1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
9 JH/01 Avoid costly startup code when not strictly needed. This reduces time
10 for some exim process initialisations. It does mean that the logging
11 of TLS configuration problems is only done for the daemon startup.
13 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
15 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
16 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
19 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
20 DKIM signing is being done. Previously a transport_filter would always
21 disable CHUNKING, falling back to traditional DATA.
23 JH/05 Regard command-line receipients as tainted.
25 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
27 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
28 PAM library frees one of the arguments given to it, despite the
29 documentation. Therefore a plain malloc must be used.
31 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
32 on-stack buffers were used, resulting in a taint trap when DSN information
33 copied from a received message was written into the buffer.
35 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
36 the ordering of its ARC headers. This caused a crash.
38 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
39 a new record was being constructed with information from the peer, a trap
42 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
43 installation would get error messages from DMARC verify, when it hit the
44 nonexistent file indicated by the default. Distros wanting DMARC enabled
45 should both provide the file and set the option.
46 Also enforce no DMARC verification for command-line sourced messages.
48 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
49 could, depending on the platform, hang at the STARTTLS response.
51 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
52 message on a connection. Previously if one message had ARC headers and
53 the following one did not, a crash could result when adding an
54 Authentication-Results: header.
56 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
57 Exim main code and Exim-related utities. The introduction of taint
58 tracking also did many adjustments to string handling. Since then, eximon
59 frequently terminated with an assert failure.
61 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
62 check for 452 responses. This slightly helps the inefficieny of doing
63 a large alias-expansion into a recipient-limited target. The max_rcpt
64 transport option still applies (and at the current default, will override
65 the new feature). The check is done for either cause of synch, and forces
66 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
67 connection. The new facility is not tunable at this time.
69 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
70 library live data was being used, so the results became garbage. Make
71 copies while it is still usable.
73 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
74 on delivery deferred (==) and failed (**) lines (if a delivery was
75 attemtped). Previously it was only on completion (=>) lines.
77 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
78 for the expansion of the server_scram_iter and server_scram_salt options.
80 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
81 are now specifically given a NO_DATA response without hitting the system
82 resolver. The library goes on to do the now-standard TXT lookup.
83 Use of dnsdb lookups is not affected.
85 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
86 only retrieve the errormessage once. Previously two calls to dlerror()
87 were used, and the second one (for mainlog/paniclog) retrieved null
90 JH/20 Taint checking: disallow use of tainted data for
91 - the appendfile transport file and directory options
92 - the pipe transport command
93 - the autoreply transport file, log and once options
94 - file names used by the redirect router (including filter files)
96 - paths used by single-key lookups
97 Previously this was permitted.
99 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
100 adjusted the size of a major service buffer; this failed because the
101 buffer was in use at the time. Change to a compile-time increase in the
102 buffer size, when this authenticator is compiled into exim.
104 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
105 previous fast-mode was untenable in the face of glibs using mmap to
106 support larger malloc requests.
108 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
109 New values supported, if defined on system where compiled:
110 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
111 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
113 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
114 running a limited number of proceses in parallel, a benefit is gained. The
115 amount varies with the platform hardware and load. The use of the option
116 queue_run_in_order means we cannot do this, as ordering becomes
119 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
120 had introduced a string-copy (for ensuring NUL-termination) which was not
121 appropriate for that case, which can include embedded NUL bytes in the
122 block of data. Investigation showed the copy to actually be needless, the
123 data being length-specified.
125 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
126 done during a receiving connection, and both used TLS, global info was
127 used rather than per-connection info for tracking the state of data
128 queued for transmission. This could result in a connection hang.
130 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
131 Previously, when delivering serveral messages down a single connection
132 only the first would provide a SIZE. This was due to the size information
133 not being properly tracked.
135 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
136 TAI (at 37 seconds currently), pretend to be in UTC for time-related
137 expansion and logging. Previously, spurious values such as a future
138 minute could be seen.
140 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
141 it could crash from a null-deref. This could also affect the
142 ${addresses: } operator and ${readsock } item.
144 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
145 message following a mime one, the variable was not reset.
147 JH/30 When an pipelined-connect fails at the first response, assume incorrect
148 cached capability (perhaps the peer reneged?) and immediately retry in
151 JH/31 Fix spurious detection of timeout while writing to transport filter.
153 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
154 an attempt to copy the string was made before checking it.
156 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
157 taint of the lookup key was maintained; we now regard the presence in the
158 filesystem as sufficient validation.
160 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
161 argument is supplied.
163 JH/35 The dsearch lookup now requires that the directory is an absolute path.
164 Previously this was not checked, and nonempty relative paths made an
165 access under Exim's current working directory.
167 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
168 Previously no event was raised.
170 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
171 parameter supplied by the sender MAIL FROM command. Previously it was
172 ignored, and only the check_spool_space option value for the required
175 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
176 the size of the signing public-key. Previously it was instead giving
177 the size of the signature hash.
183 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
184 side, to decode a TLS 1.3 packet capture.
186 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
187 Previously the default library behaviour applied, sending two, each in
190 JH/03 Debug output for ACL now gives the config file name and line number for
193 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
195 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
197 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
198 buffer overrun for (non-chunking) other transports.
200 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
201 TLS1.3, means that a server rejecting a client certificate is not visible
202 to the client until the first read of encrypted data (typically the
203 response to EHLO). Add detection for that case and treat it as a failed
204 TLS connection attempt, so that the normal retry-in-clear can work (if
205 suitably configured).
207 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
208 and/or domain. Found and fixed by Jason Betts.
210 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
211 configuration). If a CNAME target was not a wellformed name pattern, a
214 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
215 the OS reports them interleaved with other addresses.
217 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
218 used both for input and for a verify callout, both encrypted, SMTP
219 responses being sent by the server could be lost. This resulted in
220 dropped connections and sometimes bounces generated by a peer sending
223 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
224 string. Previously it was possible to cause undefined behaviour in a
225 library routine (usually a crash). Found by "zerons".
227 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
230 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
231 API was removed, so update to use the newer ones.
233 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
234 any timeout set, is taking a long time. Previously we would hang on to a
235 rotated logfile "forever" if the input was arriving with long gaps
236 (a previous attempt to fix addressed lack, for a long time, of initial
239 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
240 shared (NFS) environment. The length of the tempfile name is now
241 4 + 16 ("hdr.$message_exim_id") which might break on file
242 systems which restrict the file name length to lower values.
245 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
246 shared (NFS) environment.
248 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
249 did for all versions <4.90). Notably -M, -m, --invert, -I may be
252 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
253 on some platforms for bit 31.
255 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
256 to changes apparently associated with TLS1.3 handling some of the APIs
257 previously used were either nonfunctional or inappropriate. Strings
258 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
259 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
260 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
261 This affects log line X= elements, the $tls_{in,out}_cipher variables,
262 and the use of specific cipher names in the encrypted= ACL condition.
264 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
266 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
267 verification result was not updated unless hosts_require_ocsp applied.
269 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
270 queue_list_requires_admin set to false, non-admin users were denied the
273 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
274 directory-of-certs mode. Previously they were advertised despite the
277 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
278 A single TCP connection by a client will now hold a TLS connection open
279 for multiple message deliveries, by default. Previoud the default was to
282 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
283 default. If built with the facility, DANE will be used. The facility
284 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
286 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
287 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
288 must be defined and you must still, unless you define DISABLE_TLS, manage
289 the the include-dir and library-file requirements that go with that
290 choice. Non-TLS builds are still supported.
292 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
295 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
296 default. If the platform supports and has the facility enabled, it will
297 be requested on all coneections.
299 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
300 controlled by the build-time option SUPPORT_PIPE_CONNECT.
302 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
304 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
305 success-DSN messages. Previously the From: header was always the default
306 one for these; the option was ignored.
308 JH/28 Fix the timeout on smtp response to apply to the whole response.
309 Previously it was reset for every read, so a teergrubing peer sending
310 single bytes within the time limit could extend the connection for a
311 long time. Credit to Qualsys Security Advisory Team for the discovery.
313 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
314 delivery address, which leaked information of the results of local
315 forwarding. Change to the original envelope recipient address, per
318 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
319 requested. Previously not bounce was generated and a log entry of
320 error ignored was made.
322 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
324 JH/32 Introduce a general tainting mechanism for values read from the input
325 channel, and values derived from them. Refuse to expand any tainted
326 values, to catch one form of exploit.
328 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
329 was unused and the unexpanded text used for the test. Found and
330 fixed by Ruben Jenster.
332 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
333 an attempt to use a TLS library read routine dereffed a nul pointer,
336 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
337 them in our smtp response.
339 JH/36 Have the general router option retry_use_local_part default to true when
340 any of the restrictive preconditions are set (to anything). Previously it
341 was only for check_local user. The change removes one item of manual
342 configuration which is required for proper retries when a remote router
343 handles a subset of addresses for a domain.
345 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
346 link count into consideration.
348 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
349 caused the extension of big_buffer, the following lines were ignored.
351 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
352 accordance with RFC 2308. Previously there was no expiry, so a longlived
353 receive process (eg. due to ACL delays) versus a short SOA value could
356 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
358 JH/39 Promote DMARC support to mainline.
360 JH/40 Bug 2452: Add a References: header to DSNs.
362 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
363 parameters. The relevant library call is documented as "Deprecated: This
364 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
365 3.6.0, DH parameters are negotiated following RFC7919."
367 HS/06 Change the default of dnssec_request_domains to "*"
369 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
370 carried on and emitted a BDAT command, even when PIPELINING was not
373 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
374 buffer was used for the filename, resulting in a trap when tainted
375 arguments (eg. $domain) were used.
377 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
378 recommended to avoid a possible server-load attack. The feature can be
379 re-enabled via the openssl_options main cofiguration option.
381 JH/45 local_scan API: documented the current smtp_printf() call. This changed
382 for version 4.90 - adding a "more data" boolean to the arguments.
383 Bumped the ABI version number also, this having been missed previously;
384 release versions 4.90 to 4.92.3 inclusive were effectively broken in
385 respect of usage of smtp_printf() by either local_scan code or libraries
386 accessed via the ${dlfunc } expansion item. Both will need coding
387 adjustment for any calls to smtp_printf() to match the new function
388 signature; a FALSE value for the new argument is always safe.
390 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
391 the file-offset (which the Linux syscall does, and exim expects); this
392 resulted in an indefinite loop.
394 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
395 to do ARC verification. The Authentication-Results: header line added
396 by the configuration then had no ARC item.
402 JH/01 Remove code calling the customisable local_scan function, unless a new
403 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
405 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
406 non-signal-safe functions being used.
408 JH/03 Bug 2269: When presented with a received message having a stupidly large
409 number of DKIM-Signature headers, disable DKIM verification to avoid
410 a resource-consumption attack. The limit is set at twenty.
412 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
413 report of oldest_pass in ${authres } in consequence, and separate out
414 some descriptions of reasons for verification fail.
416 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
417 files in the spool were present and unlocked. A queue-runner could spot
418 them, resulting in a duplicate delivery. Fix that by doing the unlock
419 after the unlink. Investigation by Tim Stewart. Take the opportunity to
420 add more error-checking on spoolfile handling while that code is being
423 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
424 No known attacks, no CVE, this is defensive hardening.
426 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
427 a queue-runner could start a delivery while other operations were ongoing.
428 Cutthrough delivery was a common victim, resulting in duplicate delivery.
429 Found and investigated by Tim Stewart. Fix by using the open message data
430 file handle rather than opening another, and not locally closing it (which
431 releases a lock) for that case, while creating the temporary .eml format
432 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
434 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
435 $sender_verify_failure/$recipient_verify_failure to "random".
437 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
440 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
441 Previously this would segfault.
443 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
446 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
447 like zero, since the resolver should be doing this for us, But we need one
448 as a CNAME but no MX presence gets the CNAME returned; we need to check
449 that doesn't point to an MX to declare it "no MX returned" rather than
450 "error, loop". A new main option is added so the older capability of
451 following some limited number of chain links is maintained.
453 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
455 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
456 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
457 now-deprecated earlier definitions used only specified the range up to TLS
458 1.2 (in the older-version library docs).
460 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
462 JH/15 Rework TLS client-side context management. Stop using a global, and
463 explicitly pass a context around. This enables future use of TLS for
464 connections to service-daemons (eg. malware scanning) while a client smtp
465 connection is using TLS; with cutthrough connections this is quite likely.
467 JH/16 Fix ARC verification to do AS checks in reverse order.
469 JH/17 Support a "tls" option on the ${readsocket } expansion item.
471 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
472 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
473 Previously the "utf8" would be re-prepended for every additional message.
475 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
476 Previously thery were accepted, resulting in issues when attempting to
477 forward messages to a non-supporting MTA.
479 PP/02 Let -n work with printing macros too, not just options.
481 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
482 one parent address was copied, and bogus data was used at delivery-logging
483 time. Either a crash (after delivery) or bogus log data could result.
484 Discovery and analysis by Tim Stewart.
486 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
487 Previously if the string ended mid-character, we did not insert the
488 promised '?' replacement.
490 PP/04 Documentation: current string operators work on bytes, not codepoints.
492 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
493 should pack well giving a smaller memory footprint so better caching and
494 therefore performance. Group the declarations where this can't be done so
495 that the byte-sized flag variables are not interspersed among pointer
496 variables, giving a better chance of good packing by the compiler.
498 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
499 non-null, to avoid issues with sites running BATV. Previously reports were
500 sent with an empty envelope sender so looked like bounces.
502 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
503 The ignore_error flag wasn't being returned from the filter subprocess so
504 was not set for later routers. Investigation and fix by Matthias Kurz.
506 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
507 and a msg:complete for the whole, when a message is manually removed using
508 -Mrm. Developement by Matthias Kurz, hacked on by JH.
510 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
511 a "Gnu special" function, asprintf() in the DB utility binary builds; I
512 hope that is portable enough.
514 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
515 requiring a known-CA anchor certificate; make it now rely entirely on the
516 TLSA as an anchor. Checking the name on the leaf cert against the name
517 on the A-record for the host is still done for TA (but not for EE mode).
519 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
520 included in delivery lines for non-proxied connections, when compiled with
521 SUPPORT_SOCKS and running with proxy logging enabled.
523 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
524 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
525 move the existing event to fire before the normal logging of message
526 failure so that custom logging is bracketed by normal logging.
528 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
529 msg:fail:internal event. Developement by Matthias Kurz.
531 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
532 far too small for todays use of crypto signatures stored there. Go all
533 the way to the max DNS message size of 64kB, even though this might be
534 overmuch for IOT constrained device use.
536 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
537 copy a string over itself. The library routine is documented as not
538 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
540 JH/32 For main options check_spool_space and check_inode_space, where the
541 platform supports 64b integers, support more than the previous 2^31 kB
542 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
543 the previous G, M, k.
545 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
546 $authenticated_fail_id variable on authentication failure. Previously
549 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
550 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
551 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
552 GNUTLS_SEC_PARAM_MEDIUM.
554 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
555 side. Previously we would continue as if no SNI had been received.
557 JH/36 Harden the handling of string-lists. When a list consisted of a sole
558 "<" character, which should be a list-separator specification, we walked
559 off past the nul-terimation.
561 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
562 causes) even when the retry time is not yet met. Previously they were
563 not, meaning that when (say) an account was over-quota and temp-rejecting,
564 and multiple senders' messages were queued, only one sender would get
565 notified on each configured delay_warning cycle.
567 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
569 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
570 cipher-suites, an error can be left on the stack even for a succeeding
571 accept; this results in impossible error messages when a later operation
574 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
575 return error codes indicating retry. Under TLS1.3 this becomes required.
577 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
578 it only wrote the new authenticators, resulting in a lack of tracking of
579 peer changes of ESMTP extensions until the next cache flush.
581 JH/41 Fix the loop reading a message header line to check for integer overflow,
582 and more-often against header_maxsize. Previously a crafted message could
583 induce a crash of the recive process; now the message is cleanly rejected.
585 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
586 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
592 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
593 When redis_servers is set to a list of > 1 element, and the Redis servers
594 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
595 case of MOVED into a DEFER case instead, thus moving the query onto the
596 next server in the list. For a cluster of N elements, all N servers must
597 be defined in redis_servers.
599 GF/02 Catch and remove uninitialized value warning in exiqsumm
600 Check for existence of @ARGV before looking at $ARGV[0]
602 JH/01 Replace the store_release() internal interface with store_newblock(),
603 which internalises the check required to safely use the old one, plus
604 the allocate and data copy operations duplicated in both (!) of the
605 extant use locations.
607 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
608 modifier. This matches the restriction on the commandline.
610 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
611 Previously only the last row was returned.
613 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
614 we assumed that tags in the header were well-formed, and parsed the
615 element content after inspecting only the first char of the tag.
616 Assumptions at that stage could crash the receive process on malformed
619 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
620 While running the DKIM ACL we operate on the Permanent memory pool so that
621 variables created with "set" persist to the DATA ACL. Also (at any time)
622 DNS lookups that fail create cache records using the Permanent pool. But
623 expansions release any allocations made on the current pool - so a dnsdb
624 lookup expansion done in the DKIM ACL releases the memory used for the
625 DNS negative-cache, and bad things result. Solution is to switch to the
626 Main pool for expansions.
627 While we're in that code, add checks on the DNS cache during store_reset,
628 active in the testsuite.
629 Problem spotted, and debugging aided, by Wolfgang Breyha.
631 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
632 When none of the hosts presented to a transport match an already-open
633 connection, close it and proceed with the list. Previously we would
634 queue the message. Spotted by Lena with Yahoo, probably involving
637 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
638 Previously a spurious "250 OK id=" response was appended to the proper
641 JH/08 The "support for" informational output now, which built with Content
642 Scanning support, has a line for the malware scanner interfaces compiled
643 in. Interface can be individually included or not at build time.
645 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
646 by the template makefile "src/EDITME". The "STREAM" support for an older
647 ClamAV interface method is removed.
649 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
650 rows affected is given instead).
652 JH/11 The runtime Berkeley DB library version is now additionally output by
653 "exim -d -bV". Previously only the compile-time version was shown.
655 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
656 SMTP connection. Previously, when one had more receipients than the
657 first, an abortive onward connection was made. Move to full support for
658 multiple onward connections in sequence, handling cutthrough connection
659 for all multi-message initiating connections.
661 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
662 routers. Previously, a multi-recipient message would fail to match the
663 onward-connection opened for the first recipient, and cause its closure.
665 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
666 a timeout on read on a GnuTLS initiating connection, resulting in the
667 initiating connection being dropped. This mattered most when the callout
668 was marked defer_ok. Fix to keep the two timeout-detection methods
671 JH/15 Relax results from ACL control request to enable cutthrough, in
672 unsupported situations, from error to silently (except under debug)
673 ignoring. This covers use with PRDR, frozen messages, queue-only and
676 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
678 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
679 metadata, resulting in a crash in free().
681 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
682 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
683 Broken also in d185889f4, with init system revamp.
685 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
686 Previously we abruptly closed the connection after reading a malware-
687 found indication; now we go on to read the "scan ok" response line,
690 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
691 ACL. Previously, a crash would result.
693 JH/19 Speed up macro lookups during configuration file read, by skipping non-
694 macro text after a replacement (previously it was only once per line) and
695 by skipping builtin macros when searching for an uppercase lead character.
697 JH/20 DANE support moved from Experimental to mainline. The Makefile control
698 for the build is renamed.
700 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
701 was allocated for every new TLS startup, meaning one per message. Fix
702 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
704 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
705 reported the original. Fix to report (as far as possible) the ACL
706 result replacing the original.
708 JH/23 Fix memory leak during multi-message connections using STARTTLS under
709 OpenSSL. Certificate information is loaded for every new TLS startup,
710 and the resources needed to be freed.
712 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
714 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
717 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
718 DATA response info to the (existing) per-recipient response info for
719 the "C=" log element. It can have useful tracking info from the
720 destination system. Patch from Simon Arlott.
722 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
723 length value. Previously this would segfault.
725 HS/02 Support Avast multiline protoocol, this allows passing flags to
726 newer versions of the scanner.
728 JH/28 Ensure that variables possibly set during message acceptance are marked
729 dead before release of memory in the daemon loop. This stops complaints
730 about them when the debug_store option is enabled. Discovered specifically
731 for sender_rate_period, but applies to a whole set of variables.
732 Do the same for the queue-runner and queue-list loops, for variables set
733 from spool message files. Do the same for the SMTP per-message loop, for
734 certain variables indirectly set in ACL operations.
736 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
737 as a multi-recipient message from a mailinglist manager). The coding had
738 an arbitrary cutoff number of characters while checking for more input;
739 enforced by writing a NUL into the buffer. This corrupted long / fast
740 input. The problem was exposed more widely when more pipelineing of SMTP
741 responses was introduced, and one Exim system was feeding another.
742 The symptom is log complaints of SMTP syntax error (NUL chars) on the
743 receiving system, and refused recipients seen by the sending system
744 (propating to people being dropped from mailing lists).
745 Discovered and pinpointed by David Carter.
747 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
748 replaced by the ${authresults } expansion.
750 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
752 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
753 allows proper process termination in container environments.
755 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
756 Previously the "final dot" had a newline after it; ensure it is CR,LF.
758 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
759 and "err_perm", deprecated since 4.83 when the RFC-defined words
760 "temperror" and "permerror" were introduced.
762 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
763 transport-filters or DKIM-signing. The restriction was lost in the
764 consolidation of verify-callout and delivery SMTP handling.
765 Extend the restriction to also cover ARC-signing.
767 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
768 in defer=pass mode supply a 450 to the initiator. Previously the message
771 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
772 tls_require_ciphers is used as before.
774 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
775 "pass_unscanned". Only tmpfails from the scanner are written to
776 the paniclog, as they may require admin intervention (permission
777 denied, license issues). Other scanner errors (like decompression
778 bombs) do not cause a paniclog entry.
780 JH/36 Fix reinitialisation of DKIM logging variable between messages.
781 Previously it was possible to log spurious information in receive log
784 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
785 triggered odd behaviour from Outlook Express clients.
787 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
790 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
791 since the IETF WG has not yet settled on that versus the original
792 "bare" representation.
794 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
795 Previously the millisecond value corrupted the output.
796 Fix also for syslog_pid=no and log_selector +pid, for which the pid
797 corrupted the output.
803 JH/01 Rework error string handling in TLS interface so that the caller in
804 more cases is responsible for logging. This permits library-sourced
805 string to be attached to addresses during delivery, and collapses
806 pairs of long lines into single ones.
808 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
809 during configuration. Wildcards are allowed and expanded.
811 JH/02 Rework error string handling in DKIM to pass more info back to callers.
812 This permits better logging.
814 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
815 do not close it down and have the child transport start it up again on
816 the passed-on TCP connection. Instead, proxy the child (and any
817 subsequent ones) for TLS via a unix-domain socket channel. Logging is
818 affected: the continued delivery log lines do not have any DNSSEC, TLS
819 Certificate or OCSP information. TLS cipher information is still logged.
821 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
822 identical IP addresses on different listening ports. Will also affect
825 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
826 add noisy ifdef guards to special-case this sillyness.
827 Patch from Bernd Kuhls.
829 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
830 than 255 are no longer allowed.
832 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
833 Disable the session-cache too, which might reduce our load. Since we
834 currrectly use a new context for every connection, both as server and
835 client, there is no benefit for these.
836 GnuTLS appears to not support tickets server-side by default (we don't
837 call gnutls_session_ticket_enable_server()) but client side is enabled
838 by default on recent versions (3.1.3 +) unless the PFS priority string
841 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
842 <https://reproducible-builds.org/specs/source-date-epoch/>.
844 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
845 the check for any unsuccessful recipients did not notice the limit, and
846 erroneously found still-pending ones.
848 JH/08 Pipeline CHUNKING command and data together, on kernels that support
849 MSG_MORE. Only in-clear (not on TLS connections).
851 JH/09 Avoid using a temporary file during transport using dkim. Unless a
852 transport-filter is involved we can buffer the headers in memory for
853 creating the signature, and read the spool data file once for the
854 signature and again for transmission.
856 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
857 4.77 as the kernel support then wasn't solid, having issues in 64bit
858 mode. Now, it's been long enough. Add support for FreeBSD also.
860 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
861 case where the routing stage had gathered several addresses to send to
862 a host before calling the transport for the first, we previously failed
863 to close down TLS in the old transport process before passing the TCP
864 connection to the new process. The new one sent a STARTTLS command
865 which naturally failed, giving a failed delivery and bloating the retry
866 database. Investigation and fix prototype from Wolfgang Breyha.
868 JH/12 Fix check on SMTP command input synchronisation. Previously there were
869 false-negatives in the check that the sender had not preempted a response
870 or prompt from Exim (running as a server), due to that code's lack of
871 awareness of the SMTP input buffering.
873 PP/04 Add commandline_checks_require_admin option.
874 Exim drops privileges sanely, various checks such as -be aren't a
875 security problem, as long as you trust local users with access to their
876 own account. When invoked by services which pass untrusted data to
877 Exim, this might be an issue. Set this option in main configuration
878 AND make fixes to the calling application, such as using `--` to stop
881 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
882 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
883 responses to those, into a single TLS record each way (this usually means
884 a single packet). As a side issue, smtp_enforce_sync now works on TLS
887 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
888 affects you only if you're dancing at the edge of the param size limits.
889 If you are, and this message makes sense to you, then: raise the
890 configured limit or use OpenSSL 1.1. Nothing we can do for older
893 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
894 cmdline element to get the documented default one. Previously it was
897 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
898 are unsuable for non-message-specific reasons - eg. connection timeout,
899 banner-time rejection.
901 JH/16 Fix logging of delivery remote port, when specified by router, under
904 PP/06 Repair manualroute's ability to take options in any order, even if one
905 is the name of a transport.
908 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
910 JH/17 Change the list-building routines interface to use the expanding-string
911 triplet model, for better allocation and copying behaviour.
913 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
914 Previously it was constructed the first time a possibly-matching string
915 was met in the configuration file input during startup; now it is done
918 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
919 compatible one, to avoid the (poorly documented) possibility of a config
920 file in the working directory redirecting the DB files, possibly correpting
921 some existing file. CVE-2017-10140 assigned for BDB.
923 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
924 cache-hot. Previously, although the result was properly cached, the
925 initial verify call returned a defer.
927 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
928 the main verify for receipient in uncached-mode.
930 JH/22 Retire historical build files to an "unsupported" subdir. These are
931 defined as "ones for which we have no current evidence of testing".
933 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
934 if present. Previously it was ignored.
936 JH/24 Start using specified-initialisers in C structure init coding. This is
937 a C99 feature (it's 2017, so now considered safe).
939 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
940 if was a fixed-sized field and bitmask ops via macros; it is now more
943 PP/07 GitHub PR 56: Apply MariaDB build fix.
944 Patch provided by Jaroslav Škarvada.
946 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
947 during Coverity cleanups [4.87 JH/47]
948 Diagnosis and fix provided by Michael Fischer v. Mollard.
950 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
951 the right size to place the terminating semicolon on its own folded
952 line, the header hash was calculated to an incorrect value thanks to
953 the (relaxed) space the fold became.
955 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
956 and confused the parent.
958 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
959 which could crash as a result. This could lead to undeliverable messages.
961 JH/28 Logging: "next input sent too soon" now shows where input was truncated
964 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
965 matters on fast-turnover and PID-randomising systems, which were getting
966 out-of-order delivery.
968 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
969 a possibly-overlapping copy. The symptom was that "Remote host closed
970 connection in response to HELO" was logged instead of the actual 4xx
973 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
974 Previously only that bufferd was discarded, resulting in SYMTP command
977 JH/32 DKIM: when a message has multiple signatures matching an identity given
978 in dkim_verify_signers, run the dkim acl once for each. Previously only
979 one run was done. Bug 2189.
981 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
982 "panic the current process" to "deliberately defer". The panic log is
983 still written with the problem list name; the mail and reject logs now
984 get a temp-reject line for the message that was being handled, saying
985 something like "domains check lookup or other defer". The SMTP 451
986 message is still "Temporary local problem".
988 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
989 A crafted sequence of BDAT commands could result in in-use memory beeing
990 freed. CVE-2017-16943.
992 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
993 from SMTP input. Previously it was always done; now only done for DATA
994 and not BDAT commands. CVE-2017-16944.
996 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
997 to the message (such as an overlong header line). Previously this was
998 not done and we did not exit BDAT mode. Followon from the previous item
999 though a different problem.
1005 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1006 than -2003 did; needs libidn2 in addition to libidn.
1008 JH/02 The path option on a pipe transport is now expanded before use.
1010 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1011 Patch provided by "Björn", documentation fix added too.
1013 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1014 missing a wire-to-host endian conversion.
1016 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1017 close after a BDAT command line could be taken as a following command,
1018 giving a synch failure. Fix by only checking for synch immediately
1019 before acknowledging the chunk.
1021 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1022 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1023 macro. Patches provided by Josh Soref.
1025 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1026 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1027 by IANA as a service extension. However, John Klensin suggests that we
1030 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1031 the dkim code may be unix-mode line endings rather than smtp wire-format
1032 CRLF, so prepend a CR to any bare LF.
1034 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1035 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1037 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1038 the first were themselves being wrongly included in the feed into dkim
1039 processing; with most chunk sizes in use this resulted in an incorrect
1040 body hash calculated value.
1042 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1043 DKIM signature block, for verification. Although advised against by
1044 standards it is specifically not ruled illegal.
1046 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1048 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1049 missing a body hash (the bh= tag).
1051 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1052 It seems that HAProxy sends the Proxy Protocol information in clear and
1053 only then does a TLS startup, so do the same.
1055 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1056 TCP connections (such as for Spamd) unless the daemon successfully set
1057 Fast Open mode on its listening sockets. This fixes breakage seen on
1058 too-old kernels or those not configured for Fast Open, at the cost of
1059 requiring both directions being enabled for TFO, and TFO never being used
1060 by non-daemon-related Exim processes.
1062 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1063 endings, at least on the first header line. Try to canonify any that get
1064 past that check, despite the cost.
1066 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1067 now limited to an arbitrary five deep, while parsing addresses with the
1068 strip_excess_angle_brackets option enabled.
1070 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1071 instead leave the unprompted TLS handshake in socket buffer for the
1072 TLS library to consume.
1074 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1076 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1078 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1079 region in one operation, for speed, and this leaves assigned pointers
1080 dangling. Add checks run only under the testsuite which checks all
1081 variables at a store-reset and panics on a dangling pointer; add code
1082 explicitly nulling out all the variables discovered. Fixes one known
1083 bug: a transport crash, where a dangling pointer for $sending_ip_address
1084 originally assigned in a verify callout, is re-used.
1086 PP/06 Drop '.' from @INC in various Perl scripts.
1088 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1090 PP/08 Reduce a number of compilation warnings under clang; building with
1091 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1092 should be warning-free.
1094 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1096 HS/01 Fix portability problems introduced by PP/08 for platforms where
1097 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1099 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1100 chunk. This allows us to accept broken chunked messages. We need a more
1101 general solution here.
1103 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1104 already-broken messages in the queue.
1106 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1108 JH/19 Fix reference counting bug in routing-generated-address tracking.
1114 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1115 supports it and a size is available (ie. the sending peer gave us one).
1117 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1118 years of being deprecated). The replacements are the ACLs
1119 acl_smtp_mime and acl_not_smtp_mime.
1121 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1122 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1123 or even an in-clear connection were permitted. Now, if the host lookup
1124 was dnssec and dane was requested then the host is only used if the
1125 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1126 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1127 if one fails this test.
1128 This means that a poorly-configured remote DNS will make it incommunicado;
1129 but it protects against a DNS-interception attack on it.
1131 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1132 non-noisy when a race steals the message being considered.
1134 JH/05 If main configuration option tls_certificate is unset, generate a
1135 self-signed certificate for inbound TLS connections.
1137 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1138 in rewrites and routers.
1140 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1141 and logged a warning sing 4.83; now they are a configuration file error.
1143 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1144 (lacking @domain). Apply the same qualification processing as RCPT.
1146 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1148 JH/10 Support ${sha256:} applied to a string (as well as the previous
1151 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1152 a cutthrough deliver is pending, as we always want to make a connection.
1153 This also avoids re-routing the message when later placing the cutthrough
1154 connection after a verify cache hit.
1155 Do not update it with the verify result either.
1157 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1158 when routing results in more than one destination address.
1160 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1161 signing (which inhibits the cutthrough capability). Previously only
1162 the presence of an option was tested; now an expansion evaluating as
1163 empty is permissible (obviously it should depend only on data available
1164 when the cutthrough connection is made).
1166 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1167 the relevant preceding SMTP command did not note the pipelining mode.
1169 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1170 Previously they were not counted.
1172 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1173 as one having no matching records. Previously we deferred the message
1174 that needed the lookup.
1176 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1177 distinguished as "(=".
1179 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1180 for missing MX records. Previously it only worked for missing A records.
1182 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1184 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1185 after the data-go-ahead and data-ack. Patch from Jason Betts.
1187 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1188 even for a "none" policy. Patch from Tony Meyer.
1190 JH/22 Fix continued use of a connection for further deliveries. If a port was
1191 specified by a router, it must also match for the delivery to be
1194 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1195 When one of the recipients of a message was unsuitable for the connection
1196 (has no matching addresses), we lost track of needing to mark it
1197 deferred. As a result mail would be lost.
1199 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1201 JH/25 Decoding ACL controls is now done using a binary search; the source code
1202 takes up less space and should be simpler to maintain. Merge the ACL
1203 condition decode tables also, with similar effect.
1205 JH/26 Fix problem with one_time used on a redirect router which returned the
1206 parent address unchanged. A retry would see the parent address marked as
1207 delivered, so not attempt the (identical) child. As a result mail would
1210 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1211 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1212 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1215 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1216 Main config options check_{log,spool}_{inodes,space} are now
1217 100 inodes, 10MB unless set otherwise in the configuration.
1219 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1220 Previously it only applied to the main-section connection policy
1223 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1225 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1226 by me. Added RFC7919 DH primes as an alternative.
1228 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1229 are not in the system include path.
1231 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1232 GnuTLS, when a session startup failed (eg because the client disconnected)
1233 Exim did stdio operations after fclose. This was exposed by a recent
1234 change which nulled out the file handle after the fclose.
1236 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1237 signed directly by the cert-signing cert, rather than an intermediate
1238 OCSP-signing cert. This is the model used by LetsEncrypt.
1240 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1242 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1243 an incoming connection.
1245 HS/02 Bug 1802: Do not half-close the connection after sending a request
1248 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1249 fallback to "prime256v1".
1251 JH/34 SECURITY: Use proper copy of DATA command in error message.
1252 Could leak key material. Remotely exploitable. CVE-2016-9963.
1258 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1259 and 3.4.4 - once the server is enabled to respond to an OCSP request
1260 it does even when not requested, resulting in a stapling non-aware
1261 client dropping the TLS connection.
1263 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1264 support variable-length bit vectors. No functional change.
1266 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1267 The I= interface field on outgoing lines is now after the H= remote
1268 host field, same as incoming lines. There is a separate
1269 outgoing_interface log selector which allows you to disable the
1272 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1273 If not running log_selector +smtp_connection the mainlog would be held
1274 open indefinitely after a "too many connections" event, including to a
1275 deleted file after a log rotate. Leave the per net connection logging
1276 leaving it open for efficiency as that will be quickly detected by the
1277 check on the next write.
1279 HS/01 Bug 1671: Fix post transport crash.
1280 Processing the wait-<transport> messages could crash the delivery
1281 process if the message IDs didn't exist for some reason. When
1282 using 'split_spool_directory=yes' the construction of the spool
1283 file name failed already, exposing the same netto behaviour.
1285 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1286 mime_regex ACL conditions.
1288 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1289 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1290 to HELO, local diagnostic string.
1292 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1293 log line to debug. Even when configured with a tls authenticator many
1294 client connections are expected to not authenticate in this way, so
1295 an authenticate fail is not an error.
1297 HS/02 Add the Exim version string to the process info. This way exiwhat
1298 gives some more detail about the running daemon.
1300 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1301 matter for fast-change records such as DNSBLs.
1303 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1304 retry record, even if constant. There may be multiple transports with
1305 different interface settings and the retry behaviour needs to be kept
1308 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1310 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1312 JH/10 Bug 840: fix log_defer_output option of pipe transport
1314 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1315 hosts_randomize. This matters a lot when combined with mua_wrapper.
1317 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1318 ${quote_pgsql:<string>} operator.
1320 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1322 JH/14 Change header file name for freeradius-client. Relevant if compiling
1323 with Radius support; from the Gentoo tree and checked under Fedora.
1325 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1327 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1328 option transport_filter, meaning no filtering.
1330 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1332 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1333 defaults to "*" (all hosts). The variable is now available when not built
1334 with TLS, default unset, mainly to enable keeping the testsuite sane.
1335 If a server certificate is not supplied (via tls_certificate) an error is
1336 logged, and clients will find TLS connections fail on startup. Presumably
1337 they will retry in-clear.
1338 Packagers of Exim are strongly encouraged to create a server certificate
1339 at installation time.
1341 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1342 with the $config_file variable.
1344 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1345 in transport context, after the attempt, and per-recipient. The latter type
1346 is per host attempted. The event data is the error message, and the errno
1347 information encodes the lookup type (A vs. MX) used for the (first) host,
1348 and the trailing two digits of the smtp 4xx response.
1350 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1351 to write to mainlog (or rejectlog, paniclog) in the window between file
1352 creation and permissions/ownership being changed. Particularly affects
1353 installations where exicyclog is run as root, rather than exim user;
1354 result is that the running daemon panics and dies.
1356 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1358 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1359 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1360 "pri" and "weight". Note that the previous implicit priority given by the
1361 list order is no longer honoured.
1363 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1364 for DKIM processing.
1366 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1367 by defining SUPPORT_SOCKS.
1369 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1370 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1371 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1372 variables are renamed to proxy_{local,external}_{address,port}.
1374 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1375 for a build by defining SUPPORT_I18N
1377 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1378 of the query string, and make ${quote_redis:} do that quoting.
1380 JH/29 Move Events support from Experimental to mainline, enabled by default
1381 and removable for a build by defining DISABLE_EVENT.
1383 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1385 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1386 cached by the daemon.
1388 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1389 by defining LOOKUP_REDIS. The libhiredis library is required.
1391 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1392 keys are given for lookup.
1394 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1395 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1396 only supported when built with TLS support. The PolarSSL SHA routines
1397 are still used when the TLS library is too old for convenient support.
1399 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1400 openssl_options), for security. OpenSSL forces this from version 1.1.0
1401 server-side so match that on older versions.
1403 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1404 allocation for $value could be released as the expansion processing
1405 concluded, but leaving the global pointer active for it.
1407 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1408 and to use the domains and local_parts ACL conditions.
1410 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1411 incorrectly not doubled on cutthrough transmission, hence seen as a
1412 body-termination at the receiving system - resulting in truncated mails.
1413 Commonly the sender saw a TCP-level error, and retransmitted the message
1414 via the normal store-and-forward channel. This could result in duplicates
1415 received - but deduplicating mailstores were liable to retain only the
1416 initial truncated version.
1418 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1420 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1422 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1423 we're in there, support oversigning also; bug 1309.
1425 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1427 HS/04 Add support for keep_environment and add_environment options.
1429 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1430 either intentional arithmetic overflow during PRNG, or testing config-
1433 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1434 delivery resulted in actual delivery. Cancel cutthrough before DATA
1437 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1438 rejects a recipient: pass the reject to the originator.
1440 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1441 Many were false-positives and ignorable, but it's worth fixing the
1444 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1445 for the new environment-manipulation done at startup. Move the routines
1446 from being local to tls.c to being global via the os.c file.
1448 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1449 an extract embedded as result-arg for a map, the first arg for extract
1450 is unavailable so we cannot tell if this is a numbered or keyed
1451 extraction. Accept either.
1457 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1460 JH/02 The smtp transport option "multi_domain" is now expanded.
1462 JH/03 The smtp transport now requests PRDR by default, if the server offers
1465 JH/04 Certificate name checking on server certificates, when exim is a client,
1466 is now done by default. The transport option tls_verify_cert_hostnames
1467 can be used to disable this per-host. The build option
1468 EXPERIMENTAL_CERTNAMES is withdrawn.
1470 JH/05 The value of the tls_verify_certificates smtp transport and main options
1471 default to the word "system" to access the system default CA bundle.
1472 For GnuTLS, only version 3.0.20 or later.
1474 JH/06 Verification of the server certificate for a TLS connection is now tried
1475 (but not required) by default. The verification status is now logged by
1476 default, for both outbound TLS and client-certificate supplying inbound
1479 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1482 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1483 Status Notification (bounce) messages are now MIME format per RFC 3464.
1484 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1485 under the control of the dsn_advertise_hosts option, and routers may
1486 have a dsn_lasthop option.
1488 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1489 default, modifiable by a malware= option. The list separator for
1490 the options can now be changed in the usual way. Bug 68.
1492 JH/10 The smtp_receive_timeout main option is now expanded before use.
1494 JH/11 The incoming_interface log option now also enables logging of the
1495 local interface on delivery outgoing connections.
1497 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1498 if the interface and destination host and port all match.
1500 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1503 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1504 Patch from Andrew Lewis.
1506 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1507 now supports optional time-restrictions, weighting, and priority
1508 modifiers per server. Patch originally by <rommer@active.by>.
1510 JH/16 The spamd_address main option now supports a mixed list of local
1511 and remote servers. Remote servers can be IPv6 addresses, and
1512 specify a port-range.
1514 JH/17 Bug 68: The spamd_address main option now supports an optional
1515 timeout value per server.
1517 JH/18 Bug 1581: Router and transport options headers_add/remove can
1518 now have the list separator specified.
1520 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1523 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1526 JH/21 Support for the A6 type of dns record is withdrawn.
1528 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1529 rather than the verbs used.
1531 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1532 from 255 to 1024 chars.
1534 JH/24 Verification callouts now attempt to use TLS by default.
1536 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1537 are generic router options now. The defaults didn't change.
1539 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1540 Original patch from Alexander Shikoff, worked over by JH.
1542 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1543 files if ZCAT_COMMAND is not executable.
1545 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1547 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1549 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1550 Normally benign, it bites when the pair was led to by a CNAME;
1551 modern usage is to not canonicalize the domain to a CNAME target
1552 (and we were inconsistent anyway for A-only vs AAAA+A).
1554 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1556 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1557 when evaluating $sender_host_dnssec.
1559 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1560 $sender_helo_dnssec variable.
1562 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1564 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1566 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1568 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1569 documented as working, but never had. Support all but $spam_report.
1571 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1572 added for tls authenticator.
1574 HS/03 Add perl_taintmode main config option
1580 TL/01 When running the test suite, the README says that variables such as
1581 no_msglog_check are global and can be placed anywhere in a specific
1582 test's script, however it was observed that placement needed to be near
1583 the beginning for it to behave that way. Changed the runtest perl
1584 script to read through the entire script once to detect and set these
1585 variables, reset to the beginning of the script, and then run through
1586 the script parsing/test process like normal.
1588 TL/02 The BSD's have an arc4random API. One of the functions to induce
1589 adding randomness was arc4random_stir(), but it has been removed in
1590 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1591 function when detected.
1593 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1594 cause callback expansion.
1596 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1597 syntax errors in an expansion can be treated as a string instead of
1598 logging or causing an error, due to the internal use of bool_lax
1599 instead of bool when processing it.
1601 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1602 server certificates when making smtp deliveries.
1604 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1606 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1608 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1610 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1611 Merged patch from Sebastian Wiedenroth.
1613 JH/05 Fix results-pipe from transport process. Several recipients, combined
1614 with certificate use, exposed issues where response data items split
1615 over buffer boundaries were not parsed properly. This eventually
1616 resulted in duplicates being sent. This issue only became common enough
1617 to notice due to the introduction of connection certificate information,
1618 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1620 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1621 size buffer was used, resulting in syntax errors when an expansion
1624 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1625 version 3.3.6 or later.
1627 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1628 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1629 both become "event_action", the variables become $event_name, $event_data
1630 and $event_defer_errno. There is a new variable $verify_mode, usable in
1631 routers, transports and related events. The tls:cert event is now also
1632 raised for inbound connections, if the main configuration event_action
1635 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1636 early OCSP support, but no stapling (appears to be less than 1.0.0).
1638 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1639 server certificate names available under the smtp transport option
1640 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1643 JH/10 Time-related extraction expansions from certificates now use the main
1644 option "timezone" setting for output formatting, and are consistent
1645 between OpenSSL and GnuTLS compilations. Bug 1541.
1647 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1648 encoded parameter in the incoming message. Bug 1558.
1650 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1651 include certificate info, eximon was claiming there were spoolfile
1654 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1656 JH/14 Log delivery-related information more consistently, using the sequence
1657 "H=<name> [<ip>]" wherever possible.
1659 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1660 are problematic for Debian distribution, omit them from the release
1663 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1665 JH/16 Fix string representation of time values on 64bit time_t architectures.
1668 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1669 output list separator was used.
1674 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1675 checkers that were complaining about end of non-void function with no
1678 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1679 This was a regression introduced in 4.83 by another bugfix.
1681 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1683 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1684 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1690 TF/01 Correctly close the server side of TLS when forking for delivery.
1692 When a message was received over SMTP with TLS, Exim failed to clear up
1693 the incoming connection properly after forking off the child process to
1694 deliver the message. In some situations the subsequent outgoing
1695 delivery connection happened to have the same fd number as the incoming
1696 connection previously had. Exim would try to use TLS and fail, logging
1697 a "Bad file descriptor" error.
1699 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1700 utilities have not been installed.
1702 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1703 temporary space as the ACL may create new global variables.
1705 TL/01 LDAP support uses per connection or global context settings, depending
1706 upon the detected version of the libraries at build time.
1708 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1709 to extract and use the src ip:port in logging and expansions as if it
1710 were a direct connection from the outside internet. PPv2 support was
1711 updated based on HAProxy spec change in May 2014.
1713 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1715 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1716 Properly escape header and check for NULL return.
1718 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1721 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1723 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1724 characters in header names, implemented as a verify condition.
1725 Contributed by Michael Fischer v. Mollard.
1727 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1728 results permerror and temperror. Previous values are deprecated but
1729 still accepted. In a future release, err_perm and err_temp will be
1730 completely removed, which will be a backward incompatibility if the
1731 ACL tests for either of these two old results. Patch contributed by
1732 user bes-internal on the mailing list.
1734 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1736 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1737 selectors, in both main and reject logs.
1739 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1742 JH/07 Add malware type "sock" for talking to simple daemon.
1744 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1746 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1747 routers/transports under cutthrough routing.
1749 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1750 numbers. Touch up "bool" conditional to keep the same definition.
1752 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1754 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1756 JH/12 Expand items in router/transport headers_add or headers_remove lists
1757 individually rather than the list as a whole. Bug 1452.
1759 Required for reasonable handling of multiple headers_ options when
1760 they may be empty; requires that headers_remove items with embedded
1761 colons must have them doubled (or the list-separator changed).
1763 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1764 view the policy declared in the DMARC record. Currently, $dmarc_status
1765 is a combined value of both the record presence and the result of the
1768 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1770 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1771 dnslookup router and the smtp transport (applying to the forward
1774 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1775 of ldap servers used for a specific lookup. Patch provided by Heiko
1778 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1779 New variable $lookup_dnssec_authenticated for observability.
1781 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1782 Patch submitted by Lars Timman.
1784 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1786 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1787 Requires trusted mode and valid format message id, aborts otherwise.
1788 Patch contributed by Heiko Schlichting.
1790 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1791 certextract with support for various fields. Bug 1358.
1793 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1794 is requested by default, modifiable by smtp transport option
1797 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1798 operate on certificate variables to give certificate fingerprints
1799 Also new ${sha256:cert_variable}.
1801 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1803 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1804 Christian Aistleitner.
1806 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1808 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1809 file. Patch from Wolfgang Breyha.
1811 JH/25 Expand the coverage of the delivery $host and $host_address to
1812 client authenticators run in verify callout. Bug 1476.
1814 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1815 align with daemon_smtp_ports. Bug 72.
1817 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1818 support and error reporting did not work properly.
1820 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1821 and is readable. Patch from Andrew Colin Kissa.
1823 TL/14 Enhance documentation of ${run expansion and how it parses the
1824 commandline after expansion, particularly in the case when an
1825 unquoted variable expansion results in an empty value.
1827 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1829 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1830 by renaming away from it. Addresses GH issue 15, reported by
1833 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1835 TL/15 SECURITY: prevent double expansion in math comparison functions
1836 (can expand unsanitized data). Not remotely exploitable.
1843 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1845 PP/02 Make -n do something, by making it not do something.
1846 When combined with -bP, the name of an option is not output.
1848 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1851 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1852 $sender_host_name and config options to manage this, and basic check
1855 PP/05 DSCP support for outbound connections and control modifier for inbound.
1857 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1858 (Only plugin which currently uses this is kerberos4, which nobody should
1859 be using, but we should make it available and other future plugins might
1860 conceivably use it, even though it would break NAT; stuff *should* be
1861 using channel bindings instead).
1863 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1864 name; added for Sendmail compatibility; requires admin caller.
1865 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1866 just ignore it); requires trusted caller.
1867 Also parse but ignore: -Ac -Am -X<logfile>
1870 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1872 TL/02 Add +smtp_confirmation as a default logging option.
1874 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1875 Patch by Magnus Holmgren from 2007-02-20.
1877 TL/04 Bugzilla 1281 - Spec typo.
1878 Bugzilla 1283 - Spec typo.
1879 Bugzilla 1290 - Spec grammar fixes.
1881 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1883 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1885 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1886 mailing list by by Dmitry Isaikin.
1888 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1890 JH/02 Support "G" suffix to numbers in ${if comparisons.
1892 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1894 NM/01 Bugzilla 1197 - Spec typo
1895 Bugzilla 1196 - Spec examples corrections
1897 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1899 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1900 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1903 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1904 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1906 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1907 "acl {{name}{arg}...}", and optional args on acl condition
1910 JH/05 Permit multiple router/transport headers_add/remove lines.
1912 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1914 JH/07 Avoid using a waiting database for a single-message-only transport.
1915 Performance patch from Paul Fisher. Bugzilla 1262.
1917 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1920 JH/09 Add $headers_added variable, with content from use of ACL modifier
1921 add_header (but not yet added to the message). Bugzilla 199.
1923 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1924 Pulled from Bugzilla 817 by Wolfgang Breyha.
1926 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1928 (nb: this is the same fix as in Exim 4.80.1)
1930 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1933 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1936 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1938 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1939 advertise SMTP AUTH mechanism to us, instead of a generic
1940 protocol violation error. Also, make Exim more robust to bad
1941 data from the Dovecot auth socket.
1943 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1945 When a queue runner is handling a message, Exim first routes the
1946 recipient addresses, during which it prunes them based on the retry
1947 hints database. After that it attempts to deliver the message to
1948 any remaining recipients. It then updates the hints database using
1951 So if a recipient address works intermittently, it can get repeatedly
1952 deferred at routing time. The retry hints record remains fresh so the
1953 address never reaches the final cutoff time.
1955 This is a fairly common occurrence when a user is bumping up against
1956 their storage quota. Exim had some logic in its local delivery code
1957 to deal with this. However it did not apply to per-recipient defers
1958 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1960 This change adds a proper retry rule check during routing so that the
1961 final cutoff time is checked against the message's age. We only do
1962 this check if there is an address retry record and there is not a
1963 domain retry record; this implies that previous attempts to handle
1964 the address had the retry_use_local_parts option turned on. We use
1965 this as an approximation for the destination being like a local
1966 delivery, as in LMTP.
1968 I suspect this new check makes the old local delivery cutoff check
1969 redundant, but I have not verified this so I left the code in place.
1971 TF/02 Correct gecos expansion when From: is a prefix of the username.
1973 Test 0254 submits a message to Exim with the header
1977 When I ran the test suite under the user fanf2, Exim expanded
1978 the header to contain my full name, whereas it should have added
1979 a Resent-Sender: header. It erroneously treats any prefix of the
1980 username as equal to the username.
1982 This change corrects that bug.
1984 GF/01 DCC debug and logging tidyup
1985 Error conditions log to paniclog rather than rejectlog.
1986 Debug lines prefixed by "DCC: " to remove any ambiguity.
1988 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1990 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1991 Bug spotted by Jeremy Harris; was flawed since initial commit.
1992 Would have resulted in OCSP responses post-SNI triggering an Exim
1993 NULL dereference and crash.
1995 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1997 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1998 Bug detection, analysis and fix by Samuel Thibault.
1999 Bugzilla 1331, Debian bug #698092.
2001 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2003 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2004 Server implementation by Todd Lyons, client by JH.
2005 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2006 config variable "prdr_enable" controls whether the server
2007 advertises the facility. If the client requests PRDR a new
2008 acl_data_smtp_prdr ACL is called once for each recipient, after
2009 the body content is received and before the acl_smtp_data ACL.
2010 The client is controlled by both of: a hosts_try_prdr option
2011 on the smtp transport, and the server advertisement.
2012 Default client logging of deliveries and rejections involving
2013 PRDR are flagged with the string "PRDR".
2015 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2016 fclose(). Diagnosis by Todd Lyons.
2018 PP/17 Update configure.default to handle IPv6 localhost better.
2019 Patch by Alain Williams (plus minor tweaks).
2022 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2023 This is now consistent with GnuTLS, and is now documented: the
2024 previous undocumented portable approach to treating the option as
2025 unset was to force an expansion failure. That still works, and
2026 an empty string is now equivalent.
2028 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2029 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2030 not performing validation itself.
2032 PP/20 Added force_command boolean option to pipe transport.
2033 Patch from Nick Koston, of cPanel Inc.
2035 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2038 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2040 PP/21 Fix eximon continuous updating with timestamped log-files.
2041 Broken in a format-string cleanup in 4.80, missed when I repaired the
2042 other false fix of the same issue.
2043 Report and fix from Heiko Schlichting.
2046 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2047 Report from Prashanth Katuri.
2049 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2050 It's SecureTransport, so affects any MacOS clients which use the
2051 system-integrated TLS libraries, including email clients.
2053 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2054 using a MIME ACL for non-SMTP local injection.
2055 Report and assistance in diagnosis by Warren Baker.
2057 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2059 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2061 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2062 last id that failed so it may be referenced in subsequent ACL's.
2064 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2067 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2068 ldap library initialization, allowing self-signed CA's to be
2069 used. Also properly sets require_cert option later in code by
2070 using NULL (global ldap config) instead of ldap handle (per
2071 session). Bug diagnosis and testing by alxgomz.
2073 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2074 the src/util/ subdirectory.
2076 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2077 renamed to Transport Post Delivery Action by Jeremy Harris, as
2080 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2081 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2082 redis_servers = needs to be configured which will be used by the redis
2083 lookup. Patch from Warren Baker, of The Packet Hub.
2085 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2087 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2088 hostname or reverse DNS when processing a host list. Used suggestions
2089 from multiple comments on this bug.
2091 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2093 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2094 Missed a few lines, added it to make the runtest require no keyboard
2097 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2098 contains upper case chars. Make router use caseful_local_part.
2100 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2101 support when GnuTLS has been built with p11-kit.
2107 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2109 This, or similar/improved, will also be change PP/11 of 4.82.
2115 PP/01 Handle short writes when writing local log-files.
2116 In practice, only affects FreeBSD (8 onwards).
2117 Bugzilla 1053, with thanks to Dmitry Isaikin.
2119 NM/01 Bugzilla 949 - Documentation tweak
2121 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2124 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2126 PP/02 Implemented gsasl authenticator.
2128 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2130 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2131 `pkg-config foo` for cflags/libs.
2133 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2134 with rest of GSASL and with heimdal_gssapi.
2136 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2137 `pkg-config foo` for cflags/libs for the TLS implementation.
2139 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2140 properties get this fed in as external SSF. A number of robustness
2141 and debugging improvements to the cyrus_sasl authenticator.
2143 PP/08 cyrus_sasl server now expands the server_realm option.
2145 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2146 Patch by Jeremy Harris.
2148 PP/10 Added dbmjz lookup type.
2150 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2152 PP/12 MAIL args handles TAB as well as SP, for better interop with
2153 non-compliant senders.
2154 Analysis and variant patch by Todd Lyons.
2156 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2157 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2158 Patch from Dirk Mueller <dmueller@suse.com>
2160 PP/13 tls_peerdn now print-escaped for spool files.
2161 Observed some $tls_peerdn in wild which contained \n, which resulted
2162 in spool file corruption.
2164 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2165 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2166 or write after TLS renegotiation, which otherwise led to messages
2169 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2170 as a tracking header (ie: a signed header comes before the signature).
2171 Patch from Wolfgang Breyha.
2173 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2174 comma-sep list; embedded commas doubled.
2176 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2178 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2180 Report and patch from Dmitry Banschikov.
2182 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2183 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2184 failures. We appear to now support TLS1.1+ with Exim.
2186 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2187 lets Exim select keys and certificates based upon TLS SNI from client.
2188 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2189 before an outbound SMTP session. New log_selector, +tls_sni.
2191 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2192 NULL dereference. Report and patch from Alun Jones.
2194 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2195 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2196 Not seeing resolver debug output on NetBSD, but suspect this is a
2197 resolver implementation change.
2199 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2200 Left warnings. Added "eximon gdb" invocation mode.
2202 PP/21 Defaulting "accept_8bitmime" to true, not false.
2204 PP/22 Added -bw for inetd wait mode support.
2206 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2207 locate the relevant includes and libraries. Made this the default.
2209 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2210 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2212 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2213 This may cause build issues on older platforms.
2215 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2216 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2217 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2218 Added SNI support via GnuTLS too.
2219 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2221 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2223 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2224 Applied second patch from Janne, implementing suggestion to default
2225 multiple-strings-in-record handling to match SPF spec.
2227 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2229 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2230 read-only, out of scope).
2231 Patch from Wolfgang Breyha, report from Stuart Northfield.
2233 PP/29 Fix three issues highlighted by clang analyser static analysis.
2234 Only crash-plausible issue would require the Cambridge-specific
2235 iplookup router and a misconfiguration.
2236 Report from Marcin Mirosław.
2238 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2240 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2241 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2242 As part of this, removing so much warning spew let me fix some minor
2243 real issues in debug logging.
2245 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2246 assignment on my part. Fixed.
2248 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2249 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2250 Janne Snabb (who went above and beyond: thank you).
2252 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2253 string otherwise requires a connection and a bunch more work and it's
2254 relatively easy to get wrong. Should also expose TLS library linkage
2257 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2258 64-bit ${eval} (JH/03).
2260 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2261 GNU libc to support some of the 64-bit stuff, should not lead to
2262 conflicts. Defined before os.h is pulled in, so if a given platform
2263 needs to override this, it can.
2265 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2266 protection layer was required, which is not implemented.
2267 Bugzilla 1254, patch from Wolfgang Breyha.
2269 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2270 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2271 tls_dhparam take prime identifiers. Also unbreak combination of
2272 OpenSSL+DH_params+TLSSNI.
2274 PP/39 Disable SSLv2 by default in OpenSSL support.
2280 PP/01 Solaris build fix for Oracle's LDAP libraries.
2281 Bugzilla 1109, patch from Stephen Usher.
2283 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2285 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2288 TF/02 Fix a couple more cases where we did not log the error message
2289 when unlink() failed. See also change 4.74-TF/03.
2291 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2292 lock up or crash if it happened to be inside a call to libc when it
2293 got a SIGUSR1 from exiwhat.
2295 The SIGUSR1 handler appends the current process status to the process
2296 log which is later printed by exiwhat. It used to use the general
2297 purpose logging code to do this, but several functions it calls are
2298 not safe for signals.
2300 The new output code in the SIGUSR1 handler is specific to the process
2301 log, and simple enough that it's easy to inspect for signal safety.
2302 Removing some special cases also simplifies the general logging code.
2303 Removing the spurious timestamps from the process log simplifies
2306 TF/04 Improved ratelimit ACL condition.
2308 The /noupdate option has been deprecated in favour of /readonly which
2309 has clearer semantics. The /leaky, /strict, and /readonly update modes
2310 are mutually exclusive. The update mode is no longer included in the
2311 database key; it just determines when the database is updated. (This
2312 means that when you upgrade Exim will forget old rate measurements.)
2314 Exim now checks that the per_* options are used with an update mode that
2315 makes sense for the current ACL. For example, when Exim is processing a
2316 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2317 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2318 must specify per_mail/readonly. If you omit the update mode it defaults to
2319 /leaky where that makes sense (as before) or /readonly where required.
2321 The /noupdate option is now undocumented but still supported for
2322 backwards compatibility. It is equivalent to /readonly except that in
2323 ACLs where /readonly is required you may specify /leaky/noupdate or
2324 /strict/noupdate which are treated the same as /readonly.
2326 A useful new feature is the /count= option. This is a generalization
2327 of the per_byte option, so that you can measure the throughput of other
2328 aggregate values. For example, the per_byte option is now equivalent
2329 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2331 The per_rcpt option has been generalized using the /count= mechanism
2332 (though it's more complicated than the per_byte equivalence). When it is
2333 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2334 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2335 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2336 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2337 Note that using per_rcpt with a non-readonly update mode in more than
2338 one ACL will cause the recipients to be double-counted. (The per_mail
2339 and per_byte options don't have this problem.)
2341 The handling of very low rates has changed slightly. If the computed rate
2342 is less than the event's count (usually one) then this event is the first
2343 after a long gap. In this case the rate is set to the same as this event's
2344 count, so that the first message of a spam run is counted properly.
2346 The major new feature is a mechanism for counting the rate of unique
2347 events. The new per_addr option counts the number of different
2348 recipients that someone has sent messages to in the last time period. It
2349 behaves like per_rcpt if all the recipient addresses are different, but
2350 duplicate recipient addresses do not increase the measured rate. Like
2351 the /count= option this is a general mechanism, so the per_addr option
2352 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2353 example, measure the rate that a client uses different sender addresses
2354 with the options per_mail/unique=$sender_address. There are further
2355 details in the main documentation.
2357 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2359 TF/06 Removed a few PCRE remnants.
2361 TF/07 Automatically extract Exim's version number from tags in the git
2362 repository when doing development or release builds.
2364 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2365 Bugzilla 879. Patch from Paul Fisher.
2367 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2368 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2371 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2373 PP/05 Variable $av_failed, true if the AV scanner deferred.
2374 Bugzilla 1078. Patch from John Horne.
2376 PP/06 Stop make process more reliably on build failure.
2377 Bugzilla 1087. Patch from Heiko Schlittermann.
2379 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2380 Bugzilla 1089. Patch from Heiko Schlittermann.
2382 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2383 Bugzilla 1131. Patch from Holger Weiß.
2385 PP/09 Handle IPv6 addresses with SPF.
2386 Bugzilla 860. Patch from Wolfgang Breyha.
2388 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2390 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2393 PP/11 match_* no longer expand right-hand-side by default.
2394 New compile-time build option, EXPAND_LISTMATCH_RHS.
2395 New expansion conditions, "inlist", "inlisti".
2397 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2399 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2401 PP/14 fix log_write() format string regression from TF/03.
2402 Bugzilla 1152. Patch from Dmitry Isaikin.
2408 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2410 PP/02 Harmonised TLS library version reporting; only show if debugging.
2411 Layout now matches that introduced for other libraries in 4.74 PP/03.
2413 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2415 PP/04 New "dns_use_edns0" global option.
2417 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2420 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2421 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2423 TK/01 Updated PolarSSL code to 0.14.2.
2424 Bugzilla 1097. Patch from Andreas Metzler.
2426 PP/07 Catch divide-by-zero in ${eval:...}.
2427 Fixes bugzilla 1102.
2429 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2432 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2433 format-string attack -- SECURITY: remote arbitrary code execution.
2435 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2436 time unintentionally subject to list matching rules, letting the header
2437 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2438 arbitrary string expansion). This allowed for information disclosure.
2440 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2441 INT_MIN/-1 -- value coerced to INT_MAX.
2447 NM/01 Workaround for PCRE version dependency in version reporting
2450 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2451 This fixes portability to compilers other than gcc, notably
2452 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2454 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2455 makefiles for portability to HP-UX and POSIX correctness.
2457 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2458 Also via indented variable definition in the Makefile.
2459 (Debugging by Oliver Heesakkers).
2461 PP/02 Restore caching of spamd results with expanded spamd_address.
2462 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2464 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2465 Improves build reliability. Fix from: Frank Elsner
2467 NM/02 Fix wide character breakage in the rfc2047 coding
2468 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2470 NM/03 Allow underscore in dnslist lookups
2471 Fixes bug 1026. Patch from Graeme Fowler
2473 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2474 Code patches from Adam Ciarcinski of NetBSD.
2476 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2479 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2480 is logged, to avoid truncation. Patch from John Horne.
2482 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2483 Patch from Jakob Hirsch.
2485 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2486 SQL string expansion failure details.
2487 Patch from Andrey Oktyabrski.
2489 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2490 Patch from Simon Arlott.
2492 PP/09 New lookups functionality failed to compile on old gcc which rejects
2493 extern declarations in function scope.
2494 Patch from Oliver Fleischmann
2496 PP/10 Use sig_atomic_t for flags set from signal handlers.
2497 Check getgroups() return and improve debugging.
2498 Fixed developed for diagnosis in bug 927 (which turned out to be
2501 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2502 Patch from Mark Zealey.
2504 PP/12 Bugzilla 1056: Improved spamd server selection.
2505 Patch from Mark Zealey.
2507 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2508 Based on patch from Heiko Schlittermann.
2510 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2511 Patch from Uwe Doering, sign-off by Michael Haardt.
2513 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2514 variable declaration deep within a block. Bug and patch from
2517 PP/15 lookups-Makefile IRIX compatibility coercion.
2519 PP/16 Make DISABLE_DKIM build knob functional.
2521 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2522 Patch by Simon Arlott
2524 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2525 variable argument macros. Our copy now differs from upstream.
2531 TF/01 Failure to get a lock on a hints database can have serious
2532 consequences so log it to the panic log.
2534 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2535 controlled using the smtp_confirmation log selector.
2537 TF/03 Include the error message when we fail to unlink a spool file.
2539 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2540 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2541 for maintaining out-of-tree patches for some time.
2543 PP/01 Bugzilla 139: Documentation and portability issues.
2544 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2545 Handle per-OS dynamic-module compilation flags.
2547 PP/02 Let /dev/null have normal permissions.
2548 The 4.73 fixes were a little too stringent and complained about the
2549 permissions on /dev/null. Exempt it from some checks.
2550 Reported by Andreas M. Kirchwitz.
2552 PP/03 Report version information for many libraries, including
2553 Exim version information for dynamically loaded libraries. Created
2554 version.h, now support a version extension string for distributors
2555 who patch heavily. Dynamic module ABI change.
2557 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2558 privilege escalation vulnerability whereby the Exim run-time user
2559 can cause root to append content of the attacker's choosing to
2562 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2565 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2566 If dropping privileges for untrusted macros, we disabled normal logging
2567 on the basis that it would fail; for the Exim run-time user, this is not
2568 the case, and it resulted in successful deliveries going unlogged.
2569 Fixed. Reported by Andreas Metzler.
2575 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2576 only prepend for the Resent-* case. Fixes regression introduced in
2577 Exim 4.70 by NM/22 for Bugzilla 607.
2579 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2580 increasing numbers of administrators be bitten by this.
2582 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2584 PP/03 Bugzilla 994: added openssl_options main configuration option.
2586 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2588 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2590 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2592 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2593 without a peer certificate, leading to a segfault because of an
2594 assumption that peers always have certificates. Be a little more
2595 paranoid. Problem reported by Martin Tscholak.
2597 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2598 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2599 NB: ClamAV planning to remove STREAM in "middle of 2010".
2600 CL also introduces -bmalware, various -d+acl logging additions and
2601 more caution in buffer sizes.
2603 PP/09 Implemented reverse_ip expansion operator.
2605 PP/10 Bugzilla 937: provide a "debug" ACL control.
2607 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2609 PP/12 Bugzilla 973: Implement --version.
2611 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2613 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2615 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2617 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2618 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2619 ignore trailing whitespace.
2621 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2623 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2626 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2627 Notification from Dr Andrew Aitchison.
2629 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2630 ExtendedDetectionInfo response format.
2631 Notification from John Horne.
2633 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2636 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2637 XSL and documented dependency on system catalogs, with examples of how
2640 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2643 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2644 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2645 configuration file which is writeable by the Exim user or group.
2647 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2648 of configuration files to cover files specified with the -C option if
2649 they are going to be used with root privileges, not just the default
2652 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2653 option (effectively making it always true).
2655 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2656 files to be used while preserving root privileges.
2658 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2659 that rogue child processes cannot use them.
2661 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2662 run-time user, instead of root.
2664 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2665 Exim run-time user without dropping privileges.
2667 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2668 result string, instead of calling string_vformat() twice with the same
2671 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2672 for other users. Others should always drop root privileges if they use
2673 -C on the command line, even for a whitelisted configure file.
2675 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2677 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2683 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2684 $data_path, and $header_path variables; fixed documentation bugs and
2687 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2688 exipick to access non-standard spools, including the "frozen" queue
2691 NM/01 Bugzilla 965: Support mysql stored procedures.
2692 Patch from Alain Williams
2694 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2696 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2697 Patch from Andreas Metzler
2699 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2700 Patch from Kirill Miazine
2702 NM/05 Bugzilla 671: Added umask to procmail example.
2704 JJ/03 installed exipick 20100323.0, fixing doc bug
2706 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2707 directory. Notification and patch from Dan Rosenberg.
2709 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2711 TK/02 Improve log output when DKIM signing operation fails.
2713 MH/01 Treat the transport option dkim_domain as a colon separated
2714 list, not as a single string, and sign the message with each element,
2715 omitting multiple occurences of the same signer.
2717 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2718 Bugzilla 985, 986. Patch by Simon Arlott
2720 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2721 Patch by Simon Arlott
2723 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2724 MBX locking. Notification from Dan Rosenberg.
2730 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2732 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2734 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2736 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2738 NM/04 Bugzilla 915: Flag broken perl installation during build.
2744 TK/01 Added patch by Johannes Berg that expands the main option
2745 "spamd_address" if it starts with a dollar sign.
2747 TK/02 Write list of recipients to X-Envelope-Sender header when building
2748 the mbox-format spool file for content scanning (suggested by Jakob
2751 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2752 (http://www.dcc-servers.net/) support via dccifd. Activated by
2753 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2755 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2756 by Mark Daniel Reidel <mr@df.eu>.
2758 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2759 When building exim an external PCRE library is now needed -
2760 PCRE is a system library on the majority of modern systems.
2761 See entry on PCRE_LIBS in EDITME file.
2763 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2764 conversation. Added nologin parameter to request.
2765 Patch contributed by Kirill Miazine.
2767 TF/01 Do not log submission mode rewrites if they do not change the address.
2769 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2771 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2772 log files in place. Contributed by Roberto Lima.
2774 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2776 TF/03 Bugzilla 615: When checking the local_parts router precondition
2777 after a local_part_suffix or local_part_prefix option, Exim now
2778 does not use the address's named list lookup cache, since this
2779 contains cached lookups for the whole local part.
2781 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2782 Robert Millan. Documentation is in experimental-spec.txt.
2784 TF/04 Bugzilla 668: Fix parallel build (make -j).
2786 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2788 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2789 Patch provided by Jan Srzednicki.
2791 TF/05 Leading white space used to be stripped from $spam_report which
2792 wrecked the formatting. Now it is preserved.
2794 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2795 that they are available at delivery time.
2797 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2799 TF/08 TLS error reporting now respects the incoming_interface and
2800 incoming_port log selectors.
2802 TF/09 Produce a more useful error message if an SMTP transport's hosts
2803 setting expands to an empty string.
2805 NM/06 Bugzilla 744: EXPN did not work under TLS.
2806 Patch provided by Phil Pennock.
2808 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2809 Patch provided by Richard Godbee.
2811 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2812 acl_smtp_notquit, added index entry.
2814 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2815 Patch provided by Eugene Bujak.
2817 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2818 accept(). Patch provided by Maxim Dounin.
2820 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2821 Patch provided by Phil Pennock.
2823 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2825 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2826 Patch provided by Brad "anomie" Jorsch.
2828 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2829 Patch provided by Dean Brooks.
2831 TK/05 Add native DKIM support (does not depend on external libraries).
2833 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2834 Patch provided by Graeme Fowler.
2836 NM/16 Bugzilla 851: Documentation example syntax fix.
2838 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2840 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2843 NM/19 Bugzilla 745: TLS version reporting.
2844 Patch provided by Phil Pennock.
2846 NM/20 Bugzilla 167: bool: condition support.
2847 Patch provided by Phil Pennock.
2849 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2850 clients. Patch provided by Phil Pennock.
2852 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2853 Patch provided by Brad "anomie" Jorsch.
2855 NM/23 Bugzilla 687: Fix misparses in eximstats.
2856 Patch provided by Heiko Schlittermann.
2858 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2859 Patch provided by Heiko Schlittermann.
2861 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2862 plus update to original patch.
2864 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2866 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2867 Patch provided by David Brownlee.
2869 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2871 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2873 NM/30 Bugzilla 888: TLS documentation bugfixes.
2875 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2877 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2878 Unlike the original bugzilla I have changed all shell scripts in src tree.
2880 NM/33 Bugzilla 898: Transport filter timeout fix.
2881 Patch by Todd Rinaldo.
2883 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2884 Patch by Serge Demonchaux.
2886 NM/35 Bugzilla 39: Base64 decode bug fixes.
2887 Patch by Jakob Hirsch.
2889 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2891 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2893 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2895 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2901 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2902 ALT-N's libdkim that I have put here:
2903 http://duncanthrax.net/exim-experimental/
2905 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2906 were called 'true' and it seems that C99 defines that as a reserved
2907 keyword to be used with 'bool' variable types. That means you could
2908 not include C99-style headers which use bools without triggering
2909 build errors in sieve.c.
2911 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2912 as mailq or other aliases. Changed the --help handling significantly
2913 to do whats expected. exim_usage() emits usage/help information.
2915 SC/01 Added the -bylocaldomain option to eximstats.
2917 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2919 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2921 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2927 PH/01 Another patch from the Sieve maintainer.
2929 PH/02 When an IPv6 address is converted to a string for single-key lookup
2930 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2931 dots are used instead of colons so that keys in lsearch files need not
2932 contain colons. This was done some time before quoting was made available
2933 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2934 (notated using the quote facility) so as to distinguish them from IPv4
2935 keys. This meant that lookups for IP addresses in host lists did not work
2936 for iplsearch lookups.
2938 This has been fixed by arranging for IPv6 addresses to be expressed with
2939 colons if the lookup type is iplsearch. This is not incompatible, because
2940 previously such lookups could never work.
2942 The situation is now rather anomalous, since one *can* have colons in
2943 ordinary lsearch keys. However, making the change in all cases is
2944 incompatible and would probably break a number of configurations.
2946 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2949 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2950 conversion specification without a maximum field width, thereby enabling
2951 a rogue spamd server to cause a buffer overflow. While nobody in their
2952 right mind would setup Exim to query an untrusted spamd server, an
2953 attacker that gains access to a server running spamd could potentially
2954 exploit this vulnerability to run arbitrary code as the Exim user.
2956 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2957 $primary_hostname instead of what libspf2 thinks the hosts name is.
2959 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2960 a directory entry by the name of the lookup key. Previously, if a
2961 symlink pointed to a non-existing file or a file in a directory that
2962 Exim lacked permissions to read, a lookup for a key matching that
2963 symlink would fail. Now it is enough that a matching directory entry
2964 exists, symlink or not. (Bugzilla 503.)
2966 PH/03 The body_linecount and body_zerocount variables are now exported in the
2969 PH/04 Added the $dnslist_matched variable.
2971 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2972 This means they are set thereafter only if the connection becomes
2975 PH/06 Added the client_condition to authenticators so that some can be skipped
2976 by clients under certain conditions.
2978 PH/07 The error message for a badly-placed control=no_multiline_responses left
2979 "_responses" off the end of the name.
2981 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2983 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2984 (without spaces) instead of just copying the configuration text.
2986 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2988 PH/11 Added $max_received_linelength.
2990 PH/12 Added +ignore_defer and +include_defer to host lists.
2992 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2993 way in which PCRE > 7.0 is built.
2995 PH/14 Implemented queue_only_load_latch.
2997 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2998 MAIL command. The effect was to mangle the value on 64-bit systems.
3000 PH/16 Another patch from the Sieve maintainer.
3002 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3004 PH/18 If a system quota error occurred while trying to create the file for
3005 a maildir delivery, the message "Mailbox is full" was not appended to the
3006 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3007 only to a quota excession during the actual writing of the file.
3009 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3010 characters?) which causes problems in log lines. The DN values are now
3011 passed through string_printing() before being added to log lines.
3013 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3014 and InterBase are left for another time.)
3016 PH/21 Added message_body_newlines option.
3018 PH/22 Guard against possible overflow in moan_check_errorcopy().
3020 PH/23 POSIX allows open() to be a macro; guard against that.
3022 PH/24 If the recipient of an error message contained an @ in the local part
3023 (suitably quoted, of course), incorrect values were put in $domain and
3024 $local_part during the evaluation of errors_copy.
3030 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3031 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3034 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3035 issue a MAIL command.
3037 PH/02 In an ACL statement such as
3039 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3041 if a client was not listed at all, or was listed with a value other than
3042 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3043 the condition was not true (as it should be), so access was not denied.
3044 The bug was that the ! inversion was incorrectly passed on to the second
3045 item. This has been fixed.
3047 PH/03 Added additional dnslists conditions == and =& which are different from
3048 = and & when the dns lookup returns more than one IP address.
3050 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3051 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3053 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3054 FSYNC, which compiles an option called disable_fsync that allows for
3055 bypassing fsync(). The documentation is heavily laced with warnings.
3057 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3059 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3060 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3061 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3062 including adding "make clean"; (3) Added -fPIC when compiling the test
3063 dynamically loaded module, to get rid of a warning.
3065 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3066 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3067 = 0s. The bug is otherwise harmless.
3069 PH/07 There was a bug in the dovecot authenticator such that the value of
3070 $auth1 could be overwritten, and so not correctly preserved, after a
3071 successful authentication. This usually meant that the value preserved by
3072 the server_setid option was incorrect.
3074 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3076 PH/09 Installed PCRE release 7.0.
3078 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3079 run for batched SMTP input. It is now run at the start of every message
3080 in the batch. While fixing this I discovered that the process information
3081 (output by running exiwhat) was not always getting set for -bs and -bS
3082 input. This is fixed, and it now also says "batched" for BSMTP.
3084 PH/11 Added control=no_pipelining.
3086 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3087 patch, slightly modified), and move the expansion of helo_data till after
3088 the connection is made in the smtp transport (so it can use these
3091 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3093 PH/14 Added log_selector = +pid.
3095 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3097 PH/16 Add ${if forany and ${if forall.
3099 PH/17 Added dsn_from option to vary the From: line in DSNs.
3101 PH/18 Flush SMTP output before performing a callout, unless control =
3102 no_callout_flush is set.
3104 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3105 was true (the default) a successful delivery failed to delete the retry
3106 item, thus causing premature timeout of the address. The bug is now
3109 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3111 PH/21 Long custom messages for fakedefer and fakereject are now split up
3112 into multiline responses in the same way that messages for "deny" and
3113 other ACL rejections are.
3115 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3116 with slight modification.
3118 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3119 draft, changing the syntax and factoring some duplicate code".
3121 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3122 for deliveries of the second and subsequent messages over the same SMTP
3125 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3126 ${reduce, with only minor "tidies".
3128 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3130 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3131 expansion side effects.
3133 PH/27 When a message times out after an over-quota error from an Exim-imposed
3134 quota, the bounce message says "mailbox is full". This message was not
3135 being given when it was a system quota that was exceeded. It now should
3138 MH/03 Made $recipients available in local_scan(). local_scan() already has
3139 better access to the recipient list through recipients_list[], but
3140 $recipients can be useful in postmaster-provided expansion strings.
3142 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3143 in the case of a MAIL command with additional options following the
3144 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3145 were accidentally chopped off.
3147 PH/29 SMTP synchronization checks are implemented when a command is read -
3148 there is a check that no more input is waiting when there shouldn't be
3149 any. However, for some commands, a delay in an ACL can mean that it is
3150 some time before the response is written. In this time, more input might
3151 arrive, invalidly. So now there are extra checks after an ACL has run for
3152 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3153 pipelining has not been advertised.
3155 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3157 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3158 returned in the SMTP response when smtp_return_error_details was false.
3159 This has been fixed.
3161 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3162 instead of the C I/O that was originally supplied, because problems were
3163 reported on Solaris.
3165 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3166 Exim which did not show up earlier: it was assuming that a call to
3167 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3168 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3169 was a macro that became an assignment, so it seemed to work. This has
3170 changed to a proper function call with a void return, hence the compile
3171 error. Exim's code has been fixed.
3173 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3176 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3178 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3180 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3181 because it turns out to be too big or there is a timeout) there is no
3182 "Completed" line in the log. When some messages of this type were
3183 selected by exigrep, they were listed as "not completed". Others were
3184 picked up by some special patterns. I have improved the selection
3185 criteria to be more general.
3187 PH/38 The host_find_failed option in the manualroute router can now be set
3188 to "ignore", to completely ignore a host whose IP address cannot be
3189 found. If all hosts are ignored, the behaviour is controlled by the new
3190 host_all_ignored option.
3192 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3193 homing or because of multiple MX records with /mx) generated more than
3194 one IP address, and the following item turned out to be the local host,
3195 all the secondary addresses of the first item were incorrectly removed
3196 from the list, along with the local host and any following hosts (which
3197 is what is supposed to happen).
3199 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3200 whoever called Exim into the -H file. In the case of the daemon it was
3201 behaving confusingly. When first started, it used values for whoever
3202 started the daemon, but after a SIGHUP it used the Exim user (because it
3203 calls itself on a restart). I have changed the code so that it now always
3206 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3207 message are rejected with the same error (e.g. no authentication or bad
3208 sender address), and a DATA command is nevertheless sent (as can happen
3209 with PIPELINING or a stupid MUA), the error message that was given to the
3210 RCPT commands is included in the rejection of the DATA command. This is
3211 intended to be helpful for MUAs that show only the final error to their
3214 PH/42 Another patch from the Sieve maintainer.
3216 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3217 Eximstats - Fixed some broken HTML links and added missing column headers
3219 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3220 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3222 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3224 PH/43 Yet another patch from the Sieve maintainer.
3226 PH/44 I found a way to check for a TCP/IP connection going away before sending
3227 the response to the final '.' that terminates a message, but only in the
3228 case where the client has not sent further data following the '.'
3229 (unfortunately, this is allowed). However, in many cases there won't be
3230 any further data because there won't be any more messages to send. A call
3231 to select() can be used: if it shows that the input is "ready", there is
3232 either input waiting, or the socket has been closed. An attempt to read
3233 the next input character can distinguish the two cases. Previously, Exim
3234 would have sent an OK response which the client would never have see.
3235 This could lead to message repetition. This fix should cure that, at
3236 least in a lot of common cases.
3238 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3239 advertised in response to EHLO.
3245 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3246 fixed by 4.65/MH/01 (is this a record?) are fixed:
3248 (i) An empty string was always treated as zero by the numeric comparison
3249 operators. This behaviour has been restored.
3251 (ii) It is documented that the numeric comparison operators always treat
3252 their arguments as decimal numbers. This was broken in that numbers
3253 starting with 0 were being interpreted as octal.
3255 While fixing these problems I realized that there was another issue that
3256 hadn't been noticed. Values of message_size_limit (both the global option
3257 and the transport option) were treated as octal if they started with 0.
3258 The documentation was vague. These values are now always treated as
3259 decimal, and I will make that clear in the documentation.
3265 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3266 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3269 MH/01 Don't check that the operands of numeric comparison operators are
3270 integers when their expansion is in "skipping" mode (fixes bug
3271 introduced by 4.64-PH/07).
3273 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3274 child addresses, Exim now panics and dies. Previously, because the count
3275 is held in a short int, deliveries were likely to be lost. As such a
3276 large number of recipients for a single message is ridiculous
3277 (performance will be very, very poor), I have chosen to impose a limit
3278 rather than extend the field.
3284 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3285 leftover -K file (the existence of which was triggered by #402).
3286 While we were at it, introduced process PID as part of the -K
3287 filename. This should rule out race conditions when creating
3290 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3291 processing considerably. Previous code took too long for large mails,
3292 triggering a timeout which in turn triggers #401.
3294 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3295 in the DK code in transports.c. sendfile() is not really portable,
3296 hence the _LINUX specificness.
3298 TF/01 In the add_headers option to the mail command in an Exim filter,
3299 there was a bug that Exim would claim a syntax error in any
3300 header after the first one which had an odd number of characters
3303 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3304 callout verification, Exim cached a "reject" for the entire domain. This
3305 is correct for most verifications, but it is not correct for a recipient
3306 verification with use_sender or use_postmaster set, because in that case
3307 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3308 case of MAIL FROM:<> rejection from other early rejections (e.g.
3309 rejection of HELO). When verifying a recipient using a non-null MAIL
3310 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3311 Whatever the result of the callout, the value of the domain cache is
3312 left unchanged (for any other kind of callout, getting as far as trying
3313 RCPT means that the domain itself is ok).
3315 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3318 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3319 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3320 session, the daemon ignores ECONNECT errors and logs others; it now
3321 ignores EPIPE as well.
3323 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3324 (quoted-printable decoding).
3326 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3327 later the small subsequent patch to fix an introduced bug.
3329 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3331 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3333 PH/08 An error is now given if message_size_limit is specified negative.
3335 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3336 to be given (somewhat) arbitrary names.
3338 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3341 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3342 miscellaneous code fixes
3344 PH/10 Added the log_reject_target ACL modifier to specify where to log
3347 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3348 hostname. This is wrong, because it relates to the incoming message (and
3349 probably the interface on which it is arriving) and not to the outgoing
3350 callout (which could be using a different interface). This has been
3351 changed to use the value of the helo_data option from the smtp transport
3352 instead - this is what is used when a message is actually being sent. If
3353 there is no remote transport (possible with a router that sets up host
3354 addresses), $smtp_active_hostname is used.
3356 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3357 tweaks were necessary in order to get it to work (see also 21 below):
3358 (a) The code assumed that strncpy() returns a negative number on buffer
3359 overflow, which isn't the case. Replaced with Exim's string_format()
3361 (b) There were several signed/unsigned issues. I just did the minimum
3362 hacking in of casts. There is scope for a larger refactoring.
3363 (c) The code used strcasecmp() which is not a standard C function.
3364 Replaced with Exim's strcmpic() function.
3365 (d) The code set only $1; it now sets $auth1 as well.
3366 (e) A simple test gave the error "authentication client didn't specify
3367 service in request". It would seem that Dovecot has changed its
3368 interface. Fortunately there's a specification; I followed it and
3369 changed what the client sends and it appears to be working now.
3371 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3374 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3375 address A is aliased to B and C, where B exists and C does not. Without
3376 -v the output is "A verified" because verification stops after a
3377 successful redirection if more than one address is generated. However,
3378 with -v the child addresses are also verified. Exim was outputting "A
3379 failed to verify" and then showing the successful verification for C,
3380 with its parentage. It now outputs "B failed to verify", showing B's
3381 parentage before showing the successful verification of C.
3383 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3384 look up a TXT record in a specific list after matching in a combined
3387 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3388 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3389 they consult the DNS. I had assumed they would set it the way they
3390 wanted; and indeed my experiments on Linux seem to show that in some
3391 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3392 To be on the safe side, however, I have now made the interface to
3393 host_find_byname() similar to host_find_bydns(), with an argument
3394 containing the DNS resolver options. The host_find_byname() function now
3395 sets these options at its start, just as host_find_bydns() does. The smtp
3396 transport options dns_qualify_single and dns_search_parents are passed to
3397 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3398 of host_find_byname() use the default settings of RES_DEFNAMES
3399 (qualify_single) but not RES_DNSRCH (search_parents).
3401 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3402 spool_read_header() do less string testing, by means of a preliminary
3403 switch on the second character of optional "-foo" lines. (This is
3404 overdue, caused by the large number of possibilities that now exist.
3405 Originally there were few.) While I was there, I also converted the
3406 str(n)cmp tests so they don't re-test the leading "-" and the first
3407 character, in the hope this might squeeze out yet more improvement.
3409 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3410 flag allowing group syntax was set by the header_syntax check but not
3411 turned off, possible causing trouble later; (2) The flag was not being
3412 set at all for the header_verify test, causing "group"-style headers to
3413 be rejected. I have now set it in this case, and also caused header_
3414 verify to ignore an empty address taken from a group. While doing this, I
3415 came across some other cases where the code for allowing group syntax
3416 while scanning a header line wasn't quite right (mostly, not resetting
3417 the flag correctly in the right place). These bugs could have caused
3418 trouble for malformed header lines. I hope it is now all correct.
3420 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3421 with the "reply" argument non-NULL. The code, however (which originally
3422 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3423 but it didn't always do it. This confused somebody who was copying the
3424 code for some other use. I have removed all the tests.
3426 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3427 feature that was used to support insecure browsers during the U.S. crypto
3428 embargo. It requires special client support, and Exim is probably the
3429 only MTA that supported it -- and would never use it because real RSA is
3430 always available. This code has been removed, because it had the bad
3431 effect of slowing Exim down by computing (never used) parameters for the
3432 RSA_EXPORT functionality.
3434 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3435 authenticator to fail if there's a tab character in the incoming data
3436 (there should never be unless someone is messing about, as it's supposed
3437 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3438 if the connection is using TLS or if the remote IP is the same as the
3439 local IP, and the "valid-client-cert option" if a client certificate has
3442 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3443 authenticators. This can be used for authorization after authentication
3444 succeeds. (In the case of plaintext, it servers for both authentication
3447 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3448 if any retry times were supplied.
3450 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3451 connection, where there is no client IP address to check. In this
3452 situation, the verify now always succeeds.
3454 PH/25 Applied John Jetmore's -Mset patch.
3456 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3458 PH/27 In a string expansion for a processed (not raw) header when multiple
3459 headers of the same name were present, leading whitespace was being
3460 removed from all of them, but trailing whitespace was being removed only
3461 from the last one. Now trailing whitespace is removed from each header
3462 before concatenation. Completely empty headers in a concatenation (as
3463 before) are ignored.
3465 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3466 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3468 PH/29 [Removed. This was a change that I later backed out, and forgot to
3469 correct the ChangeLog entry (that I had efficiently created) before
3470 committing the later change.]
3472 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3473 address errors (4xx response to RCPT) over the same connection as other
3474 messages routed to the same hosts. Such deliveries are always "forced",
3475 so retry times are not inspected. This resulted in far too many retries
3476 for the affected addresses. The effect occurred only when there were more
3477 hosts than the hosts_max_try setting in the smtp transport when it had
3478 the 4xx errors. Those hosts that it had tried were not added to the list
3479 of hosts for which the message was waiting, so if all were tried, there
3480 was no problem. Two fixes have been applied:
3482 (i) If there are any address or message errors in an SMTP delivery, none
3483 of the hosts (tried or untried) are now added to the list of hosts
3484 for which the message is waiting, so the message should not be a
3485 candidate for sending over the same connection that was used for a
3486 successful delivery of some other message. This seems entirely
3487 reasonable: after all the message is NOT "waiting for some host".
3488 This is so "obvious" that I'm not sure why it wasn't done
3489 previously. Hope I haven't missed anything, but it can't do any
3490 harm, as the worst effect is to miss an optimization.
3492 (ii) If, despite (i), such a delivery is accidentally attempted, the
3493 routing retry time is respected, so at least it doesn't keep
3494 hammering the server.
3496 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3497 in ${readsocket because some servers need this prod.
3499 PH/32 Added some extra debug output when updating a wait-xxx database.
3501 PH/33 The hint "could be header name not terminated by colon", which has been
3502 given for certain expansion errors for a long time, was not being given
3503 for the ${if def:h_colon_omitted{... case.
3505 PH/34 The spec says: "With one important exception, whenever a domain list is
3506 being scanned, $domain contains the subject domain." There was at least
3507 one case where this was not true.
3509 PH/35 The error "getsockname() failed: connection reset by peer" was being
3510 written to the panic log as well as the main log, but it isn't really
3511 panic-worthy as it just means the connection died rather early on. I have
3512 removed the panic log writing for the ECONNRESET error when getsockname()
3515 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3516 runs only) independently of the message's sender address. This meant
3517 that, if the 4xx error was in fact related to the sender, a different
3518 message to the same recipient with a different sender could confuse
3519 things. In particular, this can happen when sending to a greylisting
3520 server, but other circumstances could also provoke similar problems.
3521 I have changed the default so that the retry time for these errors is now
3522 based a combination of the sender and recipient addresses. This change
3523 can be overridden by setting address_retry_include_sender=false in the
3526 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3527 remote server are returned as part of bounce messages. This was not
3528 happening for LMTP over a pipe (the lmtp transport), but now it is the
3529 same for both kinds of LMTP.
3531 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3532 in header lines that were in fact CNAMEs. This is no longer the case.
3534 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3535 and queue runs started by the daemon processed all messages. This has
3536 been fixed so that -R and -S can now usefully be given with -q<time>.
3538 PH/40 Import PCRE release 6.7 (fixes some bugs).
3540 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3542 PH/42 Give an error if -q is specified more than once.
3544 PH/43 Renamed the variables $interface_address and $interface_port as
3545 $received_ip_address and $received_port, to make it clear that these
3546 values apply to message reception, and not to the outgoing interface when
3547 a message is delivered. (The old names remain recognized, of course.)
3549 PH/44 There was no timeout on the connect() call when using a Unix domain
3550 socket in the ${readsocket expansion. There now is.
3552 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3553 be meaningful with "accept".
3555 SC/01 Eximstats V1.43
3556 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3558 SC/02 Eximstats V1.44
3559 Use a glob alias rather than an array ref in the generated
3560 parser. This improves both readability and performance.
3562 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3563 Collect SpamAssassin and rejection statistics.
3564 Don't display local sender or destination tables unless
3565 there is data to show.
3566 Added average volumes into the top table text output.
3568 SC/04 Eximstats V1.46
3569 Collect data on the number of addresses (recipients)
3570 as well as the number of messages.
3572 SC/05 Eximstats V1.47
3573 Added 'Message too big' to the list of mail rejection
3574 reasons (thanks to Marco Gaiarin).
3576 SC/06 Eximstats V1.48
3577 Mainlog lines which have GMT offsets and are too short to
3578 have a flag are now skipped.
3580 SC/07 Eximstats V1.49 (Alain Williams)
3581 Added the -emptyok flag.
3583 SC/08 Eximstats V1.50
3584 Fixes for obtaining the IP address from reject messages.
3586 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3587 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3588 whitespace changes from 4.64-PH/27
3590 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3593 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3594 are found, allow negative numbers in numeric criteria)
3596 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3598 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3599 to match changes made in 4.64-PH/43
3601 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3603 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3604 "domain" is actually an IP address, and force a failure. This locks out
3605 those revolvers/nameservers that support "A-for-A" lookups, in
3606 contravention of the specifications.
3608 PH/48 When a host name was looked up from an IP address, and the subsequent
3609 forward lookup of the name timed out, the host name was left in
3610 $sender_host_name, contrary to the specification.
3612 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3613 restricted to single-key lookups, Exim was not diagnosing an error if
3614 * or *@ was used with a query-style lookup.
3616 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3618 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3619 long ago, but noone interested enough thought of it. Let's just say that
3620 the "1.1" means that there are some new functions that weren't there at
3621 some point in the past.
3623 PH/51 Error processing for expansion failure of helo_data from an smtp
3624 transport during callout processing was broken.
3626 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3627 tested/used via the -bh/-bhc/-bs options.
3629 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3630 bug, fixed in subsequent PCRE releases).
3632 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3633 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3635 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3641 SC/01 Use a glob alias rather than an array ref in eximstats generated
3642 parser. This improves both readability and performance.
3644 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3645 Don't display local sender or destination tables in eximstats unless
3646 there is data to show.
3647 Added average volumes into the eximstats top table text output.
3649 SC/03 Collect data on the number of addresses (recipients) as well
3650 as the number of messages in eximstats.
3652 TF/01 Correct an error in the documentation for the redirect router. Exim
3653 does (usually) call initgroups() when daemonizing.
3655 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3656 with consistent privilege compared to when running as a daemon.
3658 TF/03 Note in the spec that $authenticated_id is not set for local
3659 submissions from trusted users.
3661 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3662 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3664 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3665 by adding some example configuration directives to the default
3666 configuration file. A little bit of work is required to uncomment the
3667 directives and define how usernames and passwords are checked, but
3668 there is now a framework to start from.
3670 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3671 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3672 without this. I don't know how relevant this is to other LDAP libraries.
3674 PH/02 Add the verb name to the "unknown ACL verb" error.
3676 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3678 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3680 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3681 directory not expanded when it should be if an expanded home directory
3682 was set for the address (which is overridden by the transport).
3684 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3687 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3688 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3689 because it is too late at that time, and has no effect.
3691 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3692 security issue with \' (bugzilla #107). I could not use the
3693 PQescapeStringConn() function, because it needs a PGconn value as one of
3696 PH/08 When testing addresses using -bt, indicate those final addresses that
3697 are duplicates that would not cause an additional delivery. At least one
3698 person was confused, thinking that -bt output corresponded to deliveries.
3699 (Suppressing duplicates isn't a good idea as you lose the information
3700 about possibly different redirections that led to the duplicates.)
3702 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3703 systems where poll() doesn't work, in particular OS X.
3705 PH/10 Added more information to debugging output for retry time not reached.
3707 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3708 operations in malware.c.
3710 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3713 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3714 syslog_duplication set false, log lines that would normally be written
3715 both the the main log and to the reject log were not written to syslog at
3718 PH/14 In the default configuration, change the use of "message" in ACL warn
3719 statements to "add_header".
3721 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3722 not followed by a command (e.g. "seen endif").
3724 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3725 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3728 PH/17 Added extra conditions to the default value of delay_warning_condition
3732 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3733 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3734 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3737 The Auto-Submitted: and various List- headers are standardised, whereas I
3738 don't think Precedence: ever was.
3740 PH/18 Refactored debugging code in route_finduser() to show more information,
3741 in particular, the error code if getpwnam() issues one.
3743 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3744 This is apparently needed in addition to the PH/07 change above to avoid
3745 any possible encoding problems.
3747 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3748 but not after initializing Perl.
3750 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3751 output them only if debugging. By default they are written stderr,
3752 apparently, which is not desirable.
3754 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3757 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3760 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3762 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3763 authenticated or an ident call has been made. Suppress the default
3764 values for $authenticated_id and $authenticated_sender (but permit -oMai
3765 and -oMas) when testing with -bh.
3767 PH/24 Re-jigged the order of the tests in the default configuration so that the
3768 tests for valid domains and recipients precede the DNS black list and CSA
3769 tests, on the grounds that those ones are more expensive.
3771 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3772 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3773 This bug exists in every version of Exim that I still have, right back to
3776 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3777 However, an attempt to turn on case-sensitivity in a regex key by
3778 including (?-i) didn't work because the subject string was already
3779 lowercased, and the effects were non-intuitive. It turns out that a
3780 one-line patch can be used to allow (?-i) to work as expected.
3786 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3787 other effects) broke the use of negated acl sub-conditions.
3789 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3792 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3793 "Deny" causes Exim to reject the incoming connection with a 554 error.
3794 Unfortunately, if there is a major crisis, such as a disk failure,
3795 tcp-wrappers gives "deny", whereas what one would like would be some
3796 kind of temporary error. A kludge has been added to help with this.
3797 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3798 554 error is used if errno is still zero or contains ENOENT (which occurs
3799 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3802 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3804 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3805 errors. Otherwise a message that provokes a temporary error (when other
3806 messages do not) can cause a whole host to time out.
3808 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3809 addresses were routed directly to files or pipes from a redirect router.
3810 File deliveries just didn't batch; pipe deliveries might have suffered
3813 PH/06 A failure to get a lock for a hints database would erroneously always say
3814 "Failed to get write lock", even when it was really a read lock.
3816 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3817 of 0600. This has been changed to use the value of the lockfile_mode
3818 option (which defaults to 0600).
3820 PH/08 Applied small patch from the Sieve maintainer.
3822 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3823 folder from quota calculations, a direct delivery into this folder messed
3824 up the contents of the maildirsize file. This was because the regex was
3825 used only to exclude .Trash (or whatever) when the size of the mailbox
3826 was calculated. There was no check that a delivery was happening into an
3827 excluded directory. This bug has been fixed by ignoring all quota
3828 processing for deliveries into excluded directories.
3830 PH/10 Added the maildirfolder_create_regex option to appendfile.
3836 PH/01 The code for finding all the local interface addresses on a FreeBSD
3837 system running IPv6 was broken. This may well have applied to all BSD
3838 systems, as well as to others that have similar system calls. The broken
3839 code found IPv4 interfaces correctly, but gave incorrect values for the
3840 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3841 that it would not match correctly against @[] and not recognize the IPv6
3844 PH/02 The ipliteral router was not recognizing addresses of the form user@
3845 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3847 PH/03 Added disable_ipv6.
3849 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3850 decoded form, because it is most often used to construct To: headers
3851 lines in autoreplies, and the decoded form may well be syntactically
3852 invalid. However, $reply_address has leading white space removed, and all
3853 newlines turned into spaces so that the autoreply transport does not
3856 PH/05 If group was specified without a user on a router, and no group or user
3857 was specified on a transport, the group from the router was ignored.
3859 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3860 for visible compile-time settings that can be used to change these
3861 numbers, for those that want even more. Backwards compatibility with old
3862 spool files has been maintained. However, going back to a previous Exim
3863 release will lost any variables that are in spool files.
3865 PH/07 Two small changes when running in the test harness: increase delay when
3866 passing a TCP/IP connection to a new process, in case the original
3867 process has to generate a bounce, and remove special handling of
3868 127.0.0.2 (sic), which is no longer necessary.
3870 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3871 be the same on different OS.
3873 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3876 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3877 whether --show-vars was specified or not
3879 JJ/02 exipick: Added support for new ACL variable spool format introduced
3882 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3883 syntactically invalid From: or Reply-to: line, and a filter used this to
3884 generate an autoreply, and therefore failed to obtain an address for the
3885 autoreply, Exim could try to deliver to a non-existent relative file
3886 name, causing unrelated and misleading errors. What now happens is that
3887 it logs this as a hard delivery error, but does not attempt to create a
3890 PH/11 The exinext utility has a -C option for testing purposes, but although
3891 the given file was scanned by exinext itself; it wasn't being passed on
3892 when Exim was called.
3894 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3895 an end-of-file indication when reading a command response.
3897 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3898 compiled. In many other places in Exim, IPv6 addresses are always
3899 recognized, so I have changed this. It also means that IPv4 domain
3900 literals of the form [IPV4:n.n.n.n] are now always recognized.
3902 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3903 used if the router is not running as root, for example, when verifying at
3904 ACL time, or when using -bh. The debugging output from this situation was
3905 non-existent - all you got was a failure to exec. I have made two
3908 (a) Failures to set uid/gid, the current directory, or a process leader
3909 in a subprocess such as that created by queryprogram now generate
3910 suitable debugging output when -d is set.
3912 (b) The queryprogram router detects when it is not running as root,
3913 outputs suitable debugging information if -d is set, and then runs
3914 the subprocess without attempting to change uid/gid.
3916 PH/15 Minor change to Makefile for building test_host (undocumented testing
3919 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3920 additional section of a DNS packet that returns MX or SRV records.
3921 Instead, it always explicitly searches for A/AAAA records. This avoids
3922 major problems that occur when a DNS server includes only records of one
3923 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3924 fixed another bug: if SRV records were looked up and the corresponding
3925 address records were *not* found in the additional section, the port
3926 values from the SRV records were lost.
3928 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3929 using the correct key (the original address) when searching the retry
3930 rules in order to find which one to use for generating the retry hint.
3932 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3933 adding the default one. Similarly, if it contains a Reply-To: header, the
3934 errors_reply_to option, if set, is not used.
3936 PH/19 When calculating a retry time, Exim used to measure the "time since
3937 failure" by looking at the "first failed" field in the retry record. Now
3938 it does not use this if it is later than than the arrival time of the
3939 message. Instead it uses the arrival time. This makes for better
3940 behaviour in cases where some deliveries succeed, thus re-setting the
3941 "first failed" field. An example is a quota failure for a huge message
3942 when small messages continue to be delivered. Without this change, the
3943 "time since failure" will always be short, possible causing more frequent
3944 delivery attempts for the huge message than are intended.
3945 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3947 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3948 $1, $2, $3) because the numerical variables can be reset during some
3949 expansion items (e.g. "match"), thereby losing the authentication data.
3951 PH/21 Make -bV show the size of off_t variables so that the test suite can
3952 decide whether to run tests for quotas > 2G.
3954 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3955 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3956 filecount value is greater than 2G or if a quota value is greater than 2G
3957 on a system where the size of off_t is not greater than 4, a panic error
3960 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3961 never match. The debug and -bh output now contains an explicit error
3962 message indicating a malformed IPv4 address or mask.
3964 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3965 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3966 PH/23 above applies.
3968 PH/25 Do not write to syslog when running in the test harness. The only
3969 occasion when this arises is a failure to open the main or panic logs
3970 (for which there is an explicit test).
3972 PH/26 Added the /no_tell option to "control=freeze".
3974 PH/27 If a host name lookup failed very early in a connection, for example, if
3975 the IP address matched host_lookup and the reverse lookup yielded a name
3976 that did not have a forward lookup, an error message of the form "no IP
3977 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3978 could be logged. Now it outputs the IP address instead of "NULL".
3980 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3981 allows the sender and the authenticated sender to be set when
3982 submitting a message from within Exim. Since child_open_exim() is
3983 documented for local_scan(), the new function should be too.
3985 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3986 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3987 results in an empty string is now treated as unset.
3989 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3991 PH/31 Added sender_verify_fail logging option.
3993 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3994 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3995 tidied the source and removed it altogether.
3997 PH/33 When a queue run was abandoned because the load average was too high, a
3998 log line was always written; now it is written only if the queue_run log
3999 selector is set. In addition, the log line for abandonment now contains
4000 information about the queue run such as the pid. This is always present
4001 in "start" and "stop" lines but was omitted from the "abandon" line.
4003 PH/34 Omit spaces between a header name and the colon in the error message that
4004 is given when verify = headers_syntax fails (if there are lots of them,
4005 the message gets confusing).
4007 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4008 names, as there are now some PTR records that contain slashes. This check
4009 is only to protect against broken name servers that fall over on strange
4010 characters, so the fact that it applies to all lookups doesn't matter.
4012 PH/36 Now that the new test suite is complete, we can remove some of the
4013 special code in Exim that was needed for the old test suite. For example,
4014 sorting DNS records because real resolvers return them in an arbitrary
4015 order. The new test suite's fake resolver always returns records in the
4018 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4019 bounces) except when queue_only is set, to avoid logging races between
4020 the different processes.
4022 PH/38 Panic-die if .include specifies a non-absolute path.
4024 PH/39 A tweak to the "H" retry rule from its user.
4026 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4027 a label. They prevented compilation on older perls.
4029 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4030 a warning to be raised on newish perls.
4032 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4033 on queue. Changes to match documented behaviour of showing count of
4034 messages matching specified criteria.
4036 PH/40 Changed the default ident timeout from 30s to 5s.
4038 PH/41 Added support for the use of login_cap features, on those BSD systems
4039 that have them, for controlling the resources used by pipe deliveries.
4041 PH/42 The content-scanning code uses fopen() to create files in which to put
4042 message data. Previously it was not paying any attention to the mode of
4043 the files. Exim runs with umask(0) because the rest of the code creates
4044 files with open(), and sets the required mode explicitly. Thus, these
4045 files were ending up world-writeable. This was not a big issue, because,
4046 being within the spool directory, they were not world-accessible. I have
4047 created a function called modefopen, which takes an additional mode
4048 argument. It sets umask(777), creates the file, chmods it to the required
4049 mode, then resets the umask. All the relevant calls to fopen() in the
4050 content scanning code have been changed to use this function.
4052 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4053 to 24 hours. This avoids potential overflow problems when processing G
4054 and H retry rules. I suspect nobody ever tinkers with this value.
4056 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4058 PH/45 When the plaintext authenticator is running as a client, the server's
4059 challenges are checked to ensure they are valid base64 strings. By
4060 default, the authentication attempt is cancelled if an invalid string is
4061 received. Setting client_ignore_invalid_base64 true ignores these errors.
4062 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4063 they are received. Thus, the responses can be made to depend on the
4064 challenges. If an invalid string is ignored, an empty string is placed in
4067 PH/46 Messages that are created by the autoreply transport now contains a
4068 References: header, in accordance with RFCs 2822 and 3834.
4070 PH/47 Added authenticated_sender_force to the smtp transport.
4072 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4074 PH/49 Installed latest patch from the Sieve maintainer.
4076 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4077 was also set, the appendfile transport was unnecessarily scanning a
4078 directory of message files (e.g. for maildir delivery) to find the count
4079 of files (along with the size), even though it did not need this
4080 information. It now does the scan only if it needs to find either the
4081 size of the count of files.
4083 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4085 PH/52 Two bugs concerned with error handling when the smtp transport is
4088 (i) Exim was not creating retry information for temporary errors given
4089 for individual recipients after the DATA command when the smtp transport
4090 was used in LMTP mode. This meant that they could be retried too
4091 frequently, and not timed out correctly.
4093 (ii) Exim was setting the flag that allows error details to be returned
4094 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4095 recipients that were returned after the DATA command.
4097 PH/53 This is related to PH/52, but is more general: for any failing address,
4098 when detailed error information was permitted to be returned to the
4099 sender, but the error was temporary, then after the final timeout, only
4100 "retry timeout exceeded" was returned. Now it returns the full error as
4101 well as "retry timeout exceeded".
4103 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4104 do this, and (what is worse) MTAs that accept it.
4106 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4107 will now be deprecated.
4109 PH/56 New os.c-cygwin from the Cygwin maintainer.
4111 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4112 formats (previously only available in exim formats via -bpr, -bpru,
4113 and -bpra. Now also available in native and exiqgrep formats)
4115 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4116 with very large, slow to parse queues
4118 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4120 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4122 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4123 responses to authentication challenges, though it was showing the
4124 challenges; (ii) I've removed the CR characters from the debug output for
4127 PH/58 Allow for the insertion of a newline as well as a space when a string
4128 is turned into more than one encoded-word during RFC 2047 encoding. The
4129 Sieve code now uses this.
4131 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4132 data_4xx, lost_connection, tls_required.
4134 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4135 message was being sent as an SMTP response.
4137 PH/61 Add -l and -k options to exicyclog.
4139 PH/62 When verifying, if an address was redirected to one new address, so that
4140 verification continued, and the new address failed or deferred after
4141 having set something in $address_data, the value of $address_data was not
4142 passed back to the ACL. This was different to the case when no
4143 redirection occurred. The value is now passed back in both cases.
4145 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4146 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4147 use login_cap.h, so on its own it isn't the distinguishing feature. The
4148 new name refers directly to the setclassresources() function.
4150 PH/65 Added configuration files for NetBSD3.
4152 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4154 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4155 is preferred over IPv4.
4157 PH/68 The bounce_return_message and bounce_return_body options were not being
4158 honoured for bounces generated during the reception of non-SMTP messages.
4159 In particular, this applied to messages rejected by the ACL. This bug has
4160 been fixed. However, if bounce_return_message is true and bounce_return_
4161 body is false, the headers that are returned for a non-SMTP message
4162 include only those that have been read before the error was detected.
4163 (In the case of an ACL rejection, they have all been read.)
4165 PH/69 The HTML version of the specification is now built in a directory called
4166 spec_html instead of spec.html, because the latter looks like a path with
4167 a MIME-type, and this confuses some software.
4169 PH/70 Catch two compiler warnings in sieve.c.
4171 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4172 function verify_get_ident() calls ip_connect() to connect a socket, but
4173 if the "connect()" function timed out, ip_connect() used to close the
4174 socket. However, verify_get_ident() also closes the socket later, and in
4175 between Exim writes to the log, which may get opened at this point. When
4176 the socket was closed in ip_connect(), the log could get the same file
4177 descriptor number as the socket. This naturally causes chaos. The fix is
4178 not to close the socket in ip_connect(); the socket should be closed by
4179 the function that creates it. There was only one place in the code where
4180 this was missing, in the iplookup router, which I don't think anybody now
4181 uses, but I've fixed it anyway.
4183 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4184 well as to direct DNS lookups. Otherwise the handling of names in host
4185 lists is inconsistent and therefore confusing.
4191 PH/01 Two changes to the default runtime configuration:
4193 (1) Move the checks for relay_from_hosts and authenticated clients from
4194 after to before the (commented out) DNS black list checks.
4196 (2) Add control=submission to the relay_from_hosts and authenticated
4197 clients checks, on the grounds that messages accepted by these
4198 statements are most likely to be submissions.
4200 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4202 (1) Generate an error if the third argument for the ${prvs expansion is
4205 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4208 (3) Reset the variables that are obtained from the first argument of
4209 ${prvscheck and used in the second argument before leaving the code,
4210 because their memory is reclaimed, so using them afterwards may do
4213 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4214 one (it's much easier than Tom thought :-).
4216 (5) Because of (4), we can now allow for the use of $prvscheck_result
4217 inside the third argument.
4219 PH/03 For some reason, the default setting of PATH when running a command from
4220 a pipe transport was just "/usr/bin". I have changed it to
4223 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4224 anything to be listed in the output from -bV.
4226 PH/05 When a filter generated an autoreply, the entire To: header line was
4227 quoted in the delivery log line, like this:
4229 => >A.N.Other <ano@some.domain> <original@ddress> ...
4231 This has been changed so that it extracts the operative address. There
4232 may be more than one such address. If so, they are comma-separated, like
4235 => >ano@some.domain,ona@other.domain <original@ddress> ...
4237 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4238 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4239 not being looked up in the rDNS to get a host name, Exim was showing the
4240 IP address twice in Received: lines, even though the IP addresses were
4241 identical. For example:
4243 Received: from [1.2.3.4] (helo=[1.2.3.4])
4245 However, if the real host name was known, it was omitting the HELO data
4246 if it matched the actual IP address. This has been tidied up so that it
4247 doesn't show the same IP address twice.
4249 PH/07 When both +timestamp and +memory debugging was on, the value given by
4250 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4251 called by the debug printing, thereby overwriting the timestamp buffer.
4252 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4254 PH/08 When the original message was included in an autoreply transport, it
4255 always said "this is a copy of the message, including all the headers",
4256 even if body_only or headers_only was set. It now gives an appropriate
4259 PH/09 Applied a patch from the Sieve maintainer which:
4261 o fixes some comments
4262 o adds the (disabled) notify extension core
4263 o adds some debug output for the result of if/elsif tests
4264 o points to the current vacation draft in the documentation
4265 and documents the missing references header update
4269 o fixes a bug in processing the envelope test (when testing
4270 multiple envelope elements, the last element determined the
4273 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4274 Electronic Mail") by including:
4276 Auto-submitted: auto-generated
4278 in the messages that it generates (bounce messages and others, such as
4279 warnings). In the case of bounce messages for non-SMTP messages, there was
4280 also a typo: it was using "Auto_submitted" (underscore instead of
4281 hyphen). Since every message generated by Exim is necessarily in response
4282 to another message, thes have all been changed to:
4284 Auto-Submitted: auto-replied
4286 in accordance with these statements in the RFC:
4288 The auto-replied keyword:
4290 - SHOULD be used on messages sent in direct response to another
4291 message by an automatic process,
4293 - MUST NOT be used on manually-generated messages,
4295 - MAY be used on Delivery Status Notifications (DSNs) and Message
4296 Disposition Notifications (MDNs),
4298 - MUST NOT be used on messages generated by automatic or periodic
4299 processes, except for messages which are automatic responses to
4302 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4303 to the default Received: header definition.
4305 PH/12 Added log selector acl_warn_skipped (default on).
4307 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4308 variables because (a) they are in the wrong storage pool and (b) even if
4309 they were copied, it wouldn't work properly because of the caching.
4311 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4312 checking when decoding. Apparently there are clients that generate
4313 overlong encoded strings. Why am I not surprised?
4315 PH/15 If the first argument of "${if match_address" was not empty, but did not
4316 contain an "@" character, Exim crashed. Now it writes a panic log message
4317 and treats the condition as false.
4319 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4321 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4322 extension "envelope-auth". The code is finished and in agreement with
4323 other implementations, but there is no documentation so far and in fact,
4324 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4325 not changing the active code.
4327 Print executed "if" and "elsif" statements when debugging is used. This
4328 helps a great deal to understand what a filter does.
4330 Document more things not specified clearly in RFC3028. I had all this
4331 sorted out, when out of a sudden new issues came to my mind. Oops."
4333 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4336 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4337 canonical form (as documented). However, after a host name lookup from
4338 the IP address, check_host() was doing a simple string comparison with
4339 addresses acquired from the DNS when checking that the found name did
4340 have the original IP as one of its addresses. Since any found IPv6
4341 addresses are likely to be in abbreviated form, the comparison could
4342 fail. Luckily, there already exists a function for doing the comparison
4343 by converting both addresses to binary, so now that is used instead of
4344 the text comparison.
4346 PH/20 There was another similar case to PH/19, when a complete host name was
4347 given in a host list; looking up its IP address could give an abbreviated
4348 form, whereas the current host's name might or might not be abbreviated.
4349 The same fix has been applied.
4355 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4356 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4359 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4360 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4362 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4364 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4365 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4366 Also, for "Auto-Submitted", treat anything other than "no" as
4367 non-personal, in accordance with RFC 3834. (Previously it treated
4368 anything starting "auto-" as non-personal.)
4370 TF/01 The control=submission/name=... option had a problem with syntax
4371 errors if the name included a slash character. The /name= option
4372 now slurps the rest of the string, so it can include any characters
4373 but it must come last in the list of options (after /sender_retain
4376 PH/05 Some modifications to the interface to the fake nameserver for the new
4384 TK/01 Added the "success_on_redirect" address verification option. See
4385 NewStuff for rationale and an example.
4387 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4389 PH/02 Patch to exigrep to allow it to work on syslog lines.
4391 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4392 fread() to skip over the body file's header line, because in Cygwin the
4393 header line is locked and is inaccessible.
4395 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4396 co-exist for some time) to make it clear that it is the Exim ID that is
4397 referenced, not the Message-ID: header line.
4399 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4400 string_format() function, because snprintf() does not exist on all
4403 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4404 ${stat: expansion item.
4406 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4407 protocol synchronization error", to keep the pedants happy.
4409 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4410 well as for IRIX systems, when gcc is being used. See the host.c source
4413 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4415 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4418 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4419 transport and to the smtp transport in LMTP mode.
4421 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4423 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4424 run a filter in a subprocess. This could lead to confusion in subsequent
4425 lookups in the parent process. There should also be a search_tidyup() at
4426 the end of the subprocess.
4428 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4429 only if the host matched helo_try_verify_hosts, which caused the
4430 verification to occur when the EHLO/HELO command was issued. The ACL just
4431 tested the remembered result. Now, if a previous verification attempt has
4432 not happened, "verify = helo" does it there and then.
4434 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4436 TK/03 Fix log output including CR from clamd.
4438 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4439 exist provoked a memory error which could cause a segfault.
4441 PH/15 Installed PCRE 6.2
4443 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4445 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4446 of the problem. Specifically, suggested +O2 rather than +O1 for the
4449 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4451 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4452 record, the port was not correctly logged when the outgoing_port log
4453 selector was set (it logged the transort's default port).
4455 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4456 fallback_hosts, and "hosts" in the smtp transport.
4458 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4459 host errors such as "Connection refused".
4461 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4462 authentication with radiusclient 0.4.9:
4464 - Error returned from rc_read_config was caught wrongly
4465 - Username/password not passed on to radius server due to wrong length.
4467 The presumption is that some radiusclient API changes for 4.51/PH/17
4468 were not taken care of correctly. The code is still untested by me (my
4469 Linux distribution still has 0.3.2 of radiusclient), but it was
4470 contributed by a Radius user.
4472 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4473 expanding the "port" option of the smtp transport.
4475 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4476 while reading a MIME header. Thanks to Tom Hughes for a patch.
4478 PH/24 Include config.h inside local_scan.h so that configuration settings are
4481 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4482 that in an ACL for RCPT (for example), you can examine exactly what was
4485 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4486 commands, but it was not correctly comparing the address with the actual
4487 client host address. Thus, it would show the EHLO address in Received:
4488 header lines when this was not necessary.
4490 PH/27 Added the % operator to ${eval:}.
4492 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4493 it should be ignoring failures (because with -C, for example, it has lost
4494 privilege). It wasn't ignoring creation failures other than "already
4497 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4500 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4501 because an input line was too long, either on its own, or by virtue of
4502 too many continuations, the temporary file was not being removed, and the
4503 return code was incorrect.
4505 PH/31 Missing "BOOL" in function definition in filtertest.c.
4507 PH/32 Applied Sieve patches from the maintainer.
4509 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4511 PH/33 Added "verify = not_blind".
4513 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4514 Local/Makefile (with some defaults set). These are used in built scripts
4515 such as exicyclog, but they have never been used in the exim_install
4516 script (though there are many overriding facilities there). I have
4517 arranged that the exim_install script now takes note of these two
4520 PH/35 Installed configuration files for Dragonfly.
4522 PH/36 When a locally submitted message by a trusted user did not contain a
4523 From: header, and the sender address was obtained from -f or from an SMTP
4524 MAIL command, and the trusted user did not use -F to supply a sender
4525 name, $originator_name was incorrectly used when constructing a From:
4526 header. Furthermore, $originator_name was used for submission mode
4527 messages from external hosts without From: headers in a similar way,
4528 which is clearly wrong.
4530 PH/37 Added control=suppress_local_fixups.
4532 PH/38 When log_selector = +received_sender was set, and the addition of the
4533 sender made the log line's construction buffer exactly full, or one byte
4534 less than full, an overflow happened when the terminating "\n" was
4537 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4538 when the result of a list match is failure because a DNS lookup failed.
4540 PH/40 RM_COMMAND is now used in the building process.
4542 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4543 the "build-* directories that it finds.
4545 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4546 address in a domain literal was a prefix of an interface address.
4548 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4549 when verifying a sender address, unless rewrite_headers is false.
4551 PH/44 Wrote a long comment about why errors_to addresses are verified as
4552 recipients, not senders.
4554 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4555 the ratelimit ACL was added.
4557 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4559 PH/46 Added extra information about PostgreSQL errors to the error string.
4561 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4562 avoiding the need to install special zones in a real server. This is
4563 backwards compatible; if it can't find the fake resolver, it drops back.
4564 Thus, both old and new test suites can be run.
4566 TF/02 Added util/ratelimit.pl
4568 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4569 clock is set back in time.
4571 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4572 Candler <B.Candler@pobox.com>.
4574 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4575 OK for addresses that are the result of redirections.
4577 PH/48 A number of further additions for the benefit of the new test suite,
4578 including a fake gethostbyname() that interfaces to the fake DNS resolver
4581 TF/06 The fix for widen_domains has also been applied to qualify_single and
4582 search_parents which are the other dnslookup options that can cause
4585 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4588 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4590 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4591 service" if it encountered an MX record with an empty target hostname.
4592 The message is now "an MX or SRV record indicated no SMTP service".
4594 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4595 if the DNS of the sending site is misconfigured. This is quite a
4596 common situation. This change restores the behaviour of treating a
4597 helo verification defer as a failure.
4599 PH/51 If self=fail was set on a router, the bounce message did not include the
4600 actual error message.
4606 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4608 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4609 command itself ended in error, the underlying message about the transport
4610 filter timeout was being overwritten with the pipe command error. Now the
4611 underlying error message should be appended to the second error message.
4613 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4615 PH/02 Reduce the amount of output that "make" produces by default. Full output
4616 can still be requested.
4618 PH/03 The warning log line about a condition test deferring for a "warn" verb
4619 was being output only once per connection, rather than after each
4620 occurrence (because it was using the same function as for successful
4621 "warn" verbs). This seems wrong, so I have changed it.
4623 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4624 it should not have, which might have caused a crash in the right
4625 circumstances, but probably never did.
4627 PH/04 Installed a modified version of Tony Finch's patch to make submission
4628 mode fix the return path as well as the Sender: header line, and to
4629 add a /name= option so that you can make the user's friendly name appear
4632 TF/03 Added the control = fakedefer ACL modifier.
4634 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4635 Mark Lowes for thorough testing.
4637 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4639 TK/03 Merged latest SRS patch from Miles Wilton.
4641 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4642 with the definition in sysexits.h (which is #included earlier).
4643 Fortunately, Exim does not actually use EX_OK. The code used to try to
4644 preserve the sysexits.h value, by assuming that macro definitions were
4645 scanned for macro replacements. I have been disabused of this notion,
4646 so now the code just undefines EX_OK before #including unistd.h.
4648 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4649 in the smtp transport. When a block could not be written in a single
4650 write() function, the timeout was being re-applied to each part-write.
4651 This seems wrong - if the receiver was accepting one byte at a time it
4652 would take for ever. The timeout is now adjusted when this happens. It
4653 doesn't have to be particularly precise.
4655 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4656 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4658 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4659 without a domain if the check to <postmaster@domain> fails.
4661 SC/01 Eximstats: added -xls and the ability to specify output files
4662 (patch written by Frank Heydlauf).
4664 SC/02 Eximstats: use FileHandles for outputting results.
4666 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4668 SC/04 Eximstats: fixed display of large numbers with -nvr option
4670 SC/05 Eximstats: fixed merging of reports with empty tables.
4672 SC/06 Eximstats: added the -include_original_destination flag
4674 SC/07 Eximstats: removed tabs and trailing whitespace.
4676 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4678 TK/06 MBOX spool code: Add real "From " MBOX separator line
4679 so the .eml file is really in mbox format (even though
4680 most programs do not really care). Patch from Alex Miller.
4682 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4683 The latter is generated from $received_to and is only set if the
4684 message has one envelope recipient. SA can use these headers,
4685 obviously out-of-the-box. Patch from Alex Miller.
4687 PH/08 The ${def test on a variable was returning false if the variable's
4688 value was "0", contrary to what the specification has always said!
4689 The result should be true unless the variable is empty.
4691 PH/09 The syntax error of a character other than { following "${if
4692 def:variable_name" (after optional whitespace) was not being diagnosed.
4693 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4694 accidental colon was present, for example, could give incorrect results.
4696 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4697 result is used (not including appendfile, where other changes are about
4700 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4701 This involved changing a lot of size variables from int to off_t. It
4702 should work with maildirs and everything.
4704 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4705 spamd dying while we are connected to it.
4707 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4710 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4711 function for BDB 4.3.
4713 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4715 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4716 the sources for PCRE have been re-arranged and more files are now
4719 PH/15 The code I had for printing potentially long long variables in PH/11
4720 above was not the best (it lost precision). The length of off_t variables
4721 is now inspected at build time, and an appropriate printing format (%ld
4722 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4723 to be "long long int" or "long int". This is needed for the internal
4724 formatting function string_vformat().
4726 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4727 the configuration file to be ":syslog", then the script "guesses" where
4728 the logs files are, rather than using the compiled in default. In our
4729 case the guess is not the same as the compiled default, so the script
4730 suddenly stopped working when I started to use syslog. The patch checks
4731 to see if log_file_path is "". If so, it attempts to read it from exim
4732 with no configuration file to get the compiled in version, before it
4733 falls back to the previous guessing code."
4735 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4736 implementing BATV in an Exim configuration. See NewStuff for the gory
4739 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4740 Makefile that are specific to HP-UX.
4742 PH/18 If the "use_postmaster" option was set for a recipient callout together
4743 with the "random" option, the postmaster address was used as the MAIL
4744 FROM address for the random test, but not for the subsequent recipient
4745 test. It is now used for both.
4747 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4748 patch removes a few documentation additions to RFC 3028, because the
4749 latest draft now contains them. It adds the new en;ascii-case comparator
4750 and a new error check for 8bit text in MIME parts. Comparator and
4751 require names are now matched exactly. I enabled the subaddress
4752 extension, but it is not well tested yet (read: it works for me)."
4754 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4755 rework some of the code of TK/09 above to avoid the hardwired use of
4756 "%lld" and "long long". Replaced the call to snprintf() with a call to
4759 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4760 records point to non-existent hosts", "retry timeout exceeded", and
4761 "retry time not reached for any host after a long failure period".
4763 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4764 experimental DomainKeys support:
4766 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4767 (2) On an error such as an illegally used "control", the wrong name for
4768 the control was given.
4770 These problems did NOT occur unless DomainKeys support was compiled.
4772 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4774 PH/24 Added ${if match_ip condition.
4776 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4777 databases so that it will be absolutely obvious if a crash occurs in the
4778 DB library. This is a regular occurrence (often caused by mis-matched
4781 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4782 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4783 picked up on a user's system that detects such things. There doesn't seem
4784 to be a gcc warning option for this - only an attribute that has to be
4785 put on the function's prototype. It seems that in Fedora Core 4 they have
4786 set this on a number of new functions. No doubt there will be more in due
4789 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4790 specify a transport. However, if an address that was verified by such a
4791 router was the subject of a callout, Exim crashed because it tried to
4792 read the rcpt_include_affixes from the non-existent transport. Now it
4793 just assumes that the setting of that option is false. This bug was
4794 introduced by 4.51/PH/31.
4796 PH/28 Changed -d+all to exclude +memory, because that information is very
4797 rarely of interest, but it makes the output a lot bigger. People tend to
4798 do -d+all out of habit.
4800 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4801 code in os-type was giving problems when libc.so lives in lib64, like on
4804 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4805 aren't the modern standard, and it seems that some systems' include files
4806 don't always have them. Exim was already checking for some of the newer
4807 ones like T_AAAA, and defining it itself. I've added checks for all the
4808 record types that Exim uses.
4810 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4811 not automatically generating a new one, as it is supposed to. This
4812 prevented TLS from working. If the file did exist, but contained invalid
4813 data, a new version was generated, as expected. It was only the case of a
4814 non-existent file that was broken.
4816 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4817 with a change in libdomainkeys > 0.64.
4819 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4820 from DNS. If the selector record carries the flag, it now has
4821 precedence over the domain-wide flag.
4823 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4825 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4826 the use of a port name that isn't defined in /etc/services) occurred, the
4827 message was deferred as in a normal delivery, and thus remained on the
4828 spool, instead of being failed because of the mua_wrapper setting. This
4829 is now fixed, and I tidied up some of the mua_wrapper messages at the
4832 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4833 the messages in a hash of arrays rather than using individual hashes.
4834 This is a bit cleaner and results in dramatic memory savings, albeit
4835 at a slight CPU cost.
4837 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4838 as requested by Marc Sherman.
4840 SC/10 Eximstats: added histograms for user specified patterns as requested
4843 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4845 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4846 fopen() in the content-scanning modules that did not already have it.
4852 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4853 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4855 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4857 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4859 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4860 to test Sieve filters that use "vacation".
4862 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4863 that changes the way the GnuTLS parameters are stored in the cache file.
4864 The new format can be generated externally. For backward compatibility,
4865 if the data in the cache doesn't make sense, Exim assumes it has read an
4866 old-format file, and it generates new data and writes a new file. This
4867 means that you can't go back to an older release without removing the
4870 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4871 work if there are any delivery delays because "one_time" forces the
4872 parent to be marked "delivered", so its unseen clone is never tried
4873 again. For this reason, Exim now forbids the simultaneous setting of
4876 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4877 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4878 entry if you want to know the details. The fix, however, neglected to
4879 consider the case when local delivery batching is involved. The test for
4880 "previously delivered" was not happening when checking to see if an
4881 address could be batched with a previous (undelivered) one; under
4882 certain circumstances this could lead to multiple deliveries to the same
4885 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4886 in its include files, and this causes problems building Exim.
4888 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4889 header_syntax) but Exim was just ignoring anything given after a slash.
4890 In particular, this caused confusion with an attempt to use "verify =
4891 reverse_host_lookup/defer_ok". An error is now given when options are
4892 supplied for verify items that do not have them. (Maybe reverse_host_
4893 lookup should have a defer_ok option, but that's a different point.)
4895 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4896 defined by RFC 821) to 2048, because there were problems with some AUTH
4897 commands, and RFC 1869 says the size should be increased for extended
4898 SMTP commands that take arguments.
4900 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4903 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4904 "unknown" error; now it says that the functionality isn't in the binary.
4906 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4907 an address' error message when a string expansion fails (syntax or
4908 whatever). Otherwise the password may appear in the log. Following change
4909 PH/42 below, there is no longer a chance of it appearing in a bounce
4912 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4914 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4915 of its addresses was ever tried. (Bugzilla bug #2.)
4917 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4918 the result incorrectly in the debug output. (It correctly added a newline
4919 to what was transported.)
4921 TF/01 Added $received_time.
4923 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4924 commented out examples of how to interface to a virus scanner and to
4925 SpamAssassin. Also added commented examples of av_scanner and
4926 spamd_address settings.
4928 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4929 and controls are allowed in which ACLs. There were a couple of minor
4930 errors. Some of the entries in the conditions table (which is a table of
4931 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4932 negation of where the condition IS allowed.
4934 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4936 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4937 header file does not have a version number, so I've had to invent a new
4938 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4939 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4940 radiusclient), but it was contributed by a Radius user.
4942 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4943 files or directories, for OpenSSL.
4945 PH/19 When an Exim process that is running as root has to create an Exim log
4946 file, it does so in a subprocess that runs as exim:exim so as to get the
4947 ownership right at creation (otherwise, other Exim processes might see
4948 the file with the wrong ownership). There was no test for failure of this
4949 fork() call, which would lead to the process getting stuck as it waited
4950 for a non-existent subprocess. Forks do occasionally fail when resources
4951 run out. I reviewed all the other calls to fork(); they all seem to check
4954 PH/20 When checking for unexpected SMTP input at connect time (before writing
4955 the banner), Exim was not dealing correctly with a non-positive return
4956 from the read() function. If the client had disconnected by this time,
4957 the result was a log entry for a synchronization error with an empty
4958 string after "input=" when read() returned zero. If read() returned -1
4959 (an event I could not check), uninitialized data bytes were printed.
4960 There were reports of junk text (parts of files, etc) appearing after
4963 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4965 PH/22 Added support for macro redefinition, and (re)definition in between
4966 driver and ACL definitions.
4968 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4969 forgetting to use the resulting value; it was using the unexpanded value.
4971 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4972 hadn't been configured. The fix is from Juergen Kreileder, who
4973 understands it better than I do:
4975 "Here's what I see happening with three configured cyrus_sasl
4976 authenticators configured (plain, login, cram-md5):
4978 On startup auth_cyrus_sasl_init() gets called for each of these.
4979 This means three calls to sasl_listmech() without a specified mech_list.
4980 => SASL tests which mechs of all available mechs actually work
4981 => three warnings about OTP not working
4982 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4984 With the patch, sasl_listmech() also gets called three times. But now
4985 SASL's mech_list option is set to the server_mech specified in the the
4986 authenticator. Or in other words, the answer from sasl_listmech()
4987 gets limited to just the mech you're testing for (which is different
4989 => the return list contains just 'plain' or 'login', 'cram-md5' or
4990 nothing depending on the value of ob->server_mech.
4992 I've just tested the patch: Authentication still works fine,
4993 unavailable mechs specified in the exim configuration are still
4994 caught, and the auth.log warnings about OTP are gone."
4996 PH/25 When debugging is enabled, the contents of the command line are added
4997 to the debugging output, even when log_selector=+arguments is not
5000 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5001 answer is "GNU", and only if the return is "GNU/something" is the answer
5004 PH/27 $acl_verify_message is now set immediately after the failure of a
5005 verification in an ACL, and so is available in subsequent modifiers. In
5006 particular, the message can be preserved by coding like this:
5008 warn !verify = sender
5009 set acl_m0 = $acl_verify_message
5011 Previously, $acl_verify_message was set only while expanding "message"
5012 and "log_message" when a very denied access.
5014 PH/28 Modified OS/os.c-Linux with
5016 -#ifndef OS_LOAD_AVERAGE
5017 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5019 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5020 nomenclature these days.)
5022 PH/29 Installed patch from the Sieve maintainer that adds the options
5023 sieve_useraddress and sieve_subaddress to the redirect router.
5025 PH/30 In these circumstances:
5026 . Two addresses routed to the same list of hosts;
5027 . First host does not offer TLS;
5028 . First host accepts first address;
5029 . First host gives temporary error to second address;
5030 . Second host offers TLS and a TLS session is established;
5031 . Second host accepts second address.
5032 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5033 and peerdn, if requested) that were in fact used only for the second
5036 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5037 attention to any local part prefix or suffix that was matched by the
5038 router that accepted the address. It now behaves in the same way as it
5039 does for delivery: the affixes are removed from the local part unless
5040 rcpt_include_affixes is set on the transport.
5042 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5043 timeout during the DATA phase of an incoming message.
5045 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5046 applied a patch sanctioned by the Sieve maintainer.
5048 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5049 the uid or gid is negative. A case of a negative gid caused this to be
5050 noticed. The fix allows for either to be negative.
5052 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5053 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5056 PH/36 The change PH/12 above was broken. Fixed it.
5058 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5059 the grounds that routing the same address twice would always produce the
5060 same answer. This might have been true once, but it is certainly no
5061 longer true now. Routing a child address may depend on the previous
5062 routing that produced that child. Some complicated redirection strategies
5063 went wrong when messages had multiple recipients, and made Exim's
5064 behaviour dependent on the order in which the addresses were given.
5066 I have moved the duplicate checking until after the routing is complete.
5067 Exim scans the addresses that are assigned to local and remote
5068 transports, and removes any duplicates. This means that more work will be
5069 done, as duplicates will always all be routed, but duplicates are
5070 presumably rare, so I don't expect this is of any significance.
5072 For deliveries to pipes, files, and autoreplies, the duplicate checking
5073 still happens during the routing process, since they are not going to be
5076 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5077 It corrects a timeout issue with spamd. This is Ian's comment: "The
5078 background is that sometimes spamd either never reads data from a
5079 connection it has accepted, or it never writes response data. The exiscan
5080 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5081 blindly assumes that writes won't block so it may never time out."
5083 PH/39 Allow G after quota size as well as K and M.
5085 PH/40 The value set for $authenticated_id in an authenticator may not contain
5086 binary zeroes or newlines because the value is written to log lines and
5087 to spool files. There was no check on this. Now the value is run through
5088 the string_printing() function so that such characters are converted to
5089 printable escape sequences.
5091 PH/41 $message_linecount is a new variable that contains the total number of
5092 lines in the message. Compare $body_linecount, which is the count for the
5095 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5096 bounce and delay warning messages, except in certain special cases, which
5099 (a) An SMTP error message from a remote host;
5100 (b) A message specified in a :fail: redirection;
5101 (c) A message specified in a "fail" command in a system filter;
5102 (d) A message specified in a FAIL return from the queryprogram router;
5103 (e) A message specified by the cannot_route_message router option.
5105 In these cases only, Exim does include the error details in bounce and
5106 warning messages. There are also a few cases where bland messages such
5107 as "unrouteable address" or "local delivery error" are given.
5109 PH/43 $value is now also set for the "else" part of a ${run expansion.
5111 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5112 being worked on, but at least Exim now implements the latest version to
5115 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5116 process to complete was treated as a delivery failure, a timeout while
5117 writing the message to the pipe was logged, but erroneously treated as a
5118 successful delivery. Such timeouts include transport filter timeouts. For
5119 consistency with the overall process timeout, these timeouts are now
5120 treated as errors, giving rise to delivery failures by default. However,
5121 there is now a new Boolean option for the pipe transport called
5122 timeout_defer, which, if set TRUE, converts the failures into defers for
5123 both kinds of timeout. A transport filter timeout is now identified in
5126 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5127 systems where "make" and "gmake" are different, calling "gmake" at top
5128 level broke things. I've arranged for the value of $(MAKE) to be passed
5129 from the Makefile to this script so that it can call the same version of
5133 A note about Exim versions 4.44 and 4.50
5134 ----------------------------------------
5136 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5137 changes of various kinds. As a consequence, a big documentation update was
5138 needed. This delayed the release for rather longer than seemed good, especially
5139 in the light of a couple of (minor) security issues. Therefore, the changes
5140 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5141 release. So 4.44 and 4.50 are in effect two different branches that both start
5144 I have left the 4.50 change log unchanged; it contains all the changes since
5145 4.43. The change log for 4.44 is below; many of its items are identical to
5146 those for 4.50. This seems to be the most sensible way to preserve the
5147 historical information.
5153 1. Minor wording change to the doc/README.SIEVE file.
5155 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5156 computation of the current number of files was incorrect.
5158 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5159 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5162 4. Give more explanation in the error message when the command for a transport
5163 filter fails to execute.
5165 5. There are several places where Exim runs a non-Exim command in a
5166 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5167 was being done only for the command run by the queryprogram router. It is
5168 now done for all such subprocesses. The other cases are: ${run, transport
5169 filters, and the commands run by the lmtp and pipe transports.
5171 6. Added CONFIGURE_GROUP build-time option.
5173 7. Some older OS have a limit of 256 on the maximum number of file
5174 descriptors. Exim was using setrlimit() to set 1000 as a large value
5175 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5176 systems. I've change it so that if it can't get 1000, it tries for 256.
5178 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5179 was an oversight, and furthermore, ever since the addition of extra
5180 controls (e.g. 4.43/32), the checks on when to allow different forms of
5181 "control" were broken. There should now be diagnostics for all cases when a
5182 control that does not make sense is encountered.
5184 9. Added the /retain_sender option to "control=submission".
5186 10. $recipients is now available in the predata ACL (oversight).
5188 11. Tidy the search cache before the fork to do a delivery from a message
5189 received from the command line. Otherwise the child will trigger a lookup
5190 failure and thereby defer the delivery if it tries to use (for example) a
5191 cached ldap connection that the parent has called unbind on.
5193 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5194 of $address_data from the recipient verification was clobbered by the
5195 sender verification.
5197 13. The value of address_data from a sender verification is now available in
5198 $sender_address_data in subsequent conditions in the ACL statement.
5200 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5202 15. Added a new option "connect=<time>" to callout options, to set a different
5205 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5206 was its contents. (It was OK if the option was not defined at all.)
5208 17. A "Completed" log line is now written for messages that are removed from
5209 the spool by the -Mrm option.
5211 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5212 information about exactly what failed.
5214 19. Added -dd to debug only the daemon process.
5216 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5217 handles timeouts, both on the server side and network timeouts. Renamed the
5218 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5220 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5221 It is now set to "smtps".
5223 22. $host_address is now set to the target address during the checking of
5224 ignore_target_hosts.
5226 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5227 being passed; this would have caused $sender_host_name to have been used if
5228 matching the list had actually called for a host name (not very likely,
5229 since this list is usually IP addresses). A host name is now passed as
5232 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5233 code that specifies a non-restarting handler (typically sigaction() in
5234 modern systems) in an attempt to fix a rare and obscure crash bug.
5236 25. Narrowed the window for a race in the daemon that could cause it to ignore
5237 SIGCHLD signals. This is not a major problem, because they are used only to
5238 wake it up if nothing else does.
5240 26. A malformed maildirsize file could cause Exim to calculate negative values
5241 for the mailbox size or file count. Odd effects could occur as a result.
5242 The maildirsize information is now recalculated if the size or filecount
5245 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5246 support for a long time. Removed HAVE_SYS_VFS_H.
5248 28. Installed the latest version of exipick from John Jetmore.
5250 29. In an address list, if the pattern was not a regular expression, an empty
5251 subject address (from a bounce message) matched only if the pattern was an
5252 empty string. Non-empty patterns were not even tested. This was the wrong
5253 because it is perfectly reasonable to use an empty address as part of a
5254 database query. An empty address is now tested by patterns that are
5255 lookups. However, all the other forms of pattern expect the subject to
5256 contain a local part and a domain, and therefore, for them, an empty
5257 address still always fails if the pattern is not itself empty.
5259 30. Exim went into a mad DNS loop when attempting to do a callout where the
5260 host was specified on an smtp transport, and looking it up yielded more
5261 than one IP address.
5263 31. Re-factored the code for checking spool and log partition space into a
5264 function that finds that data and another that does the check. The former
5265 is then used to implement four new variables: $spool_space, $log_space,
5266 $spool_inodes, and $log_inodes.
5268 32. The RFC2047 encoding function was originally intended for short strings
5269 such as real names; it was not keeping to the 75-character limit for
5270 encoded words that the RFC imposes. It now respects the limit, and
5271 generates multiple encoded words if necessary. To be on the safe side, I
5272 have increased the buffer size for the ${rfc2047: expansion operator from
5275 33. It is now permitted to omit both strings after an "if" condition; if the
5276 condition is true, the result is "true". As before, when the second string
5277 is omitted, a false condition yields an empty string. This makes it less
5278 cumbersome to write custom ACL and router conditions.
5280 34. Failure to deliver a bounce message always caused it to be frozen, even if
5281 there was an errors_to setting on the router. The errors_to setting is now
5284 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5285 canonical form (fully expanded) before being placed in
5286 $sender_host_address.
5288 36. The table in the code that translates DNS record types into text (T_A to
5289 "A" for instance) was missing entries for NS and CNAME. It is just possible
5290 that this could have caused confusion if both these types were looked up
5291 for the same domain, because the text type is used as part of Exim's
5292 per-process caching. But the chance of anyone hitting this buglet seems
5295 37. The dnsdb lookup has been extended in a number of ways.
5297 (1) There is a new type, "zns", which walks up the domain tree until it
5298 finds some nameserver records. It should be used with care.
5300 (2) There is a new type, "mxh", which is like "mx" except that it returns
5301 just the host names, not the priorities.
5303 (3) It is now possible to give a list of domains (or IP addresses) to be
5304 looked up. The behaviour when one of the lookups defers can be
5305 controlled by a keyword.
5307 (4) It is now possible to specify the separator character for use when
5308 multiple records are returned.
5310 38. The dnslists ACL condition has been extended: it is now possible to supply
5311 a list of IP addresses and/or domains to be looked up in a particular DNS
5314 39. Added log_selector=+queue_time_overall.
5316 40. When running the queue in the test harness, wait just a tad after forking a
5317 delivery process, to get repeatability of debugging output.
5319 41. Include certificate and key file names in error message when GnuTLS fails
5320 to set them up, because the GnuTLS error message doesn't include the name
5321 of the failing file when there is a problem reading it.
5323 42. Allow both -bf and -bF in the same test run.
5325 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5327 44. The "Exiscan patch" is now merged into the mainline Exim source.
5329 45. Sometimes the final signoff response after QUIT could fail to get
5330 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5331 < 0 before doing a fflush(). This bug looks as though it goes back to the
5332 introduction of TLS in release 3.20, but "sometimes" must have been rare
5333 because the tests only now provoked it.
5335 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5336 (this can affect the format of dates).
5338 47. exim_tidydb, when checking for the continued existence of a message for
5339 which it has found a message-specific retry record, was not finding
5340 messages that were in split spool directories. Consequently, it was
5341 deleting retry records that should have stayed in existence.
5343 48. Steve fixed some bugs in eximstats.
5345 49. The SPA authentication driver was not abandoning authentication and moving
5346 on to the next authenticator when an expansion was forced to fail,
5347 contradicting the general specification for all authenticators. Instead it
5348 was generating a temporary error. It now behaves as specified.
5350 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5351 (the order specifies the preference for clients). The order is now AES256,
5352 AES128, 3DES, ARCFOUR128.
5354 51. Small patch to Sieve code - explicitly set From: when generating an
5357 52. Exim crashed if a remote delivery caused a very long error message to be
5358 recorded - for instance if somebody sent an entire SpamAssassin report back
5359 as a large number of 550 error lines. This bug was coincidentally fixed by
5360 increasing the size of one of Exim's internal buffers (big_buffer) that
5361 happened as part of the Exiscan merge. However, to be on the safe side, I
5362 have made the code more robust (and fixed the comments that describe what
5365 53. Now that there can be additional text after "Completed" in log lines (if
5366 the queue_time_overall log selector is set), a one-byte patch to exigrep
5367 was needed to allow it to recognize "Completed" as not the last thing in
5370 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5371 patch that reportedly fixes this has been added. I am not expert enough to
5372 create a test for it. This is what the patch creator wrote:
5374 "I found a little strange behaviour of ldap code when working with
5375 Windows 2003 AD Domain, where users was placed in more than one
5376 Organization Units. When I tried to give exim partial DN, the exit code
5377 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5378 But simultaneously result of request was absolutely normal ldap result,
5379 so I produce this patch..."
5381 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5382 REFERENCE, so I have modified the code to exclude the patch when that macro
5385 55. Some experimental protocols are using DNS PTR records for new purposes. The
5386 keys for these records are domain names, not reversed IP addresses. The
5387 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5388 leaves it alone. Component reversal etc. now happens only for IP addresses.
5391 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5393 57. Double the size of the debug message buffer (to 2048) so that more of very
5394 long debug lines gets shown.
5396 58. The exicyclog utility now does better if the number of log files to keep
5397 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5399 59. Two changes related to the smtp_active_hostname option:
5401 (1) $smtp_active_hostname is now available as a variable.
5402 (2) The default for smtp_banner uses $smtp_active_hostname instead
5403 of $primary_hostname.
5405 60. The host_aton() function is supposed to be passed a string that is known
5406 to be a valid IP address. However, in the case of IPv6 addresses, it was
5407 not checking this. This is a hostage to fortune. Exim now panics and dies
5408 if the condition is not met. A case was found where this could be provoked
5409 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5410 components; fortuitously, this particular loophole had already been fixed
5411 by change 4.50/55 above.
5413 If there are any other similar loopholes, the new check in host_aton()
5414 itself should stop them being exploited. The report I received stated that
5415 data on the command line could provoke the exploit when Exim was running as
5416 exim, but did not say which command line option was involved. All I could
5417 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5418 running as the user.
5421 61. There was a buffer overflow vulnerability in the SPA authentication code
5422 (which came originally from the Samba project). I have added a test to the
5423 spa_base64_to_bits() function which I hope fixes it.
5426 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5427 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5429 63. The daemon start-up calls getloadavg() while still root for those OS that
5430 need the first call to be done as root, but it missed one case: when
5431 deliver_queue_load_max is set with deliver_drop_privilege. This is
5432 necessary for the benefit of the queue runner, because there is no re-exec
5433 when deliver_drop_privilege is set.
5435 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5436 This has been fixed.
5438 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5439 was in use, was not putting the data itself into the right store pool;
5440 consequently, it could be overwritten for a subsequent message in the same
5441 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5444 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5446 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5447 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5448 the function were treating the return as a boolean value, which happened to
5449 work because 0=false and not-0=true, but is not correct code.
5451 68. The host_aton() function was not handling scoped IPv6 addresses (those
5452 with, for example, "%eth0" on the end) correctly.
5454 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5455 negated items (that is, ~something) in unsigned ints. Some compilers
5456 apparently mutter when there is no cast.
5458 70. If an address verification called from an ACL failed, and did not produce a
5459 user-specific message (i.e. there was only a "system" message), nothing was
5460 put in $acl_verify_message. In this situation, it now puts the system
5463 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5464 session; change 4.31/43 added the unwanted input to the log line - except
5465 that it did not do this in the start of session case. It now does.
5467 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5468 This is wrong and can cause the other end to generate a synchronization
5469 error if it is another Exim or anything else that does the synchronization
5470 check. A QUIT command is no longer sent after a timeout.
5472 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5473 during host lookups.
5475 74. The defer_ok option of callout verification was not working if it was used
5476 when verifying addresses in header lines, that is, for this case:
5478 verify = header_sender/callout=defer_ok
5480 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5481 those file descriptors could be used for SMTP connections. If anything
5482 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5483 could be sent to the SMTP client, causing chaos. The daemon now opens
5484 stdin, stdout, and stderr to /dev/null when it puts itself into the
5487 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5488 log by default. The user can override this with suitable Perl magic.
5490 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5491 the log message when discard triggers, was not working for the DATA ACL or
5492 for the non-SMTP ACL.
5494 78. Error message wording change in sieve.c.
5496 79. If smtp_accept_max_per_host was set, the number of connections could be
5497 restricted to fewer than expected, because the daemon was trying to set up
5498 a new connection before checking whether the processes handling previous
5499 connections had finished. The check for completed processes is now done
5500 earlier. On busy systems, this bug wouldn't be noticed because something
5501 else would have woken the daemon, and it would have reaped the completed
5504 80. If a message was submitted locally by a user whose login name contained one
5505 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5506 It caused a spool format error. I have fixed the spool reading code. A
5507 related problem was that the "from" clause in the Received: line became
5508 illegal because of the space(s). It is now covered by ${quote_local_part.
5510 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5513 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5515 83. Patch from Sieve maintainer to fix unterminated string problem in
5516 "vacation" handling.
5518 84. Some minor changes to the Linux configuration files to help with other
5519 OS variants using glibc.
5521 85. One more patch for Sieve to update vacation handling to latest spec.
5524 ----------------------------------------------------
5525 See the note above about the 4.44 and 4.50 releases.
5526 ----------------------------------------------------
5532 1. Change 4.43/35 introduced a bug that caused file counts to be
5533 incorrectly computed when quota_filecount was set in an appendfile
5536 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5537 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5540 3. Give more explanation in the error message when the command for a transport
5541 filter fails to execute.
5543 4. There are several places where Exim runs a non-Exim command in a
5544 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5545 was being done only for the command run by the queryprogram router. It is
5546 now done for all such subprocesses. The other cases are: ${run, transport
5547 filters, and the commands run by the lmtp and pipe transports.
5549 5. Some older OS have a limit of 256 on the maximum number of file
5550 descriptors. Exim was using setrlimit() to set 1000 as a large value
5551 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5552 systems. I've change it so that if it can't get 1000, it tries for 256.
5554 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5555 was an oversight, and furthermore, ever since the addition of extra
5556 controls (e.g. 4.43/32), the checks on when to allow different forms of
5557 "control" were broken. There should now be diagnostics for all cases when a
5558 control that does not make sense is encountered.
5560 7. $recipients is now available in the predata ACL (oversight).
5562 8. Tidy the search cache before the fork to do a delivery from a message
5563 received from the command line. Otherwise the child will trigger a lookup
5564 failure and thereby defer the delivery if it tries to use (for example) a
5565 cached ldap connection that the parent has called unbind on.
5567 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5568 of $address_data from the recipient verification was clobbered by the
5569 sender verification.
5571 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5572 was its contents. (It was OK if the option was not defined at all.)
5574 11. A "Completed" log line is now written for messages that are removed from
5575 the spool by the -Mrm option.
5577 12. $host_address is now set to the target address during the checking of
5578 ignore_target_hosts.
5580 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5581 being passed; this would have caused $sender_host_name to have been used if
5582 matching the list had actually called for a host name (not very likely,
5583 since this list is usually IP addresses). A host name is now passed as
5586 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5587 code that specifies a non-restarting handler (typically sigaction() in
5588 modern systems) in an attempt to fix a rare and obscure crash bug.
5590 15. Narrowed the window for a race in the daemon that could cause it to ignore
5591 SIGCHLD signals. This is not a major problem, because they are used only to
5592 wake it up if nothing else does.
5594 16. A malformed maildirsize file could cause Exim to calculate negative values
5595 for the mailbox size or file count. Odd effects could occur as a result.
5596 The maildirsize information is now recalculated if the size or filecount
5599 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5600 support for a long time. Removed HAVE_SYS_VFS_H.
5602 18. Updated exipick to current release from John Jetmore.
5604 19. Allow an empty sender to be matched against a lookup in an address list.
5605 Previously the only cases considered were a regular expression, or an
5608 20. Exim went into a mad DNS lookup loop when doing a callout where the
5609 host was specified on the transport, if the DNS lookup yielded more than
5612 21. The RFC2047 encoding function was originally intended for short strings
5613 such as real names; it was not keeping to the 75-character limit for
5614 encoded words that the RFC imposes. It now respects the limit, and
5615 generates multiple encoded words if necessary. To be on the safe side, I
5616 have increased the buffer size for the ${rfc2047: expansion operator from
5619 22. Failure to deliver a bounce message always caused it to be frozen, even if
5620 there was an errors_to setting on the router. The errors_to setting is now
5623 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5624 canonical form (fully expanded) before being placed in
5625 $sender_host_address.
5627 24. Updated eximstats to version 1.33
5629 25. Include certificate and key file names in error message when GnuTLS fails
5630 to set them up, because the GnuTLS error message doesn't include the name
5631 of the failing file when there is a problem reading it.
5633 26. Expand error message when OpenSSL has problems setting up cert/key files.
5636 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5637 (this can affect the format of dates).
5639 28. exim_tidydb, when checking for the continued existence of a message for
5640 which it has found a message-specific retry record, was not finding
5641 messages that were in split spool directories. Consequently, it was
5642 deleting retry records that should have stayed in existence.
5644 29. eximstats updated to version 1.35
5645 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5646 1.35 - bugfix such that pie charts by volume are generated correctly
5648 30. The SPA authentication driver was not abandoning authentication and moving
5649 on to the next authenticator when an expansion was forced to fail,
5650 contradicting the general specification for all authenticators. Instead it
5651 was generating a temporary error. It now behaves as specified.
5653 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5654 (the order specifies the preference for clients). The order is now AES256,
5655 AES128, 3DES, ARCFOUR128.
5657 31. Small patch to Sieve code - explicitly set From: when generating an
5660 32. Exim crashed if a remote delivery caused a very long error message to be
5661 recorded - for instance if somebody sent an entire SpamAssassin report back
5662 as a large number of 550 error lines. This bug was coincidentally fixed by
5663 increasing the size of one of Exim's internal buffers (big_buffer) that
5664 happened as part of the Exiscan merge. However, to be on the safe side, I
5665 have made the code more robust (and fixed the comments that describe what
5668 33. Some experimental protocols are using DNS PTR records for new purposes. The
5669 keys for these records are domain names, not reversed IP addresses. The
5670 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5671 leaves it alone. Component reversal etc. now happens only for IP addresses.
5674 34. The host_aton() function is supposed to be passed a string that is known
5675 to be a valid IP address. However, in the case of IPv6 addresses, it was
5676 not checking this. This is a hostage to fortune. Exim now panics and dies
5677 if the condition is not met. A case was found where this could be provoked
5678 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5679 components; fortuitously, this particular loophole had already been fixed
5680 by change 4.50/55 or 4.44/33 above.
5682 If there are any other similar loopholes, the new check in host_aton()
5683 itself should stop them being exploited. The report I received stated that
5684 data on the command line could provoke the exploit when Exim was running as
5685 exim, but did not say which command line option was involved. All I could
5686 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5687 running as the user.
5690 35. There was a buffer overflow vulnerability in the SPA authentication code
5691 (which came originally from the Samba project). I have added a test to the
5692 spa_base64_to_bits() function which I hope fixes it.
5695 36. The daemon start-up calls getloadavg() while still root for those OS that
5696 need the first call to be done as root, but it missed one case: when
5697 deliver_queue_load_max is set with deliver_drop_privilege. This is
5698 necessary for the benefit of the queue runner, because there is no re-exec
5699 when deliver_drop_privilege is set.
5701 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5702 was in use, was not putting the data itself into the right store pool;
5703 consequently, it could be overwritten for a subsequent message in the same
5704 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5707 38. Sometimes the final signoff response after QUIT could fail to get
5708 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5709 < 0 before doing a fflush(). This bug looks as though it goes back to the
5710 introduction of TLS in release 3.20, but "sometimes" must have been rare
5711 because the tests only now provoked it.
5717 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5718 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5719 Now it returns an integer. A number of calls were still expecting a T/F
5720 return. Fortuitously, in all cases, the tests worked in OK situations,
5721 which is the norm. However, things would have gone wrong on any write
5722 failures on the smtp file descriptor. This function is used when sending
5723 messages over SMTP and also when doing verify callouts.
5725 2. When Exim is called to do synchronous delivery of a locally submitted
5726 message (the -odf or -odi options), it no longer closes stderr before doing
5729 3. Implemented the mua_wrapper option.
5731 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5733 5. Implemented the functions header_remove(), header_testname(),
5734 header_add_at_position(), and receive_remove_recipient(), and exported them
5737 6. If an ACL "warn" statement specified the addition of headers, Exim already
5738 inserted X-ACL-Warn: at the start if there was no header name. However, it
5739 was not making this test for the second and subsequent header lines if
5740 there were newlines in the string. This meant that an invalid header could
5741 be inserted if Exim was badly configured.
5743 7. Allow an ACL "warn" statement to add header lines at the start or after all
5744 the Received: headers, as well as at the end.
5746 8. Added the rcpt_4xx retry error code.
5748 9. Added postmaster_mailfrom=xxx to callout verification option.
5750 10. Added mailfrom=xxxx to the callout verification option, for verify=
5753 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5754 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5756 12. Inserted some casts to stop certain compilers warning when using pointer
5757 differences as field lengths or precisions in printf-type calls (mostly
5758 affecting debugging statements).
5760 13. Added optional readline() support for -be (dynamically loaded).
5762 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5763 same clock tick as a message's arrival, so that its received time was the
5764 same as the "first fail" time on the retry record, and that message
5765 remained on the queue past the ultimate address timeout, every queue runner
5766 would try a delivery (because it was past the ultimate address timeout) but
5767 after another failure, the ultimate address timeout, which should have then
5768 bounced the address, did not kick in. This was a "< instead of <=" error;
5769 in most cases the first failure would have been in the next clock tick
5770 after the received time, and all would be well.
5772 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5773 being recognized when the domain list was tested by the match_domain
5774 condition in an expansion string.
5776 16. Added the ${str2b64: operator.
5778 17. Exim was always calling setrlimit() to set a large limit for the number of
5779 processes, without checking whether the existing limit was already
5780 adequate. (It did check for the limit on file descriptors.) Furthermore,
5781 errors from getrlimit() and setrlimit() were being ignored. Now they are
5782 logged to the main and panic logs, but Exim does carry on, to try to do its
5783 job under whatever limits there are.
5785 18. Imported PCRE 5.0.
5787 19. Trivial typo in log message " temporarily refused connection" (the leading
5790 20. If the log selector return_path_on_delivery was set and an address was
5791 redirected to /dev/null, the delivery process crashed because it assumed
5792 that a return path would always be set for a "successful" delivery. In this
5793 case, the whole delivery is bypassed as an optimization, and therefore no
5796 21. Internal re-arrangement: the function for sending a challenge and reading
5797 a response while authentication was assuming a zero-terminated challenge
5798 string. It's now changed to take a pointer and a length, to allow for
5799 binary data in such strings.
5801 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5803 23. Exim was not respecting finduser_retries when seeking the login of the
5804 uid under which it was called; it was always trying 10 times. (The default
5805 setting of finduser_retries is zero.) Also, it was sleeping after the final
5806 failure, which is pointless.
5808 24. Implemented tls_on_connect_ports.
5810 25. Implemented acl_smtp_predata.
5812 26. If the domain in control=submission is set empty, Exim assumes that the
5813 authenticated id is a complete email address when it generates From: or
5814 Sender: header lines.
5816 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5817 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5818 chown and chgrp in /bin and hostname in /usr/bin.
5820 28. Exim was keeping the "process log" file open after each use, just as it
5821 does for the main log. This opens the possibility of it remaining open for
5822 long periods when the USR1 signal hits a daemon. Occasional processlog
5823 errors were reported, that could have been caused by this. Anyway, it seems
5824 much more sensible not to leave this file open at all, so that is what now
5827 29. The long-running daemon process does not normally write to the log once it
5828 has entered its main loop, and it closes the log before doing so. This is
5829 so that log files can straightforwardly be renamed and moved. However,
5830 there are a couple of unusual error situations where the daemon does write
5831 log entries, and I had neglected to close the log afterwards.
5833 30. The text of an SMTP error response that was received during a remote
5834 delivery was being truncated at 512 bytes. This is too short for some of
5835 the long messages that one sometimes sees. I've increased the limit to
5838 31. It is now possible to make retry rules that apply only when a message has a
5839 specific sender, in particular, an empty sender.
5841 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5842 it possible to be selective about when SMTP synchronization is enforced.
5844 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5846 32. Implemented hosts_connection_nolog.
5848 33. Added an ACL for QUIT.
5850 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5853 35. Added mailbox_size and mailbox_filecount to appendfile.
5855 36. Added control = no_multiline_responses to ACLs.
5857 37. There was a bug in the logic of the code that waits for the clock to tick
5858 in the case where the clock went backwards by a substantial amount such
5859 that the microsecond fraction of "now" was more than the microsecond
5860 fraction of "then" (but the whole seconds number was less).
5862 38. Added support for the libradius Radius client library this is found on
5863 FreeBSD (previously only the radiusclient library was supported).
5869 1. When certain lookups returned multiple values in the form name=value, the
5870 quoting of the values was not always being done properly. Specifically:
5871 (a) If the value started with a double quote, but contained no whitespace,
5873 (b) If the value contained whitespace other than a space character (i.e.
5874 tabs or newlines or carriage returns) it was not quoted.
5875 This fix has been applied to the mysql and pgsql lookups by writing a
5876 separate quoting function and calling it from the lookup code. The fix
5877 should probably also be applied to nisplus, ibase and oracle lookups, but
5878 since I cannot test any of those, I have not disturbed their existing code.
5880 2. A hit in the callout cache for a specific address caused a log line with no
5881 reason for rejecting RCPT. Now it says "Previous (cached) callout
5882 verification failure".
5884 3. There was an off-by-one bug in the queryprogram router. An over-long
5885 return line was truncated at 256 instead of 255 characters, thereby
5886 overflowing its buffer with the terminating zero. As well as fixing this, I
5887 have increased the buffer size to 1024 (and made a note to document this).
5889 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5890 when Exim is waiting for an SMTP response from a remote server, Exim
5891 restarts its select() call on the socket, thereby resetting its timeout.
5892 This is not a problem when such interrupts are rare. Somebody set up a cron
5893 job to run exiwhat every 2 minutes, which is less than the normal select()
5894 timeout (5 or 10 minutes). This meant that the select() timeout never
5895 kicked in because it was always reset. I have fixed this by comparing the
5896 time when an interrupt arrives with the time at the start of the first call
5897 to select(). If more time than the timeout has elapsed, the interrupt is
5898 treated as a timeout.
5900 5. Some internal re-factoring in preparation for the addition of Sieve
5901 extensions (by MH). In particular, the "personal" test is moved to a
5902 separate function, and given an option for scanning Cc: and Bcc: (which is
5903 not set for Exim filters).
5905 6. When Exim created an email address using the login of the caller as the
5906 local part (e.g. when creating a From: or Sender: header line), it was not
5907 quoting the local part when it contained special characters such as @.
5909 7. Installed new OpenBSD configuration files.
5911 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5912 try to make them clearer.
5914 9. Callout options, other than the timeout value, were being ignored when
5915 verifying sender addresses in header lines. For example, when using
5917 verify = header_sender/callout=no_cache
5919 the cache was (incorrectly) being used.
5921 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5922 only the Cygwin environment.
5924 11. When return_path_on_delivery was set as a log selector, if different remote
5925 addresses in the same message used different return paths and parallel
5926 remote delivery occurred, the wrong values would sometimes be logged.
5927 (Whenever a remote delivery process finished, the return path value from
5928 the most recently started remote delivery process was logged.)
5930 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5931 lines when AUTH and/or TLS are in use. This is the "received protocol"
5932 field. Exim used to use "asmtp" for authenticated SMTP, without any
5933 indication (in the protocol name) for TLS use. Now it follows the RFC and
5934 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5935 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5936 names appear in log lines as well as in Received: header lines.
5938 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5939 extensions, and comparison tests, and to fix some bugs.
5941 14. Changes to the "personal" filter test:
5943 (1) The test was buggy in that it was just doing the equivalent of
5944 "contains" tests on header lines. For example, if a user's address was
5945 anne@some.where, the "personal" test would incorrectly be true for
5947 To: susanne@some.where
5949 This test is now done by extracting each address from the header in turn,
5950 and checking the entire address. Other tests that are part of "personal"
5951 are now done using regular expressions (for example, to check local parts
5952 of addresses in From: header lines).
5954 (2) The list of non-personal local parts in From: addresses has been
5955 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5956 taken from the Sieve specification recommendations.
5958 (3) If the message contains any header line starting with "List-" it is
5959 treated as non-personal.
5961 (4) The test for "circular" in the Subject: header line has been removed
5962 because it now seems ill-conceived.
5964 15. Minor typos in src/EDITME comments corrected.
5966 16. Installed latest exipick from John Jetmore.
5968 17. If headers_add on a router specified a text string that was too long for
5969 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5970 of string_sprintf() is now avoided.
5972 18. $message_body_size was not set (it was always zero) when running the DATA
5973 ACL and the local_scan() function.
5975 19. For the "mail" command in an Exim filter, no default was being set for
5976 the once_repeat time, causing a random time value to be used if "once" was
5977 specified. (If the value happened to be <= 0, no repeat happened.) The
5978 default is now 0s, meaning "never repeat". The "vacation" command was OK
5979 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5980 (I found it when inspecting the code).
5982 20. There is now an overall timeout for performing a callout verification. It
5983 defaults to 4 times the callout timeout, which applies to individual SMTP
5984 commands during the callout. The overall timeout applies when there is more
5985 than one host that can be tried. The timeout is checked before trying the
5986 next host. This prevents very long delays if there are a large number of
5987 hosts and all are timing out (e.g. when the network connections are timing
5988 out). The value of the overall timeout can be changed by specifying an
5989 additional sub-option for "callout", called "maxwait". For example:
5991 verify = sender/callout=5s,maxwait=20s
5993 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5994 to the end before writing, but this should make it even safer).
5996 22. Exim was forgetting that it had advertised PIPELINING for the second and
5997 subsequent messages on an SMTP connection. It was also not resetting its
5998 memory on STARTTLS and an internal HELO.
6000 23. When Exim logs an SMTP synchronization error within a session, it now
6001 records whether PIPELINING has been advertised or not.
6003 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6004 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6005 rather than long int.
6007 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6009 26. Added the never_mail option to autoreply.
6015 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6016 crash if the getsockname() call failed; this can happen if a connection is
6017 closed very soon after it is established. The problem was simply in the
6018 order in which certain operations were done, causing Exim to try to write
6019 to the SMTP stream before it had set up the file descriptor. The bug has
6020 been fixed by making things happen in the correct order.
6026 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6027 before the connection was closed, thus losing the rejection response.
6029 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6030 some early Solaris releases, but causes trouble in current releases where
6031 socklen_t is defined.
6033 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6036 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6039 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6040 that releasing the top of it at the end releases what was used for sub-
6041 expansions (unless the block got too big). However, discard this block if
6042 the first thing is a variable or header, so that we can use its block when
6043 it is dynamic (useful for very large $message_headers, for example).
6045 6. Lookups now cache *every* query, not just the most recent. A new, separate
6046 store pool is used for this. It can be recovered when all lookup caches are
6047 flushed. Lookups now release memory at the end of their result strings.
6048 This has involved some general refactoring of the lookup sources.
6050 7. Some code has been added to the store_xxx() functions to reduce the amount
6051 of flapping under certain conditions.
6053 8. log_incoming_interface used to affect only the <= reception log lines. Now
6054 it causes the local interface and port to be added to several more SMTP log
6055 lines, for example "SMTP connection from", and rejection lines.
6057 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6059 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6061 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6062 could be overwritten at the end of the current message (or the start of a
6063 new message if it was set in a HELO ACL). The value is now preserved for
6064 the duration of the SMTP connection.
6066 12. If a transport had a headers_rewrite setting, and a matching header line
6067 contained an unqualified address, that address was qualified, even if it
6068 did not match any rewriting rules. The underlying bug was that the values
6069 of the flags that permit the existence of unqualified sender and recipient
6070 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6071 non-local messages, and by -bnq for local messages) were not being
6072 preserved with the message after it was received.
6074 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6075 "next input=" as part of the text comprising the host identity instead of
6076 the correct text. The code was using the same buffer for two different
6077 strings. However, depending on which order the printing function evaluated
6078 its arguments, the bug did not always show up. Under Linux, for example, my
6079 test suite worked just fine.
6081 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6082 doesn't work with some older versions of Perl. It has been changed to "my",
6083 which in any case is probably the better facility to use.
6085 15. A really picky compiler found some instances of statements for creating
6086 error messages that either had too many or two few arguments for the format
6089 16. The size of the buffer for calls to the DNS resolver has been increased
6090 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6091 for addresses that have a lot of PTR records. This alleviates a problem; it
6092 does not fully solve it.
6094 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6095 buffer now truncates the list and logs the incident, which is the same
6096 action as happens when Exim is looking up a host name and its aliases.
6097 Previously in this situation something unpredictable would happen;
6098 sometimes it was "internal error: store_reset failed".
6100 18. If a server dropped the connection unexpectedly when an Exim client was
6101 using GnuTLS and trying to read a response, the client delivery process
6102 crashed while trying to generate an error log message.
6104 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6105 string, for example:
6107 warn message = H1: something\nH2: something
6109 the text was added as a single header line from Exim's point of view
6110 though it ended up OK in the delivered message. However, searching for the
6111 second and subsequent header lines using $h_h2: did not work. This has been
6112 fixed. Similarly, if a system filter added multiple headers in this way,
6113 the routers could not see them.
6115 20. Expanded the error message when iplsearch is called with an invalid key to
6116 suggest using net-iplsearch in a host list.
6118 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6119 ACLs are no longer actually imposed (and a message to that effect is
6122 22. If a "gecos" field in a passwd entry contained escaped characters, in
6123 particular, if it contained a \" sequence, Exim got it wrong when building
6124 a From: or a Sender: header from that name. A second bug also caused
6125 incorrect handling when an unquoted " was present following a character
6126 that needed quoting.
6128 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6129 was not being matched caselessly.
6131 24. Arranged for all hyphens in the exim.8 source to be escaped with
6134 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6135 properly. Recipient callout cache records were still being keyed to include
6136 the sender, even when use_sender was set false. This led to far more
6137 callouts that were necessary. The sender is no longer included in the key
6138 when use_sender is false.
6140 26. Added "control = submission" modifier to ACLs.
6142 27. Added the ${base62d: operator to decode base 62 numbers.
6144 28. dnsdb lookups can now access SRV records.
6146 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6147 the configuration file.
6149 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6150 (-v) mode. This makes the output for a verbose queue run more intelligible.
6152 31. Added a use_postmaster feature to recipient callouts.
6154 32. Added the $body_zerocount variable, containing the number of binary zero
6155 bytes in the message body.
6157 33. The time of last modification of the "new" subdirectory is now used as the
6158 "mailbox time last read" when there is a quota error for a maildir
6161 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6163 35. Added +ignore_unknown as a special item in host lists.
6165 36. Code for decoding IPv6 addresses in host lists is now included, even if
6166 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6167 address was recognized as an IP address, but was then not correctly decoded
6168 into binary, causing unexpected and incorrect effects when compared with
6175 1. Very minor rewording of debugging text in manualroute to say "list of
6176 hosts" instead of "hostlist".
6178 2. If verify=header_syntax was set, and a header line with an unqualified
6179 address (no domain) and a large number of spaces between the end of the
6180 name and the colon was received, the reception process suffered a buffer
6181 overflow, and (when I tested it) crashed. This was caused by some obsolete
6182 code that should have been removed. The fix is to remove it!
6184 3. When running in the test harness, delay a bit after writing a bounce
6185 message to get a bit more predictability in the log output.
6187 4. Added a call to search_tidyup() just before forking a reception process. In
6188 theory, someone could use a lookup in the expansion of smtp_accept_max_
6189 per_host which, without the tidyup, could leave open a database connection.
6191 5. Added the variables $recipient_data and $sender_data which get set from a
6192 lookup success in an ACL "recipients" or "senders" condition, or a router
6193 "senders" option, similar to $domain_data and $local_part_data.
6195 6. Moved the writing of debug_print from before to after the "senders" test
6198 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6199 problems for message scanning, either using a data ACL, or using
6200 local_scan() because the Received: header was not generated till after they
6201 were called (in order to set the time as the time of reception completion).
6202 I have revised the way this works. The header is now generated after the
6203 body is received, but before the ACL or local_scan() are called. After they
6204 are run, the timestamp in the header is updated.
6210 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6211 before starting a queue runner without re-exec. This happened only when
6212 deliver_drop_privilege was set or when the Exim user was set to root. The
6213 effect of the bug was that timeouts during subsequent deliveries caused
6214 crashes instead of being properly handled. The handler is now left at its
6215 default (and expected) setting.
6217 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6218 message, again when deliver_drop_privilege is set or Exim is run as root.
6219 The bug described in (1) was not present in this case, but the tidying up
6220 of the other signals was missing. I have made the two cases consistent.
6222 3. The ignore_target_hosts setting on a manualroute router was being ignored
6223 for hosts that were looked up using the /MX notation.
6225 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6228 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6229 operated on the sender address. After changing the $sender_address to <>
6230 for the sender address verify, Exim was re-instated it as the original
6231 (before rewriting) address, but remembering that it had rewritten it, so it
6232 wasn't rewriting it again. This bug also had the effect of breaking the
6233 sender address verification caching when the sender address was rewritten.
6235 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6236 This has been changed so that if the ip literal address matches
6237 ignore_target_hosts, the router declines.
6239 7. Added expansion conditions match_domain, match_address, and match_local_
6240 part (NOT match_host).
6242 8. The placeholder for the Received: header didn't have a length field set.
6244 9. Added code to Exim itself and to exim_lock to test for a specific race
6245 condition that could lead to file corruption when using MBX delivery. The
6246 issue is with the lockfile that is created in /tmp. If this file is removed
6247 after a process has opened it but before that process has acquired a lock,
6248 there is the potential for a second process to recreate the file and also
6249 acquire a lock. This could lead to two Exim processes writing to the file
6250 at the same time. The added code performs the same test as UW imapd; it
6251 checks after acquiring the lock that its file descriptor still refers to
6252 the same named file.
6254 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6255 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6258 11. Added the smtp_active_hostname option. If used, this will typically be made
6259 to depend on the incoming interface address. Because $interface_address is
6260 not set up until the daemon has forked a reception process, error responses
6261 that can happen earlier (such as "too many connections") no longer contain
6264 12. If an expansion in a condition on a "warn" statement fails because a lookup
6265 defers, the "warn" statement is abandoned, and the next ACL statement is
6266 processed. Previously this caused the whole ACL to be aborted.
6268 13. Added the iplsearch lookup type.
6270 14. Added ident_timeout as a log selector.
6272 15. Added tls_certificate_verified as a log selector.
6274 16. Added a global option tls_require_ciphers (compare the smtp transport
6275 option of the same name). This controls incoming TLS connections.
6277 17. I finally figured out how to make tls_require_ciphers do a similar thing
6278 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6279 before starting the TLS session.
6281 18. Tabs are now shown as \t in -bP output.
6283 19. If the log selector return_path_on_delivery was set, Exim crashed when
6284 bouncing a message because it had too many Received: header lines.
6286 20. If two routers both had headers_remove settings, and the first one included
6287 a superfluous trailing colon, the final name in the first list and the
6288 first name in the second list were incorrectly joined into one item (with a
6289 colon in the middle).
6295 1. Added -C and -D options to the exinext utility, mainly to make it easier
6296 to include in the automated testing, but these could be helpful when
6297 multiple configurations are in use.
6299 2. The exinext utility was not formatting the output nicely when there was
6300 an alternate port involved in the retry record key, nor when there was a
6301 message id as well (for retries that were specific to a specific message
6302 and a specific host). It was also confused by IPv6 addresses, because of
6303 the additional colons they contain. I have fixed the IPv4 problem, and
6304 patched it up to do a reasonable job for IPv6.
6306 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6307 delivery, the log line now contains "pipelined" if PIPELINING was used.
6309 4. An SMTP transport process used to panic and die if the bind() call to set
6310 an explicit outgoing interface failed. This has been changed; it is now
6311 treated in the same way as a connect() failure.
6313 5. A reference to $sender_host_name in the part of a conditional expansion
6314 that was being skipped was still causing a DNS lookup. This no longer
6317 6. The def: expansion condition was not recognizing references to header lines
6318 that used bh_ and bheader_.
6320 7. Added the _cache feature to named lists.
6322 8. The code for checking quota_filecount in the appendfile transport was
6323 allowing one more file than it should have been.
6325 9. For compatibility with Sendmail, the command line option
6333 and sets the incoming protocol and host name (for trusted callers). The
6334 host name and its colon can be omitted when only the protocol is to be set.
6335 Note the Exim already has two private options, -pd and -ps, that refer to
6336 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6337 "s", but I don't think that's a major issue.
6339 10. A number of refactoring changes to the code, none of which should affect
6342 (a) The number of logging options was getting close to filling up the
6343 32-bit word that was used as a bit map. I have split them into two classes:
6344 those that are passed in the argument to log_write(), and those that are
6345 only ever tested independently outside of that function. These are now in
6346 separate 32-bit words, so there is plenty of room for expansion again.
6347 There is no change in the user interface or the logging behaviour.
6349 (b) When building, for example, log lines, the code previously used a
6350 macro that called string_cat() twice, in order to add two strings. This is
6351 not really sufficiently general. Furthermore, there was one instance where
6352 it was actually wrong because one of the argument was used twice, and in
6353 one call a function was used. (As it happened, calling the function twice
6354 did not affect the overall behaviour.) The macro has been replaced by a
6355 function that can join an arbitrary number of extra strings onto a growing
6358 (c) The code for expansion conditions now uses a table and a binary chop
6359 instead of a serial search (which was left over from when there were very
6360 few conditions). Also, it now recognizes conditions like "pam" even when
6361 the relevant support is not compiled in: a suitably worded error message is
6362 given if an attempt is made to use such a condition.
6364 11. Added ${time_interval:xxxxx}.
6366 12. A bug was causing one of the ddress fields not to be passed back correctly
6367 from remote delivery subprocesses. The field in question was not being
6368 subsequently used, so this caused to problems in practice.
6370 13. Added new log selectors queue_time and deliver_time.
6372 14. Might have fixed a bug in maildirsizefile handling that threw up
6373 "unexpected character" debug warnings, and recalculated the data
6374 unnecessarily. In any case, I expanded the warning message to give more
6377 15. Added the message "Restricted characters in address" to the statements in
6378 the default ACL that block characters like @ and % in local parts.
6380 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6381 Three changes have been made:
6383 (a) There was a serious bug; a negative response to MAIL caused the whole
6384 recipient domain to be cached as invalid, thereby blocking all messages
6385 to all local parts at the same domain, from all senders. This bug has
6386 been fixed. The domain is no longer cached after a negative response to
6387 MAIL if the sender used is not empty.
6389 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6392 (c) A new callout option, "use_sender" has been added for people who want
6393 the modified behaviour.
6399 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6402 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6403 indeed breaks things for older releases.
6405 3. Added additional logging to the case where there is a problem reading data
6406 from a filter that is running in a subprocess using a pipe, in order to
6407 try to track down a specific problem.
6409 4. Testing facility fudge: when running in the test harness and attempting
6410 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6411 getting "No route to host". Convert this to a timeout.
6413 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6416 6. Some OS don't have socklen_t but use size_t instead. This affects the
6417 fifth argument of getsockopt() amongst other things. This is now
6418 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6419 can be set for individual OS. I have set it for SunOS5, OSF1, and
6420 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6421 some earlier ones do not.
6423 7. Change 4.30/15 was not doing the test caselessly.
6425 8. The standard form for an IPv6 address literal was being rejected by address
6426 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6427 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6428 this, as well as the form without the "IPv6" on the front (but only when
6429 address literals are enabled, of course).
6431 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6433 10. Exim crashed if a message with an empty sender address specified by -f
6434 encountered a router with an errors_to setting. This could be provoked only
6435 by a command such as
6439 where an empty string was supplied; "<>" did not hit this bug.
6441 11. Installed PCRE release 4.5.
6443 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6444 remained set. It is now erased.
6446 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6447 times from message ids (which are base 36 rather than the normal 62).
6449 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6450 were being counted as actual protocol errors, and logged if the log
6451 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6452 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6453 and DATA following a set of rejected RCPTs do not count as protocol errors.
6454 In other words, Exim assumes they were pipelined, though this may not
6455 actually be the case. Of course, in all cases the client gets an
6456 appropriate error code.
6458 15. If a lookup fails in an ACL condition, a message about the failure may
6459 be available; it is used if testing the ACL cannot continue, because most
6460 such messages specify what the cause of the deferral is. However, some
6461 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6462 that caused an old message to be retained and used if a later statement
6463 caused a defer, replacing the real cause of the deferral.
6465 16. If an IP address had so many PTR records that the DNS lookup buffer
6466 was not large enough to hold them, Exim could crash while trying to process
6467 the truncated data. It now detects and logs this case.
6469 17. Further to 4.21/58, another change has been made: if (and only if) the
6470 first line of a message (the first header line) ends with CRLF, a bare LF
6471 in a subsequent header line has a space inserted after it, so as not to
6472 terminate the header.
6474 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6475 unnecessarily, used atoi() instead of strtol(), and didn't check the
6476 termination when getting file sizes from file names by regex.
6478 19. Completely re-implemented the support for maildirsize files, in the light
6479 of a number of problems with the previous contributed implementation
6480 (4.30/29). In particular:
6482 . If the quota is zero, the maildirsize file is maintained, but no quota is
6485 . If the maildir directory does not exist, it is created before any attempt
6486 to write a maildirsize file.
6488 . The quota value in the file is just a cache; if the quota is changed in
6489 the transport, the new value overrides.
6491 . A regular expression is available for excluding directories from the
6494 20. The autoreply transport checks the characters in options that define the
6495 message's headers; it allows continued headers, but it was checking with
6496 isspace() after an embedded newline instead of explicitly looking for a
6499 21. If all the "regular" hosts to which an address was routed had passed their
6500 expiry times, and had not reached their retry times, the address was
6501 bounced, even if fallback hosts were defined. Now Exim should go on to try
6504 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6505 equivalent code in the SMTP transport. Some hosts send humungous responses
6506 to HELO/EHLO, more than 1024 it seems.
6508 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6509 gives compiler warnings in some environments. I've now done it "properly",
6512 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6513 (because of problems with the built-in one) was declared to return uschar *
6514 instead of char *, causing compiler failure.
6516 25. Fixed a file descriptor leak when processing alias/forward files.
6518 26. Fixed a minor format string issue in dbfn.c.
6520 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6522 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6523 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6526 29. When a sender address is verified, it is cached, to save repeating the test
6527 when there is more than one recipient in a message. However, when the
6528 verification involves a callout, it is possible for different callout
6529 options to be set for different recipients. It is too complicated to keep
6530 track of this in the cache, so now Exim always runs a verification when a
6531 callout is required, relying on the callout cache for the optimization.
6532 The overhead is duplication of the address routing, but this should not be
6535 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6536 to be verified with callout=postmaster, and the main callout worked but the
6537 postmaster check failed, the verification correctly failed. However, if a
6538 subsequent RCPT command asked for sender verification *without* the
6539 postmaster check, incorrect caching caused this verification also to fail,
6542 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6543 it was not caching the DNS options (qualify_single, search_parents) that
6544 were used when the lookup failed. A subsequent lookup with different
6545 options therefore always gave the same answer, though there were cases
6546 where it should not have. (Example: a "domains = !$mx_any" option on a
6547 dnslookup router: the "domains" option is always processed without any
6548 widening, but the router might have qualify_single set.) Now Exim uses the
6549 cached value only when the same options are set.
6551 32. Added John Jetmore's "exipick" utility to the distribution.
6553 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6554 than a timeout (e.g. a certificate is required, and is not provided), an
6555 Exim server now closes the connection immediately. Previously it waited for
6556 the client to close - but if the client is SSL, it seems that they each
6557 wait for each other, leading to a delay before one of them times out.
6559 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6560 maintained 0.8.x compatibility because I don't think many are using it, and
6561 it is clearly obsolete.
6563 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6566 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6567 client certificate was expired. A simple patch fixes this, though I don't
6568 understand the full logic of why the verify callback is called multiple
6571 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6572 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6573 which causes problems with some clients (such as the Certicom SSL Plus
6574 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6575 disables the coutermeasure allowing Eudora to connect."
6577 38. Exim was not checking that a write() to a log file succeeded. This could
6578 lead to Bad Things if a log got too big, in particular if it hit a file
6579 size limit. Exim now panics and dies if it cannot write to a log file, just
6580 as it does if it cannot open a log file.
6582 39. Modified OS/Makefile-Linux so that it now contains
6584 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6586 The two -D definitions ensure that Exim is compiled with large file
6587 support, which makes it possible to handle log files that are bigger than
6590 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6591 instance) a domain was checked against a named list that involved a lookup,
6592 causing $domain_data to be set, then another domain was checked against the
6593 same list, then the first domain was re-checked, the value of $domain_data
6594 after the final check could be wrong. In particular, if the second check
6595 failed, it could be set empty. This bug probably also applied to
6598 41. The strip_trailing_dot option was not being applied to the address given
6599 with the -f command-line option.
6601 42. The code for reading a message's header from the spool was incrementing
6602 $received_count, but never initializing it. This meant that the value was
6603 incorrect (doubled) while delivering a message in the same process in which
6604 it was received. In the most common configuration of Exim, this never
6605 happens - a fresh exec is done - but it can happen when
6606 deliver_drop_privilege is set.
6608 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6609 it now includes up to 150 characters of the unexpected data in the log
6612 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6613 and building data strings. The size of both of these buffers was 10 000
6614 bytes - far larger than anybody would *ever* want, thought I. Needless to
6615 say, somebody hit the limit. I have increased the maximum line length to
6616 20 000 and the maximum data length of concatenated lines to 100 000. I have
6617 also fixed two bugs, because there was no checking on these buffers. Tsk,
6618 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6619 buffer is too small.
6621 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6622 lsearch lookups. Now it does.
6624 46. When parsing a route_list item in a manualroute router, a fixed-length
6625 buffer was used for the list of hosts. I made this 1024 bytes long,
6626 thinking that nobody would ever have a list of hosts that long. Wrong.
6627 Somebody had a whole pile of complicated expansion conditions, and the
6628 string was silently truncated, leading to an expansion error. It turns out
6629 that it is easier to change to an unlimited length (owing to other changes
6630 that have happened since this code was originally written) than to build
6631 structure for giving a limitation error. The length of the item that
6632 expands into the list of hosts is now unlimited.
6634 47. The lsearch lookup could not handle data where the length of text line was
6635 more than 4095 characters. Such lines were truncated, leading to shortened
6636 data being returned. It should now handle lines of any length.
6638 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6639 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6642 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6643 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6644 no longer happens in comment lines. A list of replacements is now placed
6645 at the head of all of the source files, except those whose only change is
6646 to replace PERL_COMMAND in the very first #! line.
6648 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6649 messages on the queue, with a bottom-up merge sort, using code contributed
6650 by Michael Haardt. This should make operations like -bp somewhat faster on
6651 large queues. It won't affect queue runners, except when queue_run_in_order
6654 51. Installed eximstats 1.31 in the distribution.
6656 52. Added support for SRV lookups to the dnslookup router.
6658 53. If an ACL referred to $message_body or $message_body_end, the value was not
6659 reset for any messages that followed in the same SMTP session.
6661 54. The store-handling optimization for building very long strings was not
6662 differentiating between the different store pools. I don't think this
6663 actually made any difference in practice, but I've tidied it.
6665 55. While running the routers to verify a sender address, $sender_address
6666 was still set to the sender address. This is wrong, because when routing to
6667 send a bounce to the sender, it would be empty. Therefore, I have changed
6668 it so that, while verifying a sender address, $sender_address is set to <>.
6669 (There is no change to what happens when verifying a recipient address.)
6671 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6672 target A or AAAA records (if not already returned) without resetting the
6673 qualify_single or search_parents options of the DNS resolver. These are
6674 inappropriate in this case because the targets of MX and SRV records must
6675 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6676 target that, when qualified, matched something in the local domain. These
6677 two options are now turned off when doing these lookups.
6679 57. It seems that at least some releases of Reiserfs (which does not have the
6680 concept of a fixed number of inodes) returns zero and not -1 for the
6681 number of available inodes. This interacted badly with check_spool_inodes,
6682 which assumed that -1 was the "no such thing" setting. What I have done is
6683 to check that the total number of inodes is greater than zero before doing
6684 the test of how many are available.
6686 58. When a "warn" ACL statement has a log_message modifier, the message is
6687 remembered, and not repeated. This is to avoid a lot of repetition when a
6688 message has many recipients that cause the same warning to be written.
6689 However, Exim was preserving the list of already written lines for an
6690 entire SMTP session, which doesn't seem right. The memory is now reset if a
6691 new message is started.
6693 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6694 debugging output unless log_rewrite was also set.
6696 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6697 of (int)(handle) when we know that handle contains (void *)(-1).
6699 61. The Exim daemon panic-logs an error return when it closes the incoming
6700 connection. However "connection reset by peer" seems to be common, and
6701 isn't really an error worthy of noting specially, so that particular error
6704 62. When Exim is trying to find all the local interfaces, it used to panic and
6705 die if the ioctl to get the interface flags failed. However, it seems that
6706 on at least one OS (Solaris 9) it is possible to have an interface that is
6707 included in the list of interfaces, but for which you get a failure error
6708 for this call. This happens when the interface is not "plumbed" into a
6709 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6710 failure of the "get flags" call assumes that the interface is down.
6712 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6713 makes life easier for people who are doing arithmetic on fields extracted
6714 from dates, where you often get leading zeros that should not be
6715 interpreted as octal.
6717 64. Added qualify_domain to the redirect router, to override the global
6720 65. If a pathologically long header line contained very many addresses (the
6721 report of this problem mentioned 10 000) and each of them was rewritten,
6722 Exim could use up a very large amount of memory. (It kept on making new
6723 copies of the header line as it rewrote, and never released the old ones.)
6724 At the expense of a bit more processing, the header rewriting function has
6725 been changed so that it no longer eats memory in this way.
6727 66. The generation of the Received: header has been moved from the time that a
6728 message starts to be received, to the time that it finishes. The timestamp
6729 in the Received: header should now be very close to that of the <= log
6730 line. There are two side-effects of this change:
6732 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6733 logged header lines no longer include the local Received: line, because
6734 it has not yet been created. The same applies to a copy of the message
6735 that is returned to a non-SMTP sender when a message is rejected.
6737 (b) When a filter file is tested using -bf, no additional Received: header
6738 is added to the test message. After some thought, I decided that this
6741 This change does not affect the value of $received_for. It is still set
6742 after address rewriting, but before local_scan() is called.
6744 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6746 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6747 gave an unhelpful panic error message, and a defer error. I have managed to
6748 change this behaviour so that it now rejects any supplied certificate,
6749 which seems right, as the list of acceptable certificates is empty.
6751 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6752 gave an unhelpful defer error. I have not managed to make this reject any
6753 supplied certificates, but the error message it gives is "no certificate
6754 supplied", which is not helpful.
6756 70. exigrep's output now also includes lines that are not associated with any
6757 message, but which match the given pattern. Implemented by a patch from
6758 Martin Sluka, which also tidied up the Perl a bit.
6760 71. Recipient callout verification, like sender verification, was using <> in
6761 the MAIL FROM command. This isn't really the right thing, since the actual
6762 sender may affect whether the remote host accepts the recipient or not. I
6763 have changed it to use the actual sender in the callout; this means that
6764 the cache record is now keyed on a recipient/sender pair, not just the
6765 recipient address. There doesn't seem to be a real danger of callout loops,
6766 since a callout by the remote host to check the sender would use <>.
6767 [SEE ABOVE: changed after hitting problems.]
6769 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6770 temporary errors. However, in the case of such a code being given after
6771 the end of a data transmission (i.e. after ".") Exim was failing to write
6772 a retry record for the message. (Yes, there was some broken host that was
6773 actually sending 8xx at this point.)
6775 73. An unknown lookup type in a host list could cause Exim to panic-die when
6776 the list was checked. (An example that provoked this was putting <; in the
6777 middle of a list instead of at the start.) If this happened during a DATA
6778 ACL check, a -D file could be left lying around. This kind of configuration
6779 error no longer causes Exim to die; instead it causes a defer error. The
6780 incident is still logged to the main and panic logs.
6782 74. Buglet left over from Exim 3 conversion. The message "too many messages
6783 in one connection" was written to the rejectlog but not the mainlog, except
6784 when address rewriting (yes!) was being logged.
6786 75. Added write_rejectlog option.
6788 76. When a system filter was run not as root (that is, when system_filter_user
6789 was set), the values of the $n variables were not being returned to the
6790 main process; thus, they were not subsequently available in the $sn
6793 77. Added +return_path_on_delivery log selector.
6795 78. A connection timeout was being treated differently from recipients deferred
6796 when testing hosts_max_try with a message that was older than the host's
6797 retry timeout. (The host should not be counted, thus allowing all hosts to
6798 be tried at least once before bouncing.) This may have been the cause of an
6799 occasionally reported bug whereby a message would remain on the queue
6800 longer than the retry timeout, but would be bounced if a delivery was
6801 forced. I say "may" because I never totally pinned down the problem;
6802 setting up timeout/retry tests is difficult. See also the next item.
6804 79. The ultimate address timeout was not being applied to errors that involved
6805 a combination of host plus message (for example, a timeout on a MAIL
6806 command). When an address resolved to a number of possible hosts, and they
6807 were not all tried for each delivery (e.g. because of hosts_max_try), a
6808 message could remain on the queue longer than the retry timeout.
6810 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6813 81. Fixed an obscure SMTP outgoing bug which required at least the following
6814 conditions: (a) there was another message waiting for the same server;
6815 (b) the server returned 5xx to all RCPT commands in the first message so
6816 that the message was not completed; (c) the server dropped the connection
6817 or gave a negative response to the RSET that Exim sends to abort the
6818 transaction. The observed case was a dropped connection after DATA that had
6819 been sent in pipelining mode. That is, the server had advertised PIPELINING
6820 but was not implementing it correctly. The effect of the bug was incorrect
6821 behaviour, such as trying another host, and this could lead to a crash.
6827 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6828 and daemon.c were passed as pointers to ints; they should have been
6829 pointers to socklen_t variables (which are typically unsigned ints).
6831 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6834 3. Fixed a really odd bug that affected only the testing scheme; patching a
6835 certain fixed string in the binary changed the value of another string that
6836 happened to be identical to the end of the original first string.
6838 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6839 name", it returns that address as the IP address. On some operating
6840 systems (e.g. Solaris), it also passes back the IP address string as the
6841 "host name". However, on others (e.g. Linux), it passes back an empty
6842 string. Exim wasn't checking for this, and was changing the host name to an
6843 empty string, assuming it had been canonicalized.
6845 5. Although rare, it is permitted to have more than one PTR record for a given
6846 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6847 all the names associated with an address, because they do in Solaris.
6848 However, it seems that they do not in Linux for data that comes from the
6849 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6850 I found this out when I moved to a new Linux workstation and tried to run
6851 the Exim test suite.
6853 To get round this problem I have changed the code so that it now does its
6854 own call to the DNS to look up PTR records when searching for a host name.
6855 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6856 addresses that are only in /etc/hosts are still found.
6858 This behaviour is, however, controlled by an option called host_lookup_
6859 order, which defaults to "bydns:byaddr". If people want to use the other
6860 order, or indeed, just use one or the other means of lookup, they can
6861 specify it in this variable.
6863 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6864 some operating systems, this comes back from gethostbyaddr() as an empty
6865 string, and this is what Exim used to test for. However, it seems that in
6866 other systems, "." is yielded. Exim now tests for this case too.
6868 7. The values of check_spool_space and check_log_space are now held internally
6869 as a number of kilobytes instead of an absolute number of bytes. If a
6870 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6871 kilobyte. This means that much larger values can be stored.
6873 8. Exim monitor: an attempt to get the action menu when not actually pointing
6874 at a message produces an empty menu entitled "No message selected". This
6875 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6876 no entries in it ("Shell widget menu has zero width and/or height"). So I
6877 have added a single, blank menu entry in this case.
6879 9. Added ${quote_local_part.
6881 10. MIME decoding is now applied to the contents of Subject: header lines when
6884 11. Now that a reference to $sender_host_address automatically causes a reverse
6885 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6886 host lookup before query-style lookups in lists that might use this
6887 variable. This has therefore been abolished, and the "net-" prefix is no
6888 longer necessary for query-style lookups.
6890 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6891 have been a typo for LFLAGS, so it has been changed.
6893 13. The install script calls Exim with "-C /dev/null" in order to find the
6894 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6895 to be output. However, since Exim outputs its version number before the
6896 error, it didn't break the script. It just looked ugly. I fixed this by
6897 always allowing "-C /dev/null" if the caller is root.
6899 14. Ignore overlarge ACL variable number when reading spool file - insurance
6900 against a later release with more variables having written the file.
6902 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6903 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6904 this, as well as the form without the "IPv6" on the front.
6906 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6907 OS/Makefile-Darwin file.
6909 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6910 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6912 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6913 variable. However, some OS use ints for inodes. I've added cast to long int
6914 to get rid of the compiler warning.
6916 19. I had forgotten to lock out "/../" in configuration file names when
6917 ALT_CONFIG_PREFIX was set.
6919 20. Routers used for verification do not need to specify transports. However,
6920 if such a router generated a host list, and callout was configured, Exim
6921 crashed, because it could not find a port number from the (non-existent)
6922 transport. It now assumes port 25 in this circumstance.
6924 21. Added the -t option to exigrep.
6926 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6927 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6928 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6929 them is set, LOOKUP_LSEARCH is forced.
6931 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6933 24. Added sender and host information to the "rejected by local_scan()" log
6934 line; previously there was no indication of these.
6936 25. Added .include_if_exists.
6938 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6939 Linux. It turns out that not all file systems support this. Apparently some
6940 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6941 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6942 not supported on the file descriptor, is now ignored when Exim is trying to
6943 sync a directory. This applies only to Linux.
6945 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6947 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6948 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6951 29. Added support for maildirsize files from supplied patch (modified a bit).
6953 30. The use of :fail: followed by an empty string could lead Exim to respond to
6954 sender verification failures with (e.g.):
6956 550 Verification failed for <xxx>
6957 550 Sender verify failed
6959 where the first response line was missing the '-' that indicates it is not
6960 the final line of the response.
6962 31. The loop for finding the name of the user that called Exim had a hardwired
6963 limit of 10; it now uses the value of finduser_retries, which is used for
6964 all other user lookups.
6966 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6969 33. Exim was neglecting to zero errno before one call of strtol() when
6970 expanding a string and expecting an integer value. On some systems this
6971 resulted in spurious "integer overflow" errors. Also, it was casting the
6972 result into an int without checking.
6974 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6975 did not work. The code looks as if it has *never* worked, though it appears
6976 to have been documented since at least release 1.62. I have made it work.
6978 35. The "timeout_DNS" error in retry rules, also documented since at least
6979 1.62, also never worked. As it isn't clear exactly what this means, and
6980 clearly it isn't a major issue, I have abolished the feature by treating it
6981 as "timeout", and writing a warning to the main and panic logs.
6983 36. The display of retry rules for -brt wasn't always showing the error code
6986 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6987 timeout_connect_A, timeout_connect_MX.
6989 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6990 to the empty sender.
6992 39. The daemon was not analysing the content of -oX till after it had closed
6993 stderr and disconnected from the controlling terminal. This meant that any
6994 syntax errors were only noted on the panic log, and the return code from
6995 the command was 0. By re-arranging the code a little, I've made the
6996 decoding happen first, so such errors now appear on stderr, and the return
6997 code is 1. However, the actual setting up of the sockets still happens in
6998 the disconnected process, so errors there are still only recorded on the
7001 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7002 connections (as happens on some IP stacks) was logged at start up time as
7003 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7004 it from "IPv6 and IPv4", which means that two separate sockets are being
7007 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7008 says whether AF_INET or AF_INET6 was passed as an argument.
7010 42. Exiwhat output was messed up when time zones were included in log
7013 43. Exiwhat now gives more information about the daemon's listening ports,
7014 and whether -tls-on-connect was used.
7016 44. The "port" option of the smtp transport is now expanded.
7018 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7019 silently ignored. Now an error message is written to the main and panic
7022 46. There's a new ACL modifier called "logwrite" which writes to a log file
7023 as soon as it is encountered.
7025 47. Added $local_user_uid and $local_user_gid at routing time.
7027 48. Exim crashed when trying to verify a sender address that was being
7030 49. Exim was recognizing only a space character after ".include". It now also
7031 recognizes a tab character.
7033 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7034 extracting the relevant information from the specification. The man page no
7035 longer contains scrambled data for the -d option, and I've added a section
7036 at the front about calling Exim under different names.
7038 51. Added "extra_headers" argument to the "mail" command in filter files.
7040 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7043 53. Installed eximstats 1.29.
7045 54. Added transport_filter_timeout as a generic transport option.
7047 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7048 Cc: header lines. This was required by RFC 822, but it not required by RFC
7051 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7052 incoming messages that did not have them. Now it does so only if the
7053 message originates locally, that is, if there is no associated remote host
7054 address. When Resent- header lines are present, this applies to the Resent-
7055 lines rather than the non-Resent- lines.
7057 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7058 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7060 58. Messages for configuration errors now include the name of the main
7061 configuration file - useful now that there may be more than one file in a
7062 list (.included file names were always shown).
7064 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7065 for those rare installations that do not start the daemon as root or run it
7066 setuid root. I've cut out the call to initgroups() if the daemon is not
7069 60. The Exim user and group can now be bound into the binary as text strings
7070 that are looked up at the start of Exim's processing.
7072 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7074 62. Added $mailstore_basename variable.
7076 63. Installed patch to sieve.c from Michael Haardt.
7078 64. When Exim failed to open the panic log after failing to open the main log,
7079 the original message it was trying to log was written to stderr and debug
7080 output, but if they were not available (the usual case in production), it
7081 was lost. Now it is written to syslog before the two lines that record the
7082 failures to open the logs.
7084 65. Users' Exim filters run in subprocesses under the user's uid. It is
7085 possible for a "deliver" command or an alias in a "personal" command to
7086 provoke an address rewrite. If logging of address rewriting is configured,
7087 this fails because the process is not running as root or exim. There may be
7088 a better way of dealing with this, but for the moment (because 4.30 needs
7089 to be released), I have disabled address rewrite logging when running a
7090 filter in a non-root, non-exim process.
7096 1. The buildconfig auxiliary program wasn't quoting the value set for
7097 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7098 not defined. This bug was masked in 4.22 by the effect that was fixed in
7101 2. Some messages that were rejected after a message id was allocated were
7102 shown as "incomplete" by exigrep. It no longer does this for messages that
7103 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7105 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7106 have allow_domain_literals set, the ID did not get logged in the <= line.
7107 Domain literals are now always recognized in Message-ID: header lines.
7109 4. The first argument for a ${extract expansion item is the key name or field
7110 number. Leading and trailing spaces in this item were not being ignored,
7111 causing some misleading effects.
7113 5. When deliver_drop_privilege was set, single queue runner processes started
7114 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7115 same command in the process it spins off) were not dropping privilege.
7117 6. When the daemon running as "exim" started a queue runner, it always
7118 re-executed Exim in the spun-off process. This is a waste of effort when
7119 deliver_drop_privilege is set. The new process now just calls the
7120 queue-runner function directly.
7126 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7129 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7130 ignored. Though the use of -oP was forcing the writing of a pid file, it
7131 was always written to the default place.
7133 3. If the message "no IP address found for host xxxx" is generated during
7134 incoming verification, it is now followed by identification of the incoming
7135 connection (so you can more easily find what provoked it).
7137 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7139 5. Added some features to "harden" Exim a bit more against certain attacks:
7141 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7142 be put in Local/Makefile. This is like the never_users runtime option,
7143 but it cannot be overridden. The default setting is "root".
7145 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7146 prefix string with which any file named in a -C command line option
7149 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7150 is retained for -C and -D only if the caller of Exim is root. Without
7151 it, the exim user may also use -C and -D and retain privilege.
7153 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7154 command line option is disabled.
7156 6. Macro names set by the -D option must start with an upper case letter, just
7157 like macro names defined in the configuration file.
7159 7. Added "dereference=" facility to LDAP.
7161 8. Two instances of the typo "uknown" in the source files are fixed.
7163 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7164 the Configure-Makefile script screwed up while processing it.
7166 10. Incorporated PCRE 4.4.
7168 11. The SMTP synchronization check was not operating right at the start of an
7169 SMTP session. For example, it could not catch a HELO sent before the client
7170 waited for the greeting. There is now a check for outstanding input at the
7171 point when the greeting is written. Because of the duplex, asynchronous
7172 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7173 way, but not yet received, when the check is performed.
7175 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7176 on TCP/IP sockets, because this apparently causes some broken clients to
7179 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7180 unchanged) from the Cygwin maintainer.
7182 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7183 was supported instead of testing for mbx. Effectively a typo.
7185 15. The spa authenticator server code was not checking that the input it
7186 received was valid base64.
7188 16. The debug output line for the "set" modifier in ACLs was not showing the
7189 name of the variable that was being set.
7191 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7193 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7194 lookup on its own. Something else was needed to trigger the lookup. For
7195 example, a match in host_lookup or the need for a host name in a host list.
7196 Now, if $sender_host_name is referenced and the host name has not yet been
7197 looked up, a lookup is performed. If the lookup fails, the variable remains
7198 empty, and $host_lookup_failed is set to "1".
7200 19. Added "eqi" as a case-independent comparison operator.
7202 20. The saslauthd authentication condition could segfault if neither service
7203 nor realm was specified.
7205 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7206 error message that was logged was misleading, and incoming SMTP
7207 connections were dropped. The message is now more accurate, and temporary
7208 errors are given to SMTP connections.
7210 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7211 (except RSET) with a 421 error, until QUIT is received. However, it was
7212 failing to send a response to QUIT.
7214 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7215 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7216 ACL is now run after the helo_try_verify_hosts code.
7218 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7219 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7220 case-independent, so other case variants are also recognized). Apparently
7221 some systems use these upper case variants.
7223 25. If more than two messages were waiting for the same host, and a transport
7224 filter was specified for the transport, Exim sent two messages over the
7225 same TCP/IP connection, and then failed with "socket operation on non-
7226 socket" when it tried to send the third.
7228 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7230 27. The extern definition of crypt16() in expand.c was not being excluded when
7231 the OS had its own crypt16() function.
7233 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7234 as a preferred synonym for return_size_limit, both as an option and as an
7237 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7239 30. Changed the default configuration ACL to relax the local part checking rule
7240 for addresses that are not in any local domains. For these addresses,
7241 slashes and pipe symbols are allowed within local parts, but the sequence
7242 /../ is explicitly forbidden.
7244 31. SPA server authentication was not clearing the challenge buffer before
7247 32. log_message in a "warn" ACL statement was writing to the reject log as
7248 well as to the main log, which contradicts the documentation and doesn't
7249 seem right (because no rejection is happening). So I have stopped it.
7251 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7252 However, I am unable to do any testing of this.
7254 34. Fixed an infelicity in the appendfile transport. When checking directories
7255 for a mailbox, to see if any needed to be created, it was accidentally
7256 using path names with one or more superfluous leading slashes; tracing
7257 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7259 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7260 contents are added to the log line that is written for every discarded
7261 recipient. (Previously a log_message setting was ignored.)
7263 36. The ${quote: operator now quotes the string if it is empty.
7265 37. The install script runs exim in order to find its version number. If for
7266 some reason other than non-existence or emptiness, which it checks, it
7267 could not run './exim', it was installing it with an empty version number,
7268 i.e. as "exim-". This error state is now caught, and the installation is
7271 38. An argument was missing from the function that creates an error message
7272 when Exim fails to connect to the socket for saslauthd authentication.
7273 This could cause Exim to crash, or give a corrupted message.
7275 39. Added isip, isip4, and isip6 to ${if conditions.
7277 40. The ACL variables $acl_xx are now saved with the message, and can be
7278 accessed later in routers, transports, and filters.
7280 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7281 strings in the file are not string-expanded.
7283 42. If a MAIL command specified a SIZE value that was too large to fit into an
7284 int variable, the check against message_size_limit failed. Such values are
7285 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7286 day this will have to be increased, but I don't think I want to be around
7287 when emails are that large.
7294 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7295 iconv() is not standard in FreeBSD.
7297 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7298 IPv6 enabled. The observed symptom was a segmentation fault on return from
7299 the function os_common_find_running_interfaces() in src/os.c.
7301 3. In the check_special_case() function in daemon.c I had used "errno" as an
7302 argument name, which causes warnings on some systems. This was basically a
7303 typo, since it was named "eno" in the comments!
7305 4. The code that waits for the clock to tick (at a resolution of some fraction
7306 of a second) so as to ensure message-id uniqueness was always waiting for
7307 at least one whole tick, when it could have waited for less. [This is
7308 almost certainly not relevant at current processor speeds, where it is
7309 unlikely to ever wait at all. But we try to future-proof.]
7311 5. The function that sleeps for a time interval that includes fractions of a
7312 second contained a race. It did not block SIGALRM between setting the
7313 timer, and suspending (a couple of lines later). If the interval was short
7314 and the sigsuspend() was delayed until after it had expired, the suspension
7315 never ended. On busy systems this could lead to processes getting stuck for
7318 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7319 process, before it forks any delivery processes. The open lookup caching
7320 mechanism meant that the open file or database connection was passed into
7321 the delivery process. The problem was that delivery processes always tidy
7322 up cached lookup data. This could cause a problem for the next delivery
7323 process started by the queue runner, because the external queue runner
7324 process does not know about the closure. So the next delivery process
7325 still has data in the lookup cache. In the case of a file lookup, there was
7326 no problem because closing a file descriptor in a subprocess doesn't affect
7327 the parent. However, if the lookup was caching a connection to a database,
7328 the connection was closed, and the second delivery process was likely to
7329 see errors such as "PGSQL: query failed: server closed the connection
7330 unexpectedly". The problem has been fixed by closing all cached lookups
7331 in a queue runner before running a delivery process.
7333 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7334 seem to have the "const" qualifier which it has on other OS. I've
7337 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7338 *of the same type* (client or server) with the same public name that an
7339 error should be diagnosed.
7341 9. When Exim looked up a host name for an IP address, but failed to find the
7342 original IP address when looking up the host name (a safety check), it
7343 output the message "<ip address> does not match any IP for NULL", which was
7344 confusing, to say the least. The bug was that the host name should have
7345 appeared instead of "NULL".
7347 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7348 user that is built into the binary, and the -C or -D options is used, root
7349 privilege is dropped before the configuration file is read. In addition,
7350 logging is switched to stderr instead of the normal log files. If the
7351 configuration then re-defines the Exim user, the unprivileged environment
7352 is probably not what is expected, so Exim logs a panic warning message (but
7355 However, if deliver_drop_privilege is set, the unprivileged state may well
7356 be exactly what is intended, so the warning has been cut out in that case,
7357 and Exim is allowed to try to write to its normal log files.
7363 1. smtp_return_error_details was not giving details for temporary sender
7364 or receiver verification errors.
7366 2. Diagnose a configuration error if two authenticators have the same public
7369 3. Exim used not to create the message log file for a message until the first
7370 delivery attempt. This could be confusing when incoming messages were held
7371 for policy or load reasons. The message log file is now created at the time
7372 the message is received, and an initial "Received" line is written to it.
7374 4. The automatically generated man page for command line options had a minor
7375 bug that caused no ill effects; however, a more serious problem was that
7376 the procedure for building the man page automatically didn't always
7377 operate. Consequently, release 4.20 contains an out-of-date version. This
7378 shouldn't happen again.
7380 5. When building Exim with embedded Perl support, the script that builds the
7381 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7382 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7384 6. The freeze_tell option was not being used for messages that were frozen on
7385 arrival, either by an ACL or by local_scan().
7387 7. Added the smtp_incomplete_transaction log selector.
7389 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7390 was accepting AUTH without a new EHLO.
7392 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7393 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7396 10. Logging of TCP/IP connections (when configured) now happens in the main
7397 daemon process instead of the child process, so that the TCP/IP connection
7398 count is more accurate (but it can never be perfect).
7400 11. The use of "drop" in a nested ACL was not being handled correctly in the
7401 outer ACL. Now, if condition failure induced by the nested "drop" causes
7402 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7403 or "require"), the connection is dropped.
7405 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7406 that yield "discard" can now be used with an "accept" or a "discard" verb,
7407 but an error is generated for any others (because I can't see a useful way
7408 to define what should happen).
7410 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7411 are now processed in the same way as lines in the Exim configuration file.
7412 In particular, continuation lines are supported.
7414 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7416 15. Added -ti meaning -t -i.
7418 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7419 domains, and warn by logging if others are found.
7421 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7422 returned from GIFCONF when Exim is trying to find the list of interfaces on
7423 a host. The code in os.c has been modified to copy each ifreq to an aligned
7424 structure in all cases.
7426 Also, in some cases, the returned ifreq's were being copied to a 'struct
7427 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7428 means the last couple of bytes of an IPv6 address could be chopped if the
7429 ifreq contained only a normal sockaddr (14 bytes storage).
7431 18. Named domain lists were not supported in the hosts_treat_as_local option.
7432 An entry such as +xxxx was not recognized, and was treated as a literal
7435 19. Ensure that header lines added by a DATA ACL are included in the reject log
7436 if the ACL subsequently rejects the message.
7438 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7439 MD5 (which is deprecated).
7441 21. When testing a filter file using -bf, Exim was writing a message when it
7442 took the sender from a "From " line in the message, but it was not doing so
7443 when it took $return_path from a Return-Path: header line. It now does.
7445 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7446 with a valid header line field name (a series of printing characters
7447 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7449 23. Changed "disc" in the source to "disk" to conform to the documentation and
7450 the book and for uniformity.
7452 24. Ignore Sendmail's -Ooption=value command line item.
7454 25. When execve() failed while trying to run a command in a pipe transport,
7455 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7456 could be confused with a return value of 69 from the command itself. This
7457 has been changed to 127, the value the shell returns if it is asked to run
7458 a non-existent command. The wording for the related log line suggests a
7459 non-existent command as the problem.
7461 26. If received_header_text expands to an empty string, do not add a Received:
7462 header line to the message. (Well, it adds a token one on the spool, but
7463 marks it "old" so that it doesn't get used or transmitted.)
7465 27. Installed eximstats 1.28 (addition of -nt option).
7467 28. There was no check for failure on the call to getsockname() in the daemon
7468 code. This can fail if there is a shortage of resources on the system, with
7469 ENOMEM, for example. A temporary error is now given on failure.
7471 29. Contrary to the C standard, it seems that in some environments, the
7472 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7473 program. Exim now does this explicitly; it affects the formatting of
7474 timestamps using strftime().
7476 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7477 complaints. I've now initialized all the variables, to avoid this.
7479 32. Header lines added by a system filter were not being "seen" during
7480 transport-time rewrites.
7482 33. The info_callback() function passed to OpenSSL is set up with type void
7483 (*)(SSL *, int, int), as described somewhere. However, when calling the
7484 function (actually a macro) that sets it up, the type void(*)() is
7485 expected. I've put in a cast to prevent warnings from picky compilers.
7487 34. If a DNS black list lookup found a CNAME record, but there were no A
7488 records associated with the domain it pointed at, Exim crashed.
7490 35. If a DNS black list lookup returned more than one A record, Exim ignored
7491 all but the first. It now scans all returned addresses if a particular IP
7492 value is being sought. In this situation, the contents of the
7493 $dnslist_value variable are a list of all the addresses, separated by a
7496 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7497 to accept a host name and all its aliases if the forward lookup for any of
7498 them yielded the IP address of the incoming connection. Now it accepts only
7499 those names whose forward lookup yields the correct IP address. Any other
7500 names are discarded. This closes a loophole whereby a rogue DNS
7501 administrator could create reverse DNS records to break through a
7502 wildcarded host restriction in an ACL.
7504 37. If a user filter or a system filter that ran in a subprocess used any of
7505 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7506 the wrong values were passed to the pipe command ($thisaddress had the
7507 value of $0, $0 had the value of $1, etc). This bug was introduced by
7508 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7510 38. Improved the line breaking for long SMTP error messages from ACLs.
7511 Previously, if there was no break point between 40 and 75 characters, Exim
7512 left the rest of the message alone. Two changes have been made: (a) I've
7513 reduced the minimum length to 35 characters; (b) if it can't find a break
7514 point between 35 and 75 characters, it looks ahead and uses the first one
7515 that it finds. This may give the occasional overlong line, but at least the
7516 remaining text gets split now.
7518 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7519 file descriptors might be low, and that setting 1000 would always raise it.
7520 It turns out that in some environments, the limit is already over 1000 and
7521 that lowering it causes trouble. So now Exim takes care not to decrease it.
7523 40. When delivering a message, the value of $return_path is set to $sender_
7524 address at the start of routing (routers may change the value). By an
7525 oversight, this default was not being set up when an address was tested by
7526 -bt or -bv, which affected the outcome if any router or filter referred to
7529 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7530 log only when "message" or "log_message" are set. However, if one of the
7531 conditions was an address verification, or a call to a nested ACL, the
7532 messages generated by the underlying test were being passed through. This
7533 no longer happens. The underlying message is available in $acl_verify_
7534 message for both "message" and "log_message" expansions, so it can be
7535 passed through if needed.
7537 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7538 new expansion $bh_ to give the encoded byte string without charset
7539 translation. Translation happens only if iconv() is available; HAVE_ICONV
7540 indicates this at build time. HEADERS_CHARSET gives the charset to
7541 translate to; headers_charset can change it in the configuration, and
7542 "headers charset" can change it in an individual filter file.
7544 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7545 that creates RFC 2047 encoded "words" labels them as that charset instead
7546 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7547 expansion operator; (ii) when Exim creates a From: line for a local
7548 message; (iii) when a header line is rewritten to include a "phrase" part.
7550 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7551 buggy, causing it to skip the first lines of messages whose message ID
7552 ended in 'D'. This would not have bitten before Exim release 4.14, because
7553 message IDs were unlikely to end in 'D' before then. The effect was to have
7554 incorrect size information for certain domains.
7556 45. #include "config.h" was missing at the start of the crypt16.c module. This
7557 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7560 46. If there was a timeout during a "random" callout check, Exim treated it as
7561 a failure of the random address, and carried on sending RSET and the real
7562 address. If the delay was just some slowness somewhere, the response to the
7563 original RCPT would be taken as a response to RSET and so on, causing
7564 mayhem of various kinds.
7566 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7567 when I implemented it. It didn't allow for the fact that some option values
7568 may legitimately be negative (e.g. size_addition), and it didn't even do
7569 the right test for positive values.
7571 48. Domain names in DNS records are case-independent. Exim always looks them up
7572 in lower case. Some resolvers return domain names in exactly the case they
7573 appear in the zone file, that is, they may contain uppercase letters. Not
7574 all resolvers do this - some return always lower case. Exim was treating a
7575 change of case by a resolver as a change of domain, similar to a widening
7576 of a domain abbreviation. This triggered its re-routing code and so it was
7577 trying to route what was effectively the same domain again. This normally
7578 caused routing to fail (because the router wouldn't handle the domain
7579 twice). Now Exim checks for this case specially, and just changes the
7580 casing of the domain that it ultimately uses when it transmits the message
7583 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7586 50. If a filter generated a file delivery with a non-absolute name (possible if
7587 no home directory exists for the router), the forbid_file option was not
7590 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7591 the existing equality matching.
7593 52. Exim was using ints instead of ino_t variables in some places where it was
7594 dealing with inode numbers.
7596 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7597 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7598 called TMPDIR, and if it finds it is different, it changes its value.
7600 54. The smtp_printf() function is now made available to local_scan() so
7601 additional output lines can be written before returning. There is also an
7602 smtp_fflush() function to enable the detection of a dropped connection.
7603 The variables smtp_input and smtp_batched_input are exported to
7606 55. Changed the default runtime configuration: the message "Unknown user"
7607 has been removed from the ACL, and instead placed on the localuser router,
7608 using the cannot_route_message feature. This means that any verification
7609 failures that generate their own messages won't get overridden. Similarly,
7610 the "Unrouteable address" message that was in the ACL for unverifiable
7611 relay addresses has also been removed.
7613 56. Added hosts_avoid_esmtp to the smtp transport.
7615 57. The exicyclog script was not checking for the esoteric option
7616 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7617 will work only if exicyclog is run under the appropriate euid.
7619 58. Following a discussion on the list, the rules by which Exim recognises line
7620 endings on incoming messages have been changed. The -dropcr and drop_cr
7621 options are now no-ops, retained only for backwards compatibility. The
7622 following line terminators are recognized: LF CRLF CR. However, special
7623 processing applies to CR:
7625 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7626 nor a local message in the state where . is a terminator.
7628 (ii) If a bare CR is encountered in a header line, an extra space is added
7629 after the line terminator so as not to end the header. The reasoning
7630 behind this is that bare CRs in header lines are most likely either
7631 to be mistakes, or people trying to play silly games.
7633 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7634 was being incorrectly given as 18 bytes larger than it should have been.
7635 This is a VOB (very old bug).
7637 60. This may never have affected anything current, but just in case it has:
7638 When the local host is found other than at the start of a list of hosts,
7639 the local host, those with the same MX, and any that follow, are discarded.
7640 When the list in question was part of a longer list of hosts, the following
7641 hosts (not currently being processed) were also being discarded. This no
7642 longer happens. I'm not sure if this situation could ever has previously
7645 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7648 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7649 header. This is something that is recommended in a new Internet Draft, and
7650 is something that is documented as being done by Sendmail. There are two
7651 possible values. For messages generated by the autoreply transport, Exim
7654 Auto-Submitted: auto-replied
7656 whereas for all other generated messages (e.g. bounces) it adds
7658 Auto-Submitted: auto-generated
7660 63. The "personal" condition in filters now includes a test for the
7661 Auto-Submitted: header. If it contains the string "auto-" the message it
7662 not considered personal.
7664 64. Added rcpt_include_affixes as a generic transport option.
7666 65. Added queue_only_override (default true).
7668 66. Added the syslog_duplication option.
7670 67. If what should have been the first header line of a message consisted of
7671 a space followed by a colon, Exim was mis-interpreting it as a header line.
7672 It isn't of course - it is syntactically invalid and should therefore be
7673 treated as the start of the message body. The misbehaviour could have
7674 caused a number of strange effects, including loss of data in subsequent
7675 header lines, and spool format errors.
7677 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7678 client host had authenticated. This control can now be exercised by an ACL
7679 for more flexibility.
7681 69. By default, callouts do not happen when testing with -bh. There is now a
7682 variant, -bhc, which does actually run the callout code, including
7683 consulting and updating the callout cache.
7685 70. Added support for saslauthd authentication, courtesy of Alexander
7688 71. If statvfs() failed on the spool or log directories while checking their
7689 size for availability, Exim confusingly gave the error "space shortage".
7690 Furthermore, in debugging mode it crashed with a floating point exception.
7691 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7692 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7693 serious problem, Exim now writes to the main and panic logs when this
7694 happens, with details of the failure. It then refuses to accept the
7695 incoming message, giving the message "spool directory problem" or "log
7696 directory problem" with a 421 code for SMTP messages.
7698 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7699 0, 1, and 2 exist, because some OS complain for execs without them (see
7700 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7701 descriptors. However, the code omitted to check that the open succeeded,
7702 causing mysterious errors if for some reason the permissions on /dev/null
7703 got screwed. Now Exim writes a message to the main and panic logs, and
7704 bombs out if it can't open /dev/null.
7706 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7707 interact so that it is all more flexible. It is supposed to remain
7708 backwards compatible. Also added extra_local_interfaces.
7710 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7711 to bomb out with an assertion failure - to the client this appears as a
7712 connection drop. This problem occurs in the part of the code that was taken
7713 from the Samba project. Fortunately, the assertion is in a very simple
7714 function, so I have fixed this by reproducing the function inline in the
7715 one place where it is called, and arranging for authentication to fail
7716 instead of killing the process with assert().
7718 75. The SPA client code was not working when the server requested OEM rather
7719 than Unicode encoding.
7721 76. Added code to make require_files with a specific uid setting more usable in
7722 the case where statting the file as root fails - usually a non-root-mounted
7723 NFS file system. When this happens and the failure is EACCES, Exim now
7724 forks a subprocess and does the per-uid checking as the relevant uid.
7726 77. Added process_log_path.
7728 78. If log_file_path was not explicitly set, a setting of check_log_space or
7729 check_log_inodes was ignored.
7731 79. If a space check for the spool or log partitions fails, the incident is now
7732 logged. Of course, in the latter case the data may get lost...
7734 80. Added the %p formatting code to string_format() so that it can be used to
7735 print addresses in debug_print(). Adjusted all the address printing in the
7736 debugging in store.c to use %p rather than %d.
7738 81. There was a concern that a line of code in smtp_in.c could overflow a
7739 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7740 initially expressed, the concern was not well-founded, because trailing
7741 spaces are removed early. However, if the trailing spaces were followed by
7742 a NULL, they did not get removed, so the overflow was possible. Two fixes
7745 (a) I re-wrote the offending code in a cleaner fashion.
7746 (b) If an incoming SMTP command contains a NULL character, it is rejected
7749 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7750 runs initgroups(), so that if the Exim user is in any additional groups,
7751 they will be used during message reception.
7757 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob