headers, Exim already
4906 inserted X-ACL-Warn: at the start if there was no header name. However, it
4907 was not making this test for the second and subsequent header lines if
4908 there were newlines in the string. This meant that an invalid header could
4909 be inserted if Exim was badly configured.
4911 7. Allow an ACL "warn" statement to add header lines at the start or after all
4912 the Received: headers, as well as at the end.
4914 8. Added the rcpt_4xx retry error code.
4916 9. Added postmaster_mailfrom=xxx to callout verification option.
4918 10. Added mailfrom=xxxx to the callout verification option, for verify=
4921 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4922 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4924 12. Inserted some casts to stop certain compilers warning when using pointer
4925 differences as field lengths or precisions in printf-type calls (mostly
4926 affecting debugging statements).
4928 13. Added optional readline() support for -be (dynamically loaded).
4930 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4931 same clock tick as a message's arrival, so that its received time was the
4932 same as the "first fail" time on the retry record, and that message
4933 remained on the queue past the ultimate address timeout, every queue runner
4934 would try a delivery (because it was past the ultimate address timeout) but
4935 after another failure, the ultimate address timeout, which should have then
4936 bounced the address, did not kick in. This was a "< instead of <=" error;
4937 in most cases the first failure would have been in the next clock tick
4938 after the received time, and all would be well.
4940 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4941 being recognized when the domain list was tested by the match_domain
4942 condition in an expansion string.
4944 16. Added the ${str2b64: operator.
4946 17. Exim was always calling setrlimit() to set a large limit for the number of
4947 processes, without checking whether the existing limit was already
4948 adequate. (It did check for the limit on file descriptors.) Furthermore,
4949 errors from getrlimit() and setrlimit() were being ignored. Now they are
4950 logged to the main and panic logs, but Exim does carry on, to try to do its
4951 job under whatever limits there are.
4953 18. Imported PCRE 5.0.
4955 19. Trivial typo in log message " temporarily refused connection" (the leading
4958 20. If the log selector return_path_on_delivery was set and an address was
4959 redirected to /dev/null, the delivery process crashed because it assumed
4960 that a return path would always be set for a "successful" delivery. In this
4961 case, the whole delivery is bypassed as an optimization, and therefore no
4964 21. Internal re-arrangement: the function for sending a challenge and reading
4965 a response while authentication was assuming a zero-terminated challenge
4966 string. It's now changed to take a pointer and a length, to allow for
4967 binary data in such strings.
4969 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4971 23. Exim was not respecting finduser_retries when seeking the login of the
4972 uid under which it was called; it was always trying 10 times. (The default
4973 setting of finduser_retries is zero.) Also, it was sleeping after the final
4974 failure, which is pointless.
4976 24. Implemented tls_on_connect_ports.
4978 25. Implemented acl_smtp_predata.
4980 26. If the domain in control=submission is set empty, Exim assumes that the
4981 authenticated id is a complete email address when it generates From: or
4982 Sender: header lines.
4984 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4985 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4986 chown and chgrp in /bin and hostname in /usr/bin.
4988 28. Exim was keeping the "process log" file open after each use, just as it
4989 does for the main log. This opens the possibility of it remaining open for
4990 long periods when the USR1 signal hits a daemon. Occasional processlog
4991 errors were reported, that could have been caused by this. Anyway, it seems
4992 much more sensible not to leave this file open at all, so that is what now
4995 29. The long-running daemon process does not normally write to the log once it
4996 has entered its main loop, and it closes the log before doing so. This is
4997 so that log files can straightforwardly be renamed and moved. However,
4998 there are a couple of unusual error situations where the daemon does write
4999 log entries, and I had neglected to close the log afterwards.
5001 30. The text of an SMTP error response that was received during a remote
5002 delivery was being truncated at 512 bytes. This is too short for some of
5003 the long messages that one sometimes sees. I've increased the limit to
5006 31. It is now possible to make retry rules that apply only when a message has a
5007 specific sender, in particular, an empty sender.
5009 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5010 it possible to be selective about when SMTP synchronization is enforced.
5012 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5014 32. Implemented hosts_connection_nolog.
5016 33. Added an ACL for QUIT.
5018 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5021 35. Added mailbox_size and mailbox_filecount to appendfile.
5023 36. Added control = no_multiline_responses to ACLs.
5025 37. There was a bug in the logic of the code that waits for the clock to tick
5026 in the case where the clock went backwards by a substantial amount such
5027 that the microsecond fraction of "now" was more than the microsecond
5028 fraction of "then" (but the whole seconds number was less).
5030 38. Added support for the libradius Radius client library this is found on
5031 FreeBSD (previously only the radiusclient library was supported).
5037 1. When certain lookups returned multiple values in the form name=value, the
5038 quoting of the values was not always being done properly. Specifically:
5039 (a) If the value started with a double quote, but contained no whitespace,
5041 (b) If the value contained whitespace other than a space character (i.e.
5042 tabs or newlines or carriage returns) it was not quoted.
5043 This fix has been applied to the mysql and pgsql lookups by writing a
5044 separate quoting function and calling it from the lookup code. The fix
5045 should probably also be applied to nisplus, ibase and oracle lookups, but
5046 since I cannot test any of those, I have not disturbed their existing code.
5048 2. A hit in the callout cache for a specific address caused a log line with no
5049 reason for rejecting RCPT. Now it says "Previous (cached) callout
5050 verification failure".
5052 3. There was an off-by-one bug in the queryprogram router. An over-long
5053 return line was truncated at 256 instead of 255 characters, thereby
5054 overflowing its buffer with the terminating zero. As well as fixing this, I
5055 have increased the buffer size to 1024 (and made a note to document this).
5057 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5058 when Exim is waiting for an SMTP response from a remote server, Exim
5059 restarts its select() call on the socket, thereby resetting its timeout.
5060 This is not a problem when such interrupts are rare. Somebody set up a cron
5061 job to run exiwhat every 2 minutes, which is less than the normal select()
5062 timeout (5 or 10 minutes). This meant that the select() timeout never
5063 kicked in because it was always reset. I have fixed this by comparing the
5064 time when an interrupt arrives with the time at the start of the first call
5065 to select(). If more time than the timeout has elapsed, the interrupt is
5066 treated as a timeout.
5068 5. Some internal re-factoring in preparation for the addition of Sieve
5069 extensions (by MH). In particular, the "personal" test is moved to a
5070 separate function, and given an option for scanning Cc: and Bcc: (which is
5071 not set for Exim filters).
5073 6. When Exim created an email address using the login of the caller as the
5074 local part (e.g. when creating a From: or Sender: header line), it was not
5075 quoting the local part when it contained special characters such as @.
5077 7. Installed new OpenBSD configuration files.
5079 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5080 try to make them clearer.
5082 9. Callout options, other than the timeout value, were being ignored when
5083 verifying sender addresses in header lines. For example, when using
5085 verify = header_sender/callout=no_cache
5087 the cache was (incorrectly) being used.
5089 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5090 only the Cygwin environment.
5092 11. When return_path_on_delivery was set as a log selector, if different remote
5093 addresses in the same message used different return paths and parallel
5094 remote delivery occurred, the wrong values would sometimes be logged.
5095 (Whenever a remote delivery process finished, the return path value from
5096 the most recently started remote delivery process was logged.)
5098 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5099 lines when AUTH and/or TLS are in use. This is the "received protocol"
5100 field. Exim used to use "asmtp" for authenticated SMTP, without any
5101 indication (in the protocol name) for TLS use. Now it follows the RFC and
5102 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5103 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5104 names appear in log lines as well as in Received: header lines.
5106 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5107 extensions, and comparison tests, and to fix some bugs.
5109 14. Changes to the "personal" filter test:
5111 (1) The test was buggy in that it was just doing the equivalent of
5112 "contains" tests on header lines. For example, if a user's address was
5113 anne@some.where, the "personal" test would incorrectly be true for
5115 To: susanne@some.where
5117 This test is now done by extracting each address from the header in turn,
5118 and checking the entire address. Other tests that are part of "personal"
5119 are now done using regular expressions (for example, to check local parts
5120 of addresses in From: header lines).
5122 (2) The list of non-personal local parts in From: addresses has been
5123 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5124 taken from the Sieve specification recommendations.
5126 (3) If the message contains any header line starting with "List-" it is
5127 treated as non-personal.
5129 (4) The test for "circular" in the Subject: header line has been removed
5130 because it now seems ill-conceived.
5132 15. Minor typos in src/EDITME comments corrected.
5134 16. Installed latest exipick from John Jetmore.
5136 17. If headers_add on a router specified a text string that was too long for
5137 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5138 of string_sprintf() is now avoided.
5140 18. $message_body_size was not set (it was always zero) when running the DATA
5141 ACL and the local_scan() function.
5143 19. For the "mail" command in an Exim filter, no default was being set for
5144 the once_repeat time, causing a random time value to be used if "once" was
5145 specified. (If the value happened to be <= 0, no repeat happened.) The
5146 default is now 0s, meaning "never repeat". The "vacation" command was OK
5147 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5148 (I found it when inspecting the code).
5150 20. There is now an overall timeout for performing a callout verification. It
5151 defaults to 4 times the callout timeout, which applies to individual SMTP
5152 commands during the callout. The overall timeout applies when there is more
5153 than one host that can be tried. The timeout is checked before trying the
5154 next host. This prevents very long delays if there are a large number of
5155 hosts and all are timing out (e.g. when the network connections are timing
5156 out). The value of the overall timeout can be changed by specifying an
5157 additional sub-option for "callout", called "maxwait". For example:
5159 verify = sender/callout=5s,maxwait=20s
5161 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5162 to the end before writing, but this should make it even safer).
5164 22. Exim was forgetting that it had advertised PIPELINING for the second and
5165 subsequent messages on an SMTP connection. It was also not resetting its
5166 memory on STARTTLS and an internal HELO.
5168 23. When Exim logs an SMTP synchronization error within a session, it now
5169 records whether PIPELINING has been advertised or not.
5171 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5172 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5173 rather than long int.
5175 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5177 26. Added the never_mail option to autoreply.
5183 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5184 crash if the getsockname() call failed; this can happen if a connection is
5185 closed very soon after it is established. The problem was simply in the
5186 order in which certain operations were done, causing Exim to try to write
5187 to the SMTP stream before it had set up the file descriptor. The bug has
5188 been fixed by making things happen in the correct order.
5194 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5195 before the connection was closed, thus losing the rejection response.
5197 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5198 some early Solaris releases, but causes trouble in current releases where
5199 socklen_t is defined.
5201 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5204 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5207 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5208 that releasing the top of it at the end releases what was used for sub-
5209 expansions (unless the block got too big). However, discard this block if
5210 the first thing is a variable or header, so that we can use its block when
5211 it is dynamic (useful for very large $message_headers, for example).
5213 6. Lookups now cache *every* query, not just the most recent. A new, separate
5214 store pool is used for this. It can be recovered when all lookup caches are
5215 flushed. Lookups now release memory at the end of their result strings.
5216 This has involved some general refactoring of the lookup sources.
5218 7. Some code has been added to the store_xxx() functions to reduce the amount
5219 of flapping under certain conditions.
5221 8. log_incoming_interface used to affect only the <= reception log lines. Now
5222 it causes the local interface and port to be added to several more SMTP log
5223 lines, for example "SMTP connection from", and rejection lines.
5225 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5227 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5229 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5230 could be overwritten at the end of the current message (or the start of a
5231 new message if it was set in a HELO ACL). The value is now preserved for
5232 the duration of the SMTP connection.
5234 12. If a transport had a headers_rewrite setting, and a matching header line
5235 contained an unqualified address, that address was qualified, even if it
5236 did not match any rewriting rules. The underlying bug was that the values
5237 of the flags that permit the existence of unqualified sender and recipient
5238 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5239 non-local messages, and by -bnq for local messages) were not being
5240 preserved with the message after it was received.
5242 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5243 "next input=" as part of the text comprising the host identity instead of
5244 the correct text. The code was using the same buffer for two different
5245 strings. However, depending on which order the printing function evaluated
5246 its arguments, the bug did not always show up. Under Linux, for example, my
5247 test suite worked just fine.
5249 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5250 doesn't work with some older versions of Perl. It has been changed to "my",
5251 which in any case is probably the better facility to use.
5253 15. A really picky compiler found some instances of statements for creating
5254 error messages that either had too many or two few arguments for the format
5257 16. The size of the buffer for calls to the DNS resolver has been increased
5258 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5259 for addresses that have a lot of PTR records. This alleviates a problem; it
5260 does not fully solve it.
5262 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5263 buffer now truncates the list and logs the incident, which is the same
5264 action as happens when Exim is looking up a host name and its aliases.
5265 Previously in this situation something unpredictable would happen;
5266 sometimes it was "internal error: store_reset failed".
5268 18. If a server dropped the connection unexpectedly when an Exim client was
5269 using GnuTLS and trying to read a response, the client delivery process
5270 crashed while trying to generate an error log message.
5272 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5273 string, for example:
5275 warn message = H1: something\nH2: something
5277 the text was added as a single header line from Exim's point of view
5278 though it ended up OK in the delivered message. However, searching for the
5279 second and subsequent header lines using $h_h2: did not work. This has been
5280 fixed. Similarly, if a system filter added multiple headers in this way,
5281 the routers could not see them.
5283 20. Expanded the error message when iplsearch is called with an invalid key to
5284 suggest using net-iplsearch in a host list.
5286 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5287 ACLs are no longer actually imposed (and a message to that effect is
5290 22. If a "gecos" field in a passwd entry contained escaped characters, in
5291 particular, if it contained a \" sequence, Exim got it wrong when building
5292 a From: or a Sender: header from that name. A second bug also caused
5293 incorrect handling when an unquoted " was present following a character
5294 that needed quoting.
5296 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5297 was not being matched caselessly.
5299 24. Arranged for all hyphens in the exim.8 source to be escaped with
5302 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5303 properly. Recipient callout cache records were still being keyed to include
5304 the sender, even when use_sender was set false. This led to far more
5305 callouts that were necessary. The sender is no longer included in the key
5306 when use_sender is false.
5308 26. Added "control = submission" modifier to ACLs.
5310 27. Added the ${base62d: operator to decode base 62 numbers.
5312 28. dnsdb lookups can now access SRV records.
5314 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5315 the configuration file.
5317 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5318 (-v) mode. This makes the output for a verbose queue run more intelligible.
5320 31. Added a use_postmaster feature to recipient callouts.
5322 32. Added the $body_zerocount variable, containing the number of binary zero
5323 bytes in the message body.
5325 33. The time of last modification of the "new" subdirectory is now used as the
5326 "mailbox time last read" when there is a quota error for a maildir
5329 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5331 35. Added +ignore_unknown as a special item in host lists.
5333 36. Code for decoding IPv6 addresses in host lists is now included, even if
5334 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5335 address was recognized as an IP address, but was then not correctly decoded
5336 into binary, causing unexpected and incorrect effects when compared with
5343 1. Very minor rewording of debugging text in manualroute to say "list of
5344 hosts" instead of "hostlist".
5346 2. If verify=header_syntax was set, and a header line with an unqualified
5347 address (no domain) and a large number of spaces between the end of the
5348 name and the colon was received, the reception process suffered a buffer
5349 overflow, and (when I tested it) crashed. This was caused by some obsolete
5350 code that should have been removed. The fix is to remove it!
5352 3. When running in the test harness, delay a bit after writing a bounce
5353 message to get a bit more predictability in the log output.
5355 4. Added a call to search_tidyup() just before forking a reception process. In
5356 theory, someone could use a lookup in the expansion of smtp_accept_max_
5357 per_host which, without the tidyup, could leave open a database connection.
5359 5. Added the variables $recipient_data and $sender_data which get set from a
5360 lookup success in an ACL "recipients" or "senders" condition, or a router
5361 "senders" option, similar to $domain_data and $local_part_data.
5363 6. Moved the writing of debug_print from before to after the "senders" test
5366 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5367 problems for message scanning, either using a data ACL, or using
5368 local_scan() because the Received: header was not generated till after they
5369 were called (in order to set the time as the time of reception completion).
5370 I have revised the way this works. The header is now generated after the
5371 body is received, but before the ACL or local_scan() are called. After they
5372 are run, the timestamp in the header is updated.
5378 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5379 before starting a queue runner without re-exec. This happened only when
5380 deliver_drop_privilege was set or when the Exim user was set to root. The
5381 effect of the bug was that timeouts during subsequent deliveries caused
5382 crashes instead of being properly handled. The handler is now left at its
5383 default (and expected) setting.
5385 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5386 message, again when deliver_drop_privilege is set or Exim is run as root.
5387 The bug described in (1) was not present in this case, but the tidying up
5388 of the other signals was missing. I have made the two cases consistent.
5390 3. The ignore_target_hosts setting on a manualroute router was being ignored
5391 for hosts that were looked up using the /MX notation.
5393 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5396 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5397 operated on the sender address. After changing the $sender_address to <>
5398 for the sender address verify, Exim was re-instated it as the original
5399 (before rewriting) address, but remembering that it had rewritten it, so it
5400 wasn't rewriting it again. This bug also had the effect of breaking the
5401 sender address verification caching when the sender address was rewritten.
5403 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5404 This has been changed so that if the ip literal address matches
5405 ignore_target_hosts, the router declines.
5407 7. Added expansion conditions match_domain, match_address, and match_local_
5408 part (NOT match_host).
5410 8. The placeholder for the Received: header didn't have a length field set.
5412 9. Added code to Exim itself and to exim_lock to test for a specific race
5413 condition that could lead to file corruption when using MBX delivery. The
5414 issue is with the lockfile that is created in /tmp. If this file is removed
5415 after a process has opened it but before that process has acquired a lock,
5416 there is the potential for a second process to recreate the file and also
5417 acquire a lock. This could lead to two Exim processes writing to the file
5418 at the same time. The added code performs the same test as UW imapd; it
5419 checks after acquiring the lock that its file descriptor still refers to
5420 the same named file.
5422 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5423 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5426 11. Added the smtp_active_hostname option. If used, this will typically be made
5427 to depend on the incoming interface address. Because $interface_address is
5428 not set up until the daemon has forked a reception process, error responses
5429 that can happen earlier (such as "too many connections") no longer contain
5432 12. If an expansion in a condition on a "warn" statement fails because a lookup
5433 defers, the "warn" statement is abandoned, and the next ACL statement is
5434 processed. Previously this caused the whole ACL to be aborted.
5436 13. Added the iplsearch lookup type.
5438 14. Added ident_timeout as a log selector.
5440 15. Added tls_certificate_verified as a log selector.
5442 16. Added a global option tls_require_ciphers (compare the smtp transport
5443 option of the same name). This controls incoming TLS connections.
5445 17. I finally figured out how to make tls_require_ciphers do a similar thing
5446 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5447 before starting the TLS session.
5449 18. Tabs are now shown as \t in -bP output.
5451 19. If the log selector return_path_on_delivery was set, Exim crashed when
5452 bouncing a message because it had too many Received: header lines.
5454 20. If two routers both had headers_remove settings, and the first one included
5455 a superfluous trailing colon, the final name in the first list and the
5456 first name in the second list were incorrectly joined into one item (with a
5457 colon in the middle).
5463 1. Added -C and -D options to the exinext utility, mainly to make it easier
5464 to include in the automated testing, but these could be helpful when
5465 multiple configurations are in use.
5467 2. The exinext utility was not formatting the output nicely when there was
5468 an alternate port involved in the retry record key, nor when there was a
5469 message id as well (for retries that were specific to a specific message
5470 and a specific host). It was also confused by IPv6 addresses, because of
5471 the additional colons they contain. I have fixed the IPv4 problem, and
5472 patched it up to do a reasonable job for IPv6.
5474 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5475 delivery, the log line now contains "pipelined" if PIPELINING was used.
5477 4. An SMTP transport process used to panic and die if the bind() call to set
5478 an explicit outgoing interface failed. This has been changed; it is now
5479 treated in the same way as a connect() failure.
5481 5. A reference to $sender_host_name in the part of a conditional expansion
5482 that was being skipped was still causing a DNS lookup. This no longer
5485 6. The def: expansion condition was not recognizing references to header lines
5486 that used bh_ and bheader_.
5488 7. Added the _cache feature to named lists.
5490 8. The code for checking quota_filecount in the appendfile transport was
5491 allowing one more file than it should have been.
5493 9. For compatibility with Sendmail, the command line option
5501 and sets the incoming protocol and host name (for trusted callers). The
5502 host name and its colon can be omitted when only the protocol is to be set.
5503 Note the Exim already has two private options, -pd and -ps, that refer to
5504 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5505 "s", but I don't think that's a major issue.
5507 10. A number of refactoring changes to the code, none of which should affect
5510 (a) The number of logging options was getting close to filling up the
5511 32-bit word that was used as a bit map. I have split them into two classes:
5512 those that are passed in the argument to log_write(), and those that are
5513 only ever tested independently outside of that function. These are now in
5514 separate 32-bit words, so there is plenty of room for expansion again.
5515 There is no change in the user interface or the logging behaviour.
5517 (b) When building, for example, log lines, the code previously used a
5518 macro that called string_cat() twice, in order to add two strings. This is
5519 not really sufficiently general. Furthermore, there was one instance where
5520 it was actually wrong because one of the argument was used twice, and in
5521 one call a function was used. (As it happened, calling the function twice
5522 did not affect the overall behaviour.) The macro has been replaced by a
5523 function that can join an arbitrary number of extra strings onto a growing
5526 (c) The code for expansion conditions now uses a table and a binary chop
5527 instead of a serial search (which was left over from when there were very
5528 few conditions). Also, it now recognizes conditions like "pam" even when
5529 the relevant support is not compiled in: a suitably worded error message is
5530 given if an attempt is made to use such a condition.
5532 11. Added ${time_interval:xxxxx}.
5534 12. A bug was causing one of the ddress fields not to be passed back correctly
5535 from remote delivery subprocesses. The field in question was not being
5536 subsequently used, so this caused to problems in practice.
5538 13. Added new log selectors queue_time and deliver_time.
5540 14. Might have fixed a bug in maildirsizefile handling that threw up
5541 "unexpected character" debug warnings, and recalculated the data
5542 unnecessarily. In any case, I expanded the warning message to give more
5545 15. Added the message "Restricted characters in address" to the statements in
5546 the default ACL that block characters like @ and % in local parts.
5548 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5549 Three changes have been made:
5551 (a) There was a serious bug; a negative response to MAIL caused the whole
5552 recipient domain to be cached as invalid, thereby blocking all messages
5553 to all local parts at the same domain, from all senders. This bug has
5554 been fixed. The domain is no longer cached after a negative response to
5555 MAIL if the sender used is not empty.
5557 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5560 (c) A new callout option, "use_sender" has been added for people who want
5561 the modified behaviour.
5567 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5570 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5571 indeed breaks things for older releases.
5573 3. Added additional logging to the case where there is a problem reading data
5574 from a filter that is running in a subprocess using a pipe, in order to
5575 try to track down a specific problem.
5577 4. Testing facility fudge: when running in the test harness and attempting
5578 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5579 getting "No route to host". Convert this to a timeout.
5581 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5584 6. Some OS don't have socklen_t but use size_t instead. This affects the
5585 fifth argument of getsockopt() amongst other things. This is now
5586 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5587 can be set for individual OS. I have set it for SunOS5, OSF1, and
5588 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5589 some earlier ones do not.
5591 7. Change 4.30/15 was not doing the test caselessly.
5593 8. The standard form for an IPv6 address literal was being rejected by address
5594 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5595 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5596 this, as well as the form without the "IPv6" on the front (but only when
5597 address literals are enabled, of course).
5599 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5601 10. Exim crashed if a message with an empty sender address specified by -f
5602 encountered a router with an errors_to setting. This could be provoked only
5603 by a command such as
5607 where an empty string was supplied; "<>" did not hit this bug.
5609 11. Installed PCRE release 4.5.
5611 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5612 remained set. It is now erased.
5614 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5615 times from message ids (which are base 36 rather than the normal 62).
5617 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5618 were being counted as actual protocol errors, and logged if the log
5619 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5620 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5621 and DATA following a set of rejected RCPTs do not count as protocol errors.
5622 In other words, Exim assumes they were pipelined, though this may not
5623 actually be the case. Of course, in all cases the client gets an
5624 appropriate error code.
5626 15. If a lookup fails in an ACL condition, a message about the failure may
5627 be available; it is used if testing the ACL cannot continue, because most
5628 such messages specify what the cause of the deferral is. However, some
5629 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5630 that caused an old message to be retained and used if a later statement
5631 caused a defer, replacing the real cause of the deferral.
5633 16. If an IP address had so many PTR records that the DNS lookup buffer
5634 was not large enough to hold them, Exim could crash while trying to process
5635 the truncated data. It now detects and logs this case.
5637 17. Further to 4.21/58, another change has been made: if (and only if) the
5638 first line of a message (the first header line) ends with CRLF, a bare LF
5639 in a subsequent header line has a space inserted after it, so as not to
5640 terminate the header.
5642 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5643 unnecessarily, used atoi() instead of strtol(), and didn't check the
5644 termination when getting file sizes from file names by regex.
5646 19. Completely re-implemented the support for maildirsize files, in the light
5647 of a number of problems with the previous contributed implementation
5648 (4.30/29). In particular:
5650 . If the quota is zero, the maildirsize file is maintained, but no quota is
5653 . If the maildir directory does not exist, it is created before any attempt
5654 to write a maildirsize file.
5656 . The quota value in the file is just a cache; if the quota is changed in
5657 the transport, the new value overrides.
5659 . A regular expression is available for excluding directories from the
5662 20. The autoreply transport checks the characters in options that define the
5663 message's headers; it allows continued headers, but it was checking with
5664 isspace() after an embedded newline instead of explicitly looking for a
5667 21. If all the "regular" hosts to which an address was routed had passed their
5668 expiry times, and had not reached their retry times, the address was
5669 bounced, even if fallback hosts were defined. Now Exim should go on to try
5672 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5673 equivalent code in the SMTP transport. Some hosts send humungous responses
5674 to HELO/EHLO, more than 1024 it seems.
5676 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5677 gives compiler warnings in some environments. I've now done it "properly",
5680 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5681 (because of problems with the built-in one) was declared to return uschar *
5682 instead of char *, causing compiler failure.
5684 25. Fixed a file descriptor leak when processing alias/forward files.
5686 26. Fixed a minor format string issue in dbfn.c.
5688 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5690 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5691 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5694 29. When a sender address is verified, it is cached, to save repeating the test
5695 when there is more than one recipient in a message. However, when the
5696 verification involves a callout, it is possible for different callout
5697 options to be set for different recipients. It is too complicated to keep
5698 track of this in the cache, so now Exim always runs a verification when a
5699 callout is required, relying on the callout cache for the optimization.
5700 The overhead is duplication of the address routing, but this should not be
5703 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5704 to be verified with callout=postmaster, and the main callout worked but the
5705 postmaster check failed, the verification correctly failed. However, if a
5706 subsequent RCPT command asked for sender verification *without* the
5707 postmaster check, incorrect caching caused this verification also to fail,
5710 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5711 it was not caching the DNS options (qualify_single, search_parents) that
5712 were used when the lookup failed. A subsequent lookup with different
5713 options therefore always gave the same answer, though there were cases
5714 where it should not have. (Example: a "domains = !$mx_any" option on a
5715 dnslookup router: the "domains" option is always processed without any
5716 widening, but the router might have qualify_single set.) Now Exim uses the
5717 cached value only when the same options are set.
5719 32. Added John Jetmore's "exipick" utility to the distribution.
5721 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5722 than a timeout (e.g. a certificate is required, and is not provided), an
5723 Exim server now closes the connection immediately. Previously it waited for
5724 the client to close - but if the client is SSL, it seems that they each
5725 wait for each other, leading to a delay before one of them times out.
5727 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5728 maintained 0.8.x compatibility because I don't think many are using it, and
5729 it is clearly obsolete.
5731 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5734 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5735 client certificate was expired. A simple patch fixes this, though I don't
5736 understand the full logic of why the verify callback is called multiple
5739 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5740 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5741 which causes problems with some clients (such as the Certicom SSL Plus
5742 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5743 disables the coutermeasure allowing Eudora to connect."
5745 38. Exim was not checking that a write() to a log file succeeded. This could
5746 lead to Bad Things if a log got too big, in particular if it hit a file
5747 size limit. Exim now panics and dies if it cannot write to a log file, just
5748 as it does if it cannot open a log file.
5750 39. Modified OS/Makefile-Linux so that it now contains
5752 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5754 The two -D definitions ensure that Exim is compiled with large file
5755 support, which makes it possible to handle log files that are bigger than
5758 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5759 instance) a domain was checked against a named list that involved a lookup,
5760 causing $domain_data to be set, then another domain was checked against the
5761 same list, then the first domain was re-checked, the value of $domain_data
5762 after the final check could be wrong. In particular, if the second check
5763 failed, it could be set empty. This bug probably also applied to
5766 41. The strip_trailing_dot option was not being applied to the address given
5767 with the -f command-line option.
5769 42. The code for reading a message's header from the spool was incrementing
5770 $received_count, but never initializing it. This meant that the value was
5771 incorrect (doubled) while delivering a message in the same process in which
5772 it was received. In the most common configuration of Exim, this never
5773 happens - a fresh exec is done - but it can happen when
5774 deliver_drop_privilege is set.
5776 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5777 it now includes up to 150 characters of the unexpected data in the log
5780 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5781 and building data strings. The size of both of these buffers was 10 000
5782 bytes - far larger than anybody would *ever* want, thought I. Needless to
5783 say, somebody hit the limit. I have increased the maximum line length to
5784 20 000 and the maximum data length of concatenated lines to 100 000. I have
5785 also fixed two bugs, because there was no checking on these buffers. Tsk,
5786 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5787 buffer is too small.
5789 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5790 lsearch lookups. Now it does.
5792 46. When parsing a route_list item in a manualroute router, a fixed-length
5793 buffer was used for the list of hosts. I made this 1024 bytes long,
5794 thinking that nobody would ever have a list of hosts that long. Wrong.
5795 Somebody had a whole pile of complicated expansion conditions, and the
5796 string was silently truncated, leading to an expansion error. It turns out
5797 that it is easier to change to an unlimited length (owing to other changes
5798 that have happened since this code was originally written) than to build
5799 structure for giving a limitation error. The length of the item that
5800 expands into the list of hosts is now unlimited.
5802 47. The lsearch lookup could not handle data where the length of text line was
5803 more than 4095 characters. Such lines were truncated, leading to shortened
5804 data being returned. It should now handle lines of any length.
5806 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5807 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5810 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5811 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5812 no longer happens in comment lines. A list of replacements is now placed
5813 at the head of all of the source files, except those whose only change is
5814 to replace PERL_COMMAND in the very first #! line.
5816 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5817 messages on the queue, with a bottom-up merge sort, using code contributed
5818 by Michael Haardt. This should make operations like -bp somewhat faster on
5819 large queues. It won't affect queue runners, except when queue_run_in_order
5822 51. Installed eximstats 1.31 in the distribution.
5824 52. Added support for SRV lookups to the dnslookup router.
5826 53. If an ACL referred to $message_body or $message_body_end, the value was not
5827 reset for any messages that followed in the same SMTP session.
5829 54. The store-handling optimization for building very long strings was not
5830 differentiating between the different store pools. I don't think this
5831 actually made any difference in practice, but I've tidied it.
5833 55. While running the routers to verify a sender address, $sender_address
5834 was still set to the sender address. This is wrong, because when routing to
5835 send a bounce to the sender, it would be empty. Therefore, I have changed
5836 it so that, while verifying a sender address, $sender_address is set to <>.
5837 (There is no change to what happens when verifying a recipient address.)
5839 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5840 target A or AAAA records (if not already returned) without resetting the
5841 qualify_single or search_parents options of the DNS resolver. These are
5842 inappropriate in this case because the targets of MX and SRV records must
5843 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5844 target that, when qualified, matched something in the local domain. These
5845 two options are now turned off when doing these lookups.
5847 57. It seems that at least some releases of Reiserfs (which does not have the
5848 concept of a fixed number of inodes) returns zero and not -1 for the
5849 number of available inodes. This interacted badly with check_spool_inodes,
5850 which assumed that -1 was the "no such thing" setting. What I have done is
5851 to check that the total number of inodes is greater than zero before doing
5852 the test of how many are available.
5854 58. When a "warn" ACL statement has a log_message modifier, the message is
5855 remembered, and not repeated. This is to avoid a lot of repetition when a
5856 message has many recipients that cause the same warning to be written.
5857 However, Exim was preserving the list of already written lines for an
5858 entire SMTP session, which doesn't seem right. The memory is now reset if a
5859 new message is started.
5861 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5862 debugging output unless log_rewrite was also set.
5864 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5865 of (int)(handle) when we know that handle contains (void *)(-1).
5867 61. The Exim daemon panic-logs an error return when it closes the incoming
5868 connection. However "connection reset by peer" seems to be common, and
5869 isn't really an error worthy of noting specially, so that particular error
5872 62. When Exim is trying to find all the local interfaces, it used to panic and
5873 die if the ioctl to get the interface flags failed. However, it seems that
5874 on at least one OS (Solaris 9) it is possible to have an interface that is
5875 included in the list of interfaces, but for which you get a failure error
5876 for this call. This happens when the interface is not "plumbed" into a
5877 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5878 failure of the "get flags" call assumes that the interface is down.
5880 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5881 makes life easier for people who are doing arithmetic on fields extracted
5882 from dates, where you often get leading zeros that should not be
5883 interpreted as octal.
5885 64. Added qualify_domain to the redirect router, to override the global
5888 65. If a pathologically long header line contained very many addresses (the
5889 report of this problem mentioned 10 000) and each of them was rewritten,
5890 Exim could use up a very large amount of memory. (It kept on making new
5891 copies of the header line as it rewrote, and never released the old ones.)
5892 At the expense of a bit more processing, the header rewriting function has
5893 been changed so that it no longer eats memory in this way.
5895 66. The generation of the Received: header has been moved from the time that a
5896 message starts to be received, to the time that it finishes. The timestamp
5897 in the Received: header should now be very close to that of the <= log
5898 line. There are two side-effects of this change:
5900 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5901 logged header lines no longer include the local Received: line, because
5902 it has not yet been created. The same applies to a copy of the message
5903 that is returned to a non-SMTP sender when a message is rejected.
5905 (b) When a filter file is tested using -bf, no additional Received: header
5906 is added to the test message. After some thought, I decided that this
5909 This change does not affect the value of $received_for. It is still set
5910 after address rewriting, but before local_scan() is called.
5912 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5914 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5915 gave an unhelpful panic error message, and a defer error. I have managed to
5916 change this behaviour so that it now rejects any supplied certificate,
5917 which seems right, as the list of acceptable certificates is empty.
5919 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5920 gave an unhelpful defer error. I have not managed to make this reject any
5921 supplied certificates, but the error message it gives is "no certificate
5922 supplied", which is not helpful.
5924 70. exigrep's output now also includes lines that are not associated with any
5925 message, but which match the given pattern. Implemented by a patch from
5926 Martin Sluka, which also tidied up the Perl a bit.
5928 71. Recipient callout verification, like sender verification, was using <> in
5929 the MAIL FROM command. This isn't really the right thing, since the actual
5930 sender may affect whether the remote host accepts the recipient or not. I
5931 have changed it to use the actual sender in the callout; this means that
5932 the cache record is now keyed on a recipient/sender pair, not just the
5933 recipient address. There doesn't seem to be a real danger of callout loops,
5934 since a callout by the remote host to check the sender would use <>.
5935 [SEE ABOVE: changed after hitting problems.]
5937 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5938 temporary errors. However, in the case of such a code being given after
5939 the end of a data transmission (i.e. after ".") Exim was failing to write
5940 a retry record for the message. (Yes, there was some broken host that was
5941 actually sending 8xx at this point.)
5943 73. An unknown lookup type in a host list could cause Exim to panic-die when
5944 the list was checked. (An example that provoked this was putting <; in the
5945 middle of a list instead of at the start.) If this happened during a DATA
5946 ACL check, a -D file could be left lying around. This kind of configuration
5947 error no longer causes Exim to die; instead it causes a defer error. The
5948 incident is still logged to the main and panic logs.
5950 74. Buglet left over from Exim 3 conversion. The message "too many messages
5951 in one connection" was written to the rejectlog but not the mainlog, except
5952 when address rewriting (yes!) was being logged.
5954 75. Added write_rejectlog option.
5956 76. When a system filter was run not as root (that is, when system_filter_user
5957 was set), the values of the $n variables were not being returned to the
5958 main process; thus, they were not subsequently available in the $sn
5961 77. Added +return_path_on_delivery log selector.
5963 78. A connection timeout was being treated differently from recipients deferred
5964 when testing hosts_max_try with a message that was older than the host's
5965 retry timeout. (The host should not be counted, thus allowing all hosts to
5966 be tried at least once before bouncing.) This may have been the cause of an
5967 occasionally reported bug whereby a message would remain on the queue
5968 longer than the retry timeout, but would be bounced if a delivery was
5969 forced. I say "may" because I never totally pinned down the problem;
5970 setting up timeout/retry tests is difficult. See also the next item.
5972 79. The ultimate address timeout was not being applied to errors that involved
5973 a combination of host plus message (for example, a timeout on a MAIL
5974 command). When an address resolved to a number of possible hosts, and they
5975 were not all tried for each delivery (e.g. because of hosts_max_try), a
5976 message could remain on the queue longer than the retry timeout.
5978 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5981 81. Fixed an obscure SMTP outgoing bug which required at least the following
5982 conditions: (a) there was another message waiting for the same server;
5983 (b) the server returned 5xx to all RCPT commands in the first message so
5984 that the message was not completed; (c) the server dropped the connection
5985 or gave a negative response to the RSET that Exim sends to abort the
5986 transaction. The observed case was a dropped connection after DATA that had
5987 been sent in pipelining mode. That is, the server had advertised PIPELINING
5988 but was not implementing it correctly. The effect of the bug was incorrect
5989 behaviour, such as trying another host, and this could lead to a crash.
5995 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5996 and daemon.c were passed as pointers to ints; they should have been
5997 pointers to socklen_t variables (which are typically unsigned ints).
5999 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6002 3. Fixed a really odd bug that affected only the testing scheme; patching a
6003 certain fixed string in the binary changed the value of another string that
6004 happened to be identical to the end of the original first string.
6006 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6007 name", it returns that address as the IP address. On some operating
6008 systems (e.g. Solaris), it also passes back the IP address string as the
6009 "host name". However, on others (e.g. Linux), it passes back an empty
6010 string. Exim wasn't checking for this, and was changing the host name to an
6011 empty string, assuming it had been canonicalized.
6013 5. Although rare, it is permitted to have more than one PTR record for a given
6014 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6015 all the names associated with an address, because they do in Solaris.
6016 However, it seems that they do not in Linux for data that comes from the
6017 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6018 I found this out when I moved to a new Linux workstation and tried to run
6019 the Exim test suite.
6021 To get round this problem I have changed the code so that it now does its
6022 own call to the DNS to look up PTR records when searching for a host name.
6023 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6024 addresses that are only in /etc/hosts are still found.
6026 This behaviour is, however, controlled by an option called host_lookup_
6027 order, which defaults to "bydns:byaddr". If people want to use the other
6028 order, or indeed, just use one or the other means of lookup, they can
6029 specify it in this variable.
6031 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6032 some operating systems, this comes back from gethostbyaddr() as an empty
6033 string, and this is what Exim used to test for. However, it seems that in
6034 other systems, "." is yielded. Exim now tests for this case too.
6036 7. The values of check_spool_space and check_log_space are now held internally
6037 as a number of kilobytes instead of an absolute number of bytes. If a
6038 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6039 kilobyte. This means that much larger values can be stored.
6041 8. Exim monitor: an attempt to get the action menu when not actually pointing
6042 at a message produces an empty menu entitled "No message selected". This
6043 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6044 no entries in it ("Shell widget menu has zero width and/or height"). So I
6045 have added a single, blank menu entry in this case.
6047 9. Added ${quote_local_part.
6049 10. MIME decoding is now applied to the contents of Subject: header lines when
6052 11. Now that a reference to $sender_host_address automatically causes a reverse
6053 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6054 host lookup before query-style lookups in lists that might use this
6055 variable. This has therefore been abolished, and the "net-" prefix is no
6056 longer necessary for query-style lookups.
6058 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6059 have been a typo for LFLAGS, so it has been changed.
6061 13. The install script calls Exim with "-C /dev/null" in order to find the
6062 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6063 to be output. However, since Exim outputs its version number before the
6064 error, it didn't break the script. It just looked ugly. I fixed this by
6065 always allowing "-C /dev/null" if the caller is root.
6067 14. Ignore overlarge ACL variable number when reading spool file - insurance
6068 against a later release with more variables having written the file.
6070 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6071 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6072 this, as well as the form without the "IPv6" on the front.
6074 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6075 OS/Makefile-Darwin file.
6077 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6078 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6080 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6081 variable. However, some OS use ints for inodes. I've added cast to long int
6082 to get rid of the compiler warning.
6084 19. I had forgotten to lock out "/../" in configuration file names when
6085 ALT_CONFIG_PREFIX was set.
6087 20. Routers used for verification do not need to specify transports. However,
6088 if such a router generated a host list, and callout was configured, Exim
6089 crashed, because it could not find a port number from the (non-existent)
6090 transport. It now assumes port 25 in this circumstance.
6092 21. Added the -t option to exigrep.
6094 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6095 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6096 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6097 them is set, LOOKUP_LSEARCH is forced.
6099 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6101 24. Added sender and host information to the "rejected by local_scan()" log
6102 line; previously there was no indication of these.
6104 25. Added .include_if_exists.
6106 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6107 Linux. It turns out that not all file systems support this. Apparently some
6108 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6109 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6110 not supported on the file descriptor, is now ignored when Exim is trying to
6111 sync a directory. This applies only to Linux.
6113 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6115 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6116 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6119 29. Added support for maildirsize files from supplied patch (modified a bit).
6121 30. The use of :fail: followed by an empty string could lead Exim to respond to
6122 sender verification failures with (e.g.):
6124 550 Verification failed for <xxx>
6125 550 Sender verify failed
6127 where the first response line was missing the '-' that indicates it is not
6128 the final line of the response.
6130 31. The loop for finding the name of the user that called Exim had a hardwired
6131 limit of 10; it now uses the value of finduser_retries, which is used for
6132 all other user lookups.
6134 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6137 33. Exim was neglecting to zero errno before one call of strtol() when
6138 expanding a string and expecting an integer value. On some systems this
6139 resulted in spurious "integer overflow" errors. Also, it was casting the
6140 result into an int without checking.
6142 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6143 did not work. The code looks as if it has *never* worked, though it appears
6144 to have been documented since at least release 1.62. I have made it work.
6146 35. The "timeout_DNS" error in retry rules, also documented since at least
6147 1.62, also never worked. As it isn't clear exactly what this means, and
6148 clearly it isn't a major issue, I have abolished the feature by treating it
6149 as "timeout", and writing a warning to the main and panic logs.
6151 36. The display of retry rules for -brt wasn't always showing the error code
6154 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6155 timeout_connect_A, timeout_connect_MX.
6157 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6158 to the empty sender.
6160 39. The daemon was not analysing the content of -oX till after it had closed
6161 stderr and disconnected from the controlling terminal. This meant that any
6162 syntax errors were only noted on the panic log, and the return code from
6163 the command was 0. By re-arranging the code a little, I've made the
6164 decoding happen first, so such errors now appear on stderr, and the return
6165 code is 1. However, the actual setting up of the sockets still happens in
6166 the disconnected process, so errors there are still only recorded on the
6169 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6170 connections (as happens on some IP stacks) was logged at start up time as
6171 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6172 it from "IPv6 and IPv4", which means that two separate sockets are being
6175 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6176 says whether AF_INET or AF_INET6 was passed as an argument.
6178 42. Exiwhat output was messed up when time zones were included in log
6181 43. Exiwhat now gives more information about the daemon's listening ports,
6182 and whether -tls-on-connect was used.
6184 44. The "port" option of the smtp transport is now expanded.
6186 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6187 silently ignored. Now an error message is written to the main and panic
6190 46. There's a new ACL modifier called "logwrite" which writes to a log file
6191 as soon as it is encountered.
6193 47. Added $local_user_uid and $local_user_gid at routing time.
6195 48. Exim crashed when trying to verify a sender address that was being
6198 49. Exim was recognizing only a space character after ".include". It now also
6199 recognizes a tab character.
6201 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6202 extracting the relevant information from the specification. The man page no
6203 longer contains scrambled data for the -d option, and I've added a section
6204 at the front about calling Exim under different names.
6206 51. Added "extra_headers" argument to the "mail" command in filter files.
6208 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6211 53. Installed eximstats 1.29.
6213 54. Added transport_filter_timeout as a generic transport option.
6215 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6216 Cc: header lines. This was required by RFC 822, but it not required by RFC
6219 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6220 incoming messages that did not have them. Now it does so only if the
6221 message originates locally, that is, if there is no associated remote host
6222 address. When Resent- header lines are present, this applies to the Resent-
6223 lines rather than the non-Resent- lines.
6225 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6226 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6228 58. Messages for configuration errors now include the name of the main
6229 configuration file - useful now that there may be more than one file in a
6230 list (.included file names were always shown).
6232 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6233 for those rare installations that do not start the daemon as root or run it
6234 setuid root. I've cut out the call to initgroups() if the daemon is not
6237 60. The Exim user and group can now be bound into the binary as text strings
6238 that are looked up at the start of Exim's processing.
6240 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6242 62. Added $mailstore_basename variable.
6244 63. Installed patch to sieve.c from Michael Haardt.
6246 64. When Exim failed to open the panic log after failing to open the main log,
6247 the original message it was trying to log was written to stderr and debug
6248 output, but if they were not available (the usual case in production), it
6249 was lost. Now it is written to syslog before the two lines that record the
6250 failures to open the logs.
6252 65. Users' Exim filters run in subprocesses under the user's uid. It is
6253 possible for a "deliver" command or an alias in a "personal" command to
6254 provoke an address rewrite. If logging of address rewriting is configured,
6255 this fails because the process is not running as root or exim. There may be
6256 a better way of dealing with this, but for the moment (because 4.30 needs
6257 to be released), I have disabled address rewrite logging when running a
6258 filter in a non-root, non-exim process.
6264 1. The buildconfig auxiliary program wasn't quoting the value set for
6265 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6266 not defined. This bug was masked in 4.22 by the effect that was fixed in
6269 2. Some messages that were rejected after a message id was allocated were
6270 shown as "incomplete" by exigrep. It no longer does this for messages that
6271 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6273 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6274 have allow_domain_literals set, the ID did not get logged in the <= line.
6275 Domain literals are now always recognized in Message-ID: header lines.
6277 4. The first argument for a ${extract expansion item is the key name or field
6278 number. Leading and trailing spaces in this item were not being ignored,
6279 causing some misleading effects.
6281 5. When deliver_drop_privilege was set, single queue runner processes started
6282 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6283 same command in the process it spins off) were not dropping privilege.
6285 6. When the daemon running as "exim" started a queue runner, it always
6286 re-executed Exim in the spun-off process. This is a waste of effort when
6287 deliver_drop_privilege is set. The new process now just calls the
6288 queue-runner function directly.
6294 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6297 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6298 ignored. Though the use of -oP was forcing the writing of a pid file, it
6299 was always written to the default place.
6301 3. If the message "no IP address found for host xxxx" is generated during
6302 incoming verification, it is now followed by identification of the incoming
6303 connection (so you can more easily find what provoked it).
6305 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6307 5. Added some features to "harden" Exim a bit more against certain attacks:
6309 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6310 be put in Local/Makefile. This is like the never_users runtime option,
6311 but it cannot be overridden. The default setting is "root".
6313 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6314 prefix string with which any file named in a -C command line option
6317 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6318 is retained for -C and -D only if the caller of Exim is root. Without
6319 it, the exim user may also use -C and -D and retain privilege.
6321 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6322 command line option is disabled.
6324 6. Macro names set by the -D option must start with an upper case letter, just
6325 like macro names defined in the configuration file.
6327 7. Added "dereference=" facility to LDAP.
6329 8. Two instances of the typo "uknown" in the source files are fixed.
6331 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6332 the Configure-Makefile script screwed up while processing it.
6334 10. Incorporated PCRE 4.4.
6336 11. The SMTP synchronization check was not operating right at the start of an
6337 SMTP session. For example, it could not catch a HELO sent before the client
6338 waited for the greeting. There is now a check for outstanding input at the
6339 point when the greeting is written. Because of the duplex, asynchronous
6340 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6341 way, but not yet received, when the check is performed.
6343 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6344 on TCP/IP sockets, because this apparently causes some broken clients to
6347 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6348 unchanged) from the Cygwin maintainer.
6350 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6351 was supported instead of testing for mbx. Effectively a typo.
6353 15. The spa authenticator server code was not checking that the input it
6354 received was valid base64.
6356 16. The debug output line for the "set" modifier in ACLs was not showing the
6357 name of the variable that was being set.
6359 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6361 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6362 lookup on its own. Something else was needed to trigger the lookup. For
6363 example, a match in host_lookup or the need for a host name in a host list.
6364 Now, if $sender_host_name is referenced and the host name has not yet been
6365 looked up, a lookup is performed. If the lookup fails, the variable remains
6366 empty, and $host_lookup_failed is set to "1".
6368 19. Added "eqi" as a case-independent comparison operator.
6370 20. The saslauthd authentication condition could segfault if neither service
6371 nor realm was specified.
6373 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6374 error message that was logged was misleading, and incoming SMTP
6375 connections were dropped. The message is now more accurate, and temporary
6376 errors are given to SMTP connections.
6378 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6379 (except RSET) with a 421 error, until QUIT is received. However, it was
6380 failing to send a response to QUIT.
6382 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6383 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6384 ACL is now run after the helo_try_verify_hosts code.
6386 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6387 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6388 case-independent, so other case variants are also recognized). Apparently
6389 some systems use these upper case variants.
6391 25. If more than two messages were waiting for the same host, and a transport
6392 filter was specified for the transport, Exim sent two messages over the
6393 same TCP/IP connection, and then failed with "socket operation on non-
6394 socket" when it tried to send the third.
6396 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6398 27. The extern definition of crypt16() in expand.c was not being excluded when
6399 the OS had its own crypt16() function.
6401 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6402 as a preferred synonym for return_size_limit, both as an option and as an
6405 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6407 30. Changed the default configuration ACL to relax the local part checking rule
6408 for addresses that are not in any local domains. For these addresses,
6409 slashes and pipe symbols are allowed within local parts, but the sequence
6410 /../ is explicitly forbidden.
6412 31. SPA server authentication was not clearing the challenge buffer before
6415 32. log_message in a "warn" ACL statement was writing to the reject log as
6416 well as to the main log, which contradicts the documentation and doesn't
6417 seem right (because no rejection is happening). So I have stopped it.
6419 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6420 However, I am unable to do any testing of this.
6422 34. Fixed an infelicity in the appendfile transport. When checking directories
6423 for a mailbox, to see if any needed to be created, it was accidentally
6424 using path names with one or more superfluous leading slashes; tracing
6425 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6427 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6428 contents are added to the log line that is written for every discarded
6429 recipient. (Previously a log_message setting was ignored.)
6431 36. The ${quote: operator now quotes the string if it is empty.
6433 37. The install script runs exim in order to find its version number. If for
6434 some reason other than non-existence or emptiness, which it checks, it
6435 could not run './exim', it was installing it with an empty version number,
6436 i.e. as "exim-". This error state is now caught, and the installation is
6439 38. An argument was missing from the function that creates an error message
6440 when Exim fails to connect to the socket for saslauthd authentication.
6441 This could cause Exim to crash, or give a corrupted message.
6443 39. Added isip, isip4, and isip6 to ${if conditions.
6445 40. The ACL variables $acl_xx are now saved with the message, and can be
6446 accessed later in routers, transports, and filters.
6448 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6449 strings in the file are not string-expanded.
6451 42. If a MAIL command specified a SIZE value that was too large to fit into an
6452 int variable, the check against message_size_limit failed. Such values are
6453 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6454 day this will have to be increased, but I don't think I want to be around
6455 when emails are that large.
6462 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6463 iconv() is not standard in FreeBSD.
6465 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6466 IPv6 enabled. The observed symptom was a segmentation fault on return from
6467 the function os_common_find_running_interfaces() in src/os.c.
6469 3. In the check_special_case() function in daemon.c I had used "errno" as an
6470 argument name, which causes warnings on some systems. This was basically a
6471 typo, since it was named "eno" in the comments!
6473 4. The code that waits for the clock to tick (at a resolution of some fraction
6474 of a second) so as to ensure message-id uniqueness was always waiting for
6475 at least one whole tick, when it could have waited for less. [This is
6476 almost certainly not relevant at current processor speeds, where it is
6477 unlikely to ever wait at all. But we try to future-proof.]
6479 5. The function that sleeps for a time interval that includes fractions of a
6480 second contained a race. It did not block SIGALRM between setting the
6481 timer, and suspending (a couple of lines later). If the interval was short
6482 and the sigsuspend() was delayed until after it had expired, the suspension
6483 never ended. On busy systems this could lead to processes getting stuck for
6486 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6487 process, before it forks any delivery processes. The open lookup caching
6488 mechanism meant that the open file or database connection was passed into
6489 the delivery process. The problem was that delivery processes always tidy
6490 up cached lookup data. This could cause a problem for the next delivery
6491 process started by the queue runner, because the external queue runner
6492 process does not know about the closure. So the next delivery process
6493 still has data in the lookup cache. In the case of a file lookup, there was
6494 no problem because closing a file descriptor in a subprocess doesn't affect
6495 the parent. However, if the lookup was caching a connection to a database,
6496 the connection was closed, and the second delivery process was likely to
6497 see errors such as "PGSQL: query failed: server closed the connection
6498 unexpectedly". The problem has been fixed by closing all cached lookups
6499 in a queue runner before running a delivery process.
6501 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6502 seem to have the "const" qualifier which it has on other OS. I've
6505 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6506 *of the same type* (client or server) with the same public name that an
6507 error should be diagnosed.
6509 9. When Exim looked up a host name for an IP address, but failed to find the
6510 original IP address when looking up the host name (a safety check), it
6511 output the message "<ip address> does not match any IP for NULL", which was
6512 confusing, to say the least. The bug was that the host name should have
6513 appeared instead of "NULL".
6515 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6516 user that is built into the binary, and the -C or -D options is used, root
6517 privilege is dropped before the configuration file is read. In addition,
6518 logging is switched to stderr instead of the normal log files. If the
6519 configuration then re-defines the Exim user, the unprivileged environment
6520 is probably not what is expected, so Exim logs a panic warning message (but
6523 However, if deliver_drop_privilege is set, the unprivileged state may well
6524 be exactly what is intended, so the warning has been cut out in that case,
6525 and Exim is allowed to try to write to its normal log files.
6531 1. smtp_return_error_details was not giving details for temporary sender
6532 or receiver verification errors.
6534 2. Diagnose a configuration error if two authenticators have the same public
6537 3. Exim used not to create the message log file for a message until the first
6538 delivery attempt. This could be confusing when incoming messages were held
6539 for policy or load reasons. The message log file is now created at the time
6540 the message is received, and an initial "Received" line is written to it.
6542 4. The automatically generated man page for command line options had a minor
6543 bug that caused no ill effects; however, a more serious problem was that
6544 the procedure for building the man page automatically didn't always
6545 operate. Consequently, release 4.20 contains an out-of-date version. This
6546 shouldn't happen again.
6548 5. When building Exim with embedded Perl support, the script that builds the
6549 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6550 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6552 6. The freeze_tell option was not being used for messages that were frozen on
6553 arrival, either by an ACL or by local_scan().
6555 7. Added the smtp_incomplete_transaction log selector.
6557 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6558 was accepting AUTH without a new EHLO.
6560 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6561 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6564 10. Logging of TCP/IP connections (when configured) now happens in the main
6565 daemon process instead of the child process, so that the TCP/IP connection
6566 count is more accurate (but it can never be perfect).
6568 11. The use of "drop" in a nested ACL was not being handled correctly in the
6569 outer ACL. Now, if condition failure induced by the nested "drop" causes
6570 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6571 or "require"), the connection is dropped.
6573 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6574 that yield "discard" can now be used with an "accept" or a "discard" verb,
6575 but an error is generated for any others (because I can't see a useful way
6576 to define what should happen).
6578 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6579 are now processed in the same way as lines in the Exim configuration file.
6580 In particular, continuation lines are supported.
6582 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6584 15. Added -ti meaning -t -i.
6586 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6587 domains, and warn by logging if others are found.
6589 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6590 returned from GIFCONF when Exim is trying to find the list of interfaces on
6591 a host. The code in os.c has been modified to copy each ifreq to an aligned
6592 structure in all cases.
6594 Also, in some cases, the returned ifreq's were being copied to a 'struct
6595 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6596 means the last couple of bytes of an IPv6 address could be chopped if the
6597 ifreq contained only a normal sockaddr (14 bytes storage).
6599 18. Named domain lists were not supported in the hosts_treat_as_local option.
6600 An entry such as +xxxx was not recognized, and was treated as a literal
6603 19. Ensure that header lines added by a DATA ACL are included in the reject log
6604 if the ACL subsequently rejects the message.
6606 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6607 MD5 (which is deprecated).
6609 21. When testing a filter file using -bf, Exim was writing a message when it
6610 took the sender from a "From " line in the message, but it was not doing so
6611 when it took $return_path from a Return-Path: header line. It now does.
6613 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6614 with a valid header line field name (a series of printing characters
6615 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6617 23. Changed "disc" in the source to "disk" to conform to the documentation and
6618 the book and for uniformity.
6620 24. Ignore Sendmail's -Ooption=value command line item.
6622 25. When execve() failed while trying to run a command in a pipe transport,
6623 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6624 could be confused with a return value of 69 from the command itself. This
6625 has been changed to 127, the value the shell returns if it is asked to run
6626 a non-existent command. The wording for the related log line suggests a
6627 non-existent command as the problem.
6629 26. If received_header_text expands to an empty string, do not add a Received:
6630 header line to the message. (Well, it adds a token one on the spool, but
6631 marks it "old" so that it doesn't get used or transmitted.)
6633 27. Installed eximstats 1.28 (addition of -nt option).
6635 28. There was no check for failure on the call to getsockname() in the daemon
6636 code. This can fail if there is a shortage of resources on the system, with
6637 ENOMEM, for example. A temporary error is now given on failure.
6639 29. Contrary to the C standard, it seems that in some environments, the
6640 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6641 program. Exim now does this explicitly; it affects the formatting of
6642 timestamps using strftime().
6644 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6645 complaints. I've now initialized all the variables, to avoid this.
6647 32. Header lines added by a system filter were not being "seen" during
6648 transport-time rewrites.
6650 33. The info_callback() function passed to OpenSSL is set up with type void
6651 (*)(SSL *, int, int), as described somewhere. However, when calling the
6652 function (actually a macro) that sets it up, the type void(*)() is
6653 expected. I've put in a cast to prevent warnings from picky compilers.
6655 34. If a DNS black list lookup found a CNAME record, but there were no A
6656 records associated with the domain it pointed at, Exim crashed.
6658 35. If a DNS black list lookup returned more than one A record, Exim ignored
6659 all but the first. It now scans all returned addresses if a particular IP
6660 value is being sought. In this situation, the contents of the
6661 $dnslist_value variable are a list of all the addresses, separated by a
6664 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6665 to accept a host name and all its aliases if the forward lookup for any of
6666 them yielded the IP address of the incoming connection. Now it accepts only
6667 those names whose forward lookup yields the correct IP address. Any other
6668 names are discarded. This closes a loophole whereby a rogue DNS
6669 administrator could create reverse DNS records to break through a
6670 wildcarded host restriction in an ACL.
6672 37. If a user filter or a system filter that ran in a subprocess used any of
6673 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6674 the wrong values were passed to the pipe command ($thisaddress had the
6675 value of $0, $0 had the value of $1, etc). This bug was introduced by
6676 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6678 38. Improved the line breaking for long SMTP error messages from ACLs.
6679 Previously, if there was no break point between 40 and 75 characters, Exim
6680 left the rest of the message alone. Two changes have been made: (a) I've
6681 reduced the minimum length to 35 characters; (b) if it can't find a break
6682 point between 35 and 75 characters, it looks ahead and uses the first one
6683 that it finds. This may give the occasional overlong line, but at least the
6684 remaining text gets split now.
6686 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6687 file descriptors might be low, and that setting 1000 would always raise it.
6688 It turns out that in some environments, the limit is already over 1000 and
6689 that lowering it causes trouble. So now Exim takes care not to decrease it.
6691 40. When delivering a message, the value of $return_path is set to $sender_
6692 address at the start of routing (routers may change the value). By an
6693 oversight, this default was not being set up when an address was tested by
6694 -bt or -bv, which affected the outcome if any router or filter referred to
6697 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6698 log only when "message" or "log_message" are set. However, if one of the
6699 conditions was an address verification, or a call to a nested ACL, the
6700 messages generated by the underlying test were being passed through. This
6701 no longer happens. The underlying message is available in $acl_verify_
6702 message for both "message" and "log_message" expansions, so it can be
6703 passed through if needed.
6705 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6706 new expansion $bh_ to give the encoded byte string without charset
6707 translation. Translation happens only if iconv() is available; HAVE_ICONV
6708 indicates this at build time. HEADERS_CHARSET gives the charset to
6709 translate to; headers_charset can change it in the configuration, and
6710 "headers charset" can change it in an individual filter file.
6712 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6713 that creates RFC 2047 encoded "words" labels them as that charset instead
6714 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6715 expansion operator; (ii) when Exim creates a From: line for a local
6716 message; (iii) when a header line is rewritten to include a "phrase" part.
6718 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6719 buggy, causing it to skip the first lines of messages whose message ID
6720 ended in 'D'. This would not have bitten before Exim release 4.14, because
6721 message IDs were unlikely to end in 'D' before then. The effect was to have
6722 incorrect size information for certain domains.
6724 45. #include "config.h" was missing at the start of the crypt16.c module. This
6725 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6728 46. If there was a timeout during a "random" callout check, Exim treated it as
6729 a failure of the random address, and carried on sending RSET and the real
6730 address. If the delay was just some slowness somewhere, the response to the
6731 original RCPT would be taken as a response to RSET and so on, causing
6732 mayhem of various kinds.
6734 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6735 when I implemented it. It didn't allow for the fact that some option values
6736 may legitimately be negative (e.g. size_addition), and it didn't even do
6737 the right test for positive values.
6739 48. Domain names in DNS records are case-independent. Exim always looks them up
6740 in lower case. Some resolvers return domain names in exactly the case they
6741 appear in the zone file, that is, they may contain uppercase letters. Not
6742 all resolvers do this - some return always lower case. Exim was treating a
6743 change of case by a resolver as a change of domain, similar to a widening
6744 of a domain abbreviation. This triggered its re-routing code and so it was
6745 trying to route what was effectively the same domain again. This normally
6746 caused routing to fail (because the router wouldn't handle the domain
6747 twice). Now Exim checks for this case specially, and just changes the
6748 casing of the domain that it ultimately uses when it transmits the message
6751 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6754 50. If a filter generated a file delivery with a non-absolute name (possible if
6755 no home directory exists for the router), the forbid_file option was not
6758 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6759 the existing equality matching.
6761 52. Exim was using ints instead of ino_t variables in some places where it was
6762 dealing with inode numbers.
6764 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6765 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6766 called TMPDIR, and if it finds it is different, it changes its value.
6768 54. The smtp_printf() function is now made available to local_scan() so
6769 additional output lines can be written before returning. There is also an
6770 smtp_fflush() function to enable the detection of a dropped connection.
6771 The variables smtp_input and smtp_batched_input are exported to
6774 55. Changed the default runtime configuration: the message "Unknown user"
6775 has been removed from the ACL, and instead placed on the localuser router,
6776 using the cannot_route_message feature. This means that any verification
6777 failures that generate their own messages won't get overridden. Similarly,
6778 the "Unrouteable address" message that was in the ACL for unverifiable
6779 relay addresses has also been removed.
6781 56. Added hosts_avoid_esmtp to the smtp transport.
6783 57. The exicyclog script was not checking for the esoteric option
6784 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6785 will work only if exicyclog is run under the appropriate euid.
6787 58. Following a discussion on the list, the rules by which Exim recognises line
6788 endings on incoming messages have been changed. The -dropcr and drop_cr
6789 options are now no-ops, retained only for backwards compatibility. The
6790 following line terminators are recognized: LF CRLF CR. However, special
6791 processing applies to CR:
6793 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6794 nor a local message in the state where . is a terminator.
6796 (ii) If a bare CR is encountered in a header line, an extra space is added
6797 after the line terminator so as not to end the header. The reasoning
6798 behind this is that bare CRs in header lines are most likely either
6799 to be mistakes, or people trying to play silly games.
6801 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6802 was being incorrectly given as 18 bytes larger than it should have been.
6803 This is a VOB (very old bug).
6805 60. This may never have affected anything current, but just in case it has:
6806 When the local host is found other than at the start of a list of hosts,
6807 the local host, those with the same MX, and any that follow, are discarded.
6808 When the list in question was part of a longer list of hosts, the following
6809 hosts (not currently being processed) were also being discarded. This no
6810 longer happens. I'm not sure if this situation could ever has previously
6813 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6816 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6817 header. This is something that is recommended in a new Internet Draft, and
6818 is something that is documented as being done by Sendmail. There are two
6819 possible values. For messages generated by the autoreply transport, Exim
6822 Auto-Submitted: auto-replied
6824 whereas for all other generated messages (e.g. bounces) it adds
6826 Auto-Submitted: auto-generated
6828 63. The "personal" condition in filters now includes a test for the
6829 Auto-Submitted: header. If it contains the string "auto-" the message it
6830 not considered personal.
6832 64. Added rcpt_include_affixes as a generic transport option.
6834 65. Added queue_only_override (default true).
6836 66. Added the syslog_duplication option.
6838 67. If what should have been the first header line of a message consisted of
6839 a space followed by a colon, Exim was mis-interpreting it as a header line.
6840 It isn't of course - it is syntactically invalid and should therefore be
6841 treated as the start of the message body. The misbehaviour could have
6842 caused a number of strange effects, including loss of data in subsequent
6843 header lines, and spool format errors.
6845 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6846 client host had authenticated. This control can now be exercised by an ACL
6847 for more flexibility.
6849 69. By default, callouts do not happen when testing with -bh. There is now a
6850 variant, -bhc, which does actually run the callout code, including
6851 consulting and updating the callout cache.
6853 70. Added support for saslauthd authentication, courtesy of Alexander
6856 71. If statvfs() failed on the spool or log directories while checking their
6857 size for availability, Exim confusingly gave the error "space shortage".
6858 Furthermore, in debugging mode it crashed with a floating point exception.
6859 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6860 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6861 serious problem, Exim now writes to the main and panic logs when this
6862 happens, with details of the failure. It then refuses to accept the
6863 incoming message, giving the message "spool directory problem" or "log
6864 directory problem" with a 421 code for SMTP messages.
6866 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6867 0, 1, and 2 exist, because some OS complain for execs without them (see
6868 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6869 descriptors. However, the code omitted to check that the open succeeded,
6870 causing mysterious errors if for some reason the permissions on /dev/null
6871 got screwed. Now Exim writes a message to the main and panic logs, and
6872 bombs out if it can't open /dev/null.
6874 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6875 interact so that it is all more flexible. It is supposed to remain
6876 backwards compatible. Also added extra_local_interfaces.
6878 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6879 to bomb out with an assertion failure - to the client this appears as a
6880 connection drop. This problem occurs in the part of the code that was taken
6881 from the Samba project. Fortunately, the assertion is in a very simple
6882 function, so I have fixed this by reproducing the function inline in the
6883 one place where it is called, and arranging for authentication to fail
6884 instead of killing the process with assert().
6886 75. The SPA client code was not working when the server requested OEM rather
6887 than Unicode encoding.
6889 76. Added code to make require_files with a specific uid setting more usable in
6890 the case where statting the file as root fails - usually a non-root-mounted
6891 NFS file system. When this happens and the failure is EACCES, Exim now
6892 forks a subprocess and does the per-uid checking as the relevant uid.
6894 77. Added process_log_path.
6896 78. If log_file_path was not explicitly set, a setting of check_log_space or
6897 check_log_inodes was ignored.
6899 79. If a space check for the spool or log partitions fails, the incident is now
6900 logged. Of course, in the latter case the data may get lost...
6902 80. Added the %p formatting code to string_format() so that it can be used to
6903 print addresses in debug_print(). Adjusted all the address printing in the
6904 debugging in store.c to use %p rather than %d.
6906 81. There was a concern that a line of code in smtp_in.c could overflow a
6907 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6908 initially expressed, the concern was not well-founded, because trailing
6909 spaces are removed early. However, if the trailing spaces were followed by
6910 a NULL, they did not get removed, so the overflow was possible. Two fixes
6913 (a) I re-wrote the offending code in a cleaner fashion.
6914 (b) If an incoming SMTP command contains a NULL character, it is rejected
6917 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6918 runs initgroups(), so that if the Exim user is in any additional groups,
6919 they will be used during message reception.
6925 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob